US 20080195579 A1
Methods and systems for extraction of transaction data for compliance monitoring, particularly useful in a system for monitoring electronic transactions of an enterprise and detecting exceptions indicating noncompliance with enterprise policies. Data extractors obtain data from various data sources and provide the data for use by a transaction analysis engine that executes computer-executable compliance policy statements against extracted data. Data extractors include one or more of following: a master extractor, a log extractor, a resync extractor, a programmatic extractor, an environmental source extractor, and an external source extractor. Data extraction is effected by using the master extractor and/or programmatic extractor to extract an initial subset of information about monitored transactions from an enterprise system, and then using the log extractor, the resync extractor, the programmatic extractor, the environmental source extractor, and/or the external source extractor to extract a second subset of information about changed data.
1. A system for extracting data relating to transactions from one or more external data sources and for providing data for use by a transaction analysis engine operative for executing one or more computer-executable compliance policy statements against extracted data, comprising:
a data communications port coupled to one or more data sources;
a master extractor program module operative for extracting an initial selected subset of information about monitored transactions from one or more data sources;
a log extractor program module operative for extracting information about monitored transactions from a log file provided by one or more data sources;
a resync extractor program module operative for extracting information from one or more data sources from which data was previously extracted in response to a resynchronize condition;
a programmatic extractor program module operative to receive information provided by programmatic operation of a data source that conducts data export; and
a storage program module for storing extracted information.
2. The system of
3. The system of
4. The system of
5. The system of
6. The system of
7. The system of
8. The system of
9. The system of
10. The system of
11. The system of
12. The system of
13. A method for extracting data relating to transactions from one or more external data sources and for providing data for use by a transaction analysis engine operative for executing one or more computer-executable compliance policy statements against extracted data, comprising the steps of:
extracting an initial selected subset of information about monitored transactions from one or more data sources;
extracting information about monitored transactions from a log file provided by one or more data sources;
extracting information from one or more data sources from which information was previously extracted to resynchronize data in response to a resynchronization condition;
receiving information provided by programmatic operation of a data source that conducts data export; and
storing extracted information for access by the transaction analysis engine.
14. The method of
15. The method of
16. The method of
17. The method of
18. The method of
19. The method of
20. The method of
21. The method of
22. The method of
23. The method of
24. The method of
25. A system for monitoring transactions of an enterprise stored in one or more enterprise systems associated with an enterprise, the transactions corresponding to one or more business transactions of the enterprise, to determine possible lack of compliance of a business transaction with one or more predetermined policies of the enterprise, comprising:
an extractor for extracting an initial selected subset of information about monitored transactions from an enterprise system;
a second extractor, responsive to changed data in the enterprise system, for extracting a second selected subset of information about the changed data;
a monitoring database for storing the initial subset of information and the second subset of information as monitoring entities;
a transaction analysis engine for executing one or more computer-executable policy statements against the monitoring database; and
a user interface, responsive to a determination from execution of a policy statement of an exception to a predetermined policy for providing an output corresponding to the exception indicating a possible lack of compliance with the predetermined policy.
26. The system of
27. The system of
28. The system of
29. The system of
30. The system of
31. The system of
32. The system of
33. The system of
34. The system of
35. The system of
36. A computer-implemented method for monitoring transactions of an enterprise stored in one or more enterprise systems associated with an enterprise, the transactions corresponding to one or more business transactions of the enterprise, to determine possible lack of compliance of a business transaction with one or more predetermined policies of the enterprise, comprising the steps of:
extracting an initial selected subset of information about monitored transactions from an enterprise system;
storing the initial subset of information in a monitoring database as a monitoring entity;
thereafter, in response to changed data in the enterprise system, extracting a second selected subset of information about the changed data;
storing the second subset of information in the monitoring database as a monitoring entity;
executing one or more computer-executable policy statements from a knowledge base against the monitoring database; and
in response to the determination from the execution of a policy statement of an exception to a predetermined policy, providing an output corresponding to an exception indicating a possible lack of compliance with the predetermined policy.
37. The method of
38. The method of
39. The method of
40. The method of
41. The method of
42. The method of
43. The method of
44. The method of
45. The method of
46. The method of
This application is a continuation of copending U.S. patent application entitled “METHODS AND SYSTEMS FOR TRANSACTION COMPLIANCE MONITORING” by Peter H. Kennis, Daniel R. Kuokka, Charles A. Coombs, Stayton D. Addison, Andrew T. Otwell, Jeffrey Z. Johnson, Patrick Taylor, and Michael E. Lortz, having application no. 11/085,725, filed on Mar. 21, 2005, which claims the benefit of and priority on U.S. Provisional Patent Application No. 60/554,784 entitled “METHODS AND SYSTEMS FOR CONTINUOUS MONITORING OF TRANSACTION DATA FLOW” BY Peter H. Kennis, Stayton D. Addison, Charles A. Coombs, Andrew T. Otwell, and Daniel R. Kuokka, filed on Mar. 19, 2004, the disclosures of which are hereby incorporated herein by reference in their entirety.
This application is also related to and incorporates by reference herein the following U.S. patent applications:
The present invention relates generally to compliance monitoring of electronic enterprise transactions, and more particularly relates to extraction of electronic data transactions within enterprise computing systems for enterprise policy compliance monitoring, anomaly detection, risk assessment, fraud deterrence, and investigation.
The growth of automated business systems, such as enterprise resource planning (ERP) and customer relationship management (CRM) applications, continues to propel productivity gains and new efficiencies in the e-business world. These business systems allow organizations to easily manage accounts payable, human resources, account receivables, inventory, payroll, and more in real-time. However, automated business systems are subject to errors, misuse, and fraud, just like manual, unautomated systems. Furthermore, automated business systems can open the door for business “hacks” resulting in asset misappropriation and significant financial losses. Both intentional and unintentional problems can jeopardize the integrity of transactions and reporting of an enterprise.
Sources of integrity compromise can be broken into categories that range from the most malicious to guiltless acts of well-meaning employees. Vulnerabilities in electronic transaction systems can: (1) permit access to target business applications to launch fraudulent schemes, (2) unknowingly introduce system errors that affect asset appropriation, such as create duplicate payments, or (3) allow system control to be overridden or circumvented, which then provides others the opportunity to abuse or misuse the system to commit fraud.
Organizations must take measures to reduce and eliminate all forms of errors, misuse, and fraud. Present day financial controls of modern business enterprises do not do enough to mitigate business risks from fraud and error within the organization. According to reports from the Association of Certified Fraud Examiners (ACFE), fraud and white collar hacks collectively drain 6 percent of a typical business enterprise's annual revenue. In 2002, these losses purportedly totaled over $600 billion. A survey by one well-known accounting firm pegged the average loss per company at greater than $2 million. Another accounting firm calls the problem of fraud and error “a bigger loss problem than viruses and worms combined.”
The ACFE study found that an average fraud scheme lasted 18 months before it was detected. More than half of the detected schemes accounted for losses greater than $100,000; nearly one in six caused losses greater than $1 million. The study also reported that nearly two-thirds of all identified fraud was detected by “accident” or employee tips.
New motivations for evaluating financial controls, including the Sarbanes-Oxley Act of 2002, have driven some enterprises to re-think their financial controls. Section 404 of the Sarbanes-Oxley Act caused the Securities and Exchange Commission (SEC) to establish rules about annual reports of certain companies, especially publicly held companies. Such rules require an annual report to contain (1) a statement of management's responsibility for establishing and maintaining an adequate internal control structure and procedures for financial reporting, as well as (2) management's assessment, as of the end of the company's most recent fiscal year, of the effectiveness of the company's internal control structure and procedures for financial reporting. Section 404 also requires the company's auditor to attest to, and report on management's assessment of the effectiveness of the company's internal controls and procedures for financial reporting in accordance with standards established by the Public Company Accounting Oversight Board. These requirements alone have triggered a search by both a company's management and auditors for solutions to the establishment and maintenance of internal control structures, which are inevitably reflected in a company policies and procedures.
The Sarbanes-Oxley Act has heightened the importance of establishing enterprise policies regarding business activities and practices, ensuring compliance to those policies, and correcting lack of compliance promptly and efficiently. Failure to establish and abide by some government-imposed requirements can result in criminal as well as civil penalties, so many businesses and other organizations are scrambling to establish policies and compliance monitoring systems.
The real-time nature of information, analysis, decision-making, and policy validation creates additional complexities in financial controls and compliance monitoring. Partly because so much information in modern business enterprises is conducted by computer systems, some businesses and government organizations are exploring whether it is feasible to implement automated transaction monitoring systems as an alternative or supplemental to traditional people-based financial controls. In the process of exploring automated monitoring systems, many enterprises are facing tradeoffs between stringent controls, operational efficiency, and business risk. While stringent systems controls may stop a small percent of insiders who intend to defraud the enterprise, stringent controls place a heavy burden on the vast majority of insiders who are honest. Theoretically, automated transaction monitoring should allow an enterprise to remove many system restrictions and rely on real-time analysis to flag transactions that do not comply with enterprise policies. However, prior efforts to provide efficient and effective automated transaction monitoring systems have not been entirely successful.
Some prior approaches to automated transaction monitoring focused on narrow fields of critical transaction data flows and were implemented to detect overt indications of profound and clear problems. Software tools that assist in recording and documenting the investigative actions of a human auditor are known (case management systems). Some functions in querying available data were automated but only so under the direction of a human operator. Such limited approaches are watchful of only a small percentage of transactions on a computer system. Problematic issues in areas outside of the monitored fields can be overlooked though such issues may result in problems in seemingly non-critical transactions, may affect critical transactions with subtlety, and may result in disperse adverse affects that amount in summations to problems deserving attention but that may go undetected.
Accordingly, there is room for improvement in automated transaction monitoring systems that are operative for establishing enterprise policies and procedures, monitoring compliance with such policies and procedures, and reporting violations or deviations from the established policies and procedures. But there are various requirements for a system that will be effective and acceptable to the business community. Automated transaction monitoring must rely upon sophisticated data acquisition and multi-perspective analysis to correlate information from ERP systems, legacy mainframe applications, network monitoring solutions, and external data sources. These various systems implement the known business functions of accounts payable, accounts receivable, general ledger, human resources & payroll, and inventory management. After collecting relevant transaction information, automated transaction monitoring solutions must analyze each transaction and the context of the transaction with the same level of scrutiny that an internal human auditor and fraud examiner would employ. This complex analysis requires a combination of domain engineering, automated link analysis, behavior, deductive analysis, and standard business intelligence.
Furthermore, an effective transaction analysis system should flag suspicious activities and attempt to distinguish real concerns from hundreds of indicators of fraud, misuse, and errors. The system should detect acts of concealment and conversion designed to circumvent standard auditing techniques. The system should preferably operate in continuous or near real-time mode, so as to detect efforts at concealment and prevent complications and expense from later remedy.
Providing an acceptable transaction monitoring and analysis system has proven a daunting task. Nonetheless, the benefits of such a system are clear: (1) transaction integrity monitoring should build an audit trail of transactions within a financial system and direct internal auditors to the most suspicious transactions, (2) transaction integrity monitoring should establish a business environment that deters employees and other insiders from breaking enterprise policies or defrauding the company, (3) transaction integrity monitoring should provide the benefits of rigorous financial controls without the administrative overhead and bureaucratic burden, (4) even if compliance with policies is not 100% or employees learn to game the system, risk managers should have a solution that keeps pace with real-time business transactions, and (5) an acceptable transaction integrity monitoring system should act as the ultimate layer of security from outsiders who penetrate the network as authorized users.
As will be described and explained in detail below, the present inventors have constructed various systems and methods that meet these and other requirements for an efficient, effective, robust, and comprehensive automated electronic transaction integrity monitoring.
Briefly described, and according to one aspect, the present invention relates to systems and methods for extracting data relating to transactions from one or more external data sources and for providing data for use by a transaction analysis engine operative for executing one or more computer-executable compliance policy statements against the extracted data. The invention is particularly suitable for use in an automated electronic transaction integrity monitoring system that allows establishment, codification, and maintenance of enterprise policies, monitors electronic transactions of the enterprise from various and possibly heterogeneous data sources, detects exceptions to established policies, reports such exceptions to authorized users such as managers and auditors, and/or provides a case management system for tracking such exceptions and their underlying transactions.
More particularly described, a system constructed in accordance with this aspect of the invention comprises a data communications port coupled to one or more data sources, and one or more program modules for extraction of transaction data from the one or more data sources, where those data sources are multiple and possibly heterogeneous sources that are continuously and repeatedly monitored for policy exceptions that may be clues indicating attention or investigation is in order with regard to particular transactions. In an embodiment of this aspect of the invention, those program modules may include one or more of the following: a master extractor program module operative for extracting an initial selected subset of information about monitored transactions from one or more data sources; a log extractor program module operative for extracting information about monitored transactions from a log file provided by one or more data sources; a resync extractor program module operative for extracting information from one or more data sources from which data was previously extracted in response to a resynchronize condition; a programmatic extractor program module operative to receive information provided by programmatic operation of a data source that conducts data export; and a storage program module for storing extracted information.
In one embodiment, at least one of the data sources is an enterprise system that stores information corresponding to enterprise data transactions. The data sources include a plurality of heterogeneous databases that store information relating to business transactions of the enterprise. In another embodiment, the system further comprises an external data source extractor program module for obtaining information relating to transactions from one or more data sources external to the enterprise system.
According to one aspect, a storage program module stores extracted information in a monitoring database. According to a related aspect, the storage program module stores extracted information in a staging database, and a mapper program module stores the extracted information in the monitoring database according to an enterprise ontology.
According to another aspect, the system further comprises an extractor file for identifying each data source and information required to access data in the data sources. The extractor file provides information identifying a source table, predetermined fields of the source table, a destination table, and predetermined fields in the destination table in which the source table fields are stored.
According to another aspect, the data sources of the system include one or more additional data sources that provide information supplemental to the monitored transactions. These additional data sources may include an external data source that provides data auxiliary to a particular monitored transaction, and/or an information technology (IT) system infrastructure equipment that provides data corresponding to IT system utilization and the IT system infrastructure equipment is selected from a group: file server, application server, firewall, router, intrusion detection equipment, authentication server. According to a related aspect, the additional data sources include an external data source that provides data ancillary to a particular monitored transaction.
In another aspect, the present invention relates to a method for extracting data relating to transactions from one or more external data sources and for providing data for use by a transaction analysis engine operative for executing one or more computer-executable compliance policy statements against extracted data. The method includes the steps of extracting an initial selected subset of information about monitored transactions from one or more data sources; extracting information about monitored transactions from a log file provided by one or more data sources; extracting information from one or more data sources from which information was previously extracted to resynchronize data in response to a resynchronization condition; receiving information provided by programmatic operation of a data source that conducts data export; and storing extracted information for access by the transaction analysis engine. In one embodiment, at least one of the data sources is an enterprise system that stores information corresponding to enterprise data transactions. These data sources include a plurality of heterogeneous databases that store information relating to business transactions of the enterprise. In another embodiment, the method further comprises the step of obtaining information relating to transactions from one or more data sources external to the enterprise system.
In one embodiment, the extracted information is stored in a monitoring database. According to an aspect of this embodiment, the step of storing extracted information comprises storing information in a staging database to minimize the computational load on the data source, retrieving the information from the staging database, and storing the extracted information in the monitoring database according to an enterprise ontology.
According to a related aspect, the method further includes the step of identifying each data source and information required to access data in the data sources in an extractor file. The extractor file provides information identifying a source table, predetermined fields of the source table, a destination table, and predetermined fields in the destination table in which the source table fields are stored.
In yet another embodiment, the data sources include one or more additional data sources that provide information supplemental to the monitored transactions. The additional data sources may include an external data source that provides data ancillary to a particular monitored transaction, and/or an information technology (IT) system infrastructure equipment that provides data corresponding to IT system utilization. The IT system infrastructure equipment is selected from the group: file server, application server, firewall, router, intrusion detection equipment, and authentication server. The additional data sources may also include an external data source that provides data ancillary to a particular monitored transaction.
In yet another aspect, the present invention relates to a system for monitoring transactions of an enterprise stored in one or more enterprise systems associated with an enterprise, the transactions corresponding to one or more business transactions of the enterprise, to determine possible lack of compliance of a business transaction with one or more predetermined policies of the enterprise. The system comprises an extractor for extracting an initial selected subset of information about monitored transactions from an enterprise system; a second extractor, responsive to changed data in the enterprise system, for extracting a second selected subset of information about the changed data; a monitoring database for storing the initial subset of information and the second subset of information as monitoring entities; a transaction analysis engine for executing one or more computer-executable policy statements against the monitoring database; and a user interface, responsive to a determination from execution of a policy statement of an exception to a predetermined policy for providing an output corresponding to the exception indicating a possible lack of compliance with the predetermined policy. In one embodiment, the selected subsets of information correspond to a predetermined selection of data fields from a predetermined selection of database tables storing business transactions. In another embodiment, the initial selected subset of information from the enterprise system comprises information corresponding to a complete set of business transactions current through a predetermined date. The second selected subset of information from the enterprise system comprises information corresponding to business transaction changes occurring subsequent to the predetermined date.
In yet another embodiment, the policy statements comprise computer-executable expressions of relationships between predetermined data items in the monitoring database, expressed in terms of an enterprise ontology, that resolve into indicators that determine exceptions. In a further embodiment, the extractors refer to stored extractor information identifying one or more data sources, access information for the data sources, selected data tables in the data sources, and selected fields in the selected tables. In yet a further embodiment, the extractors comprise computer program modules providing for a master extractor, a log extractor, a programmatic extractor, a resync extractor, an external data source extractor, and an information technology (IT) system extractor.
In a related embodiment, the system further comprises one or more additional data sources that provide information supplemental to the monitored transactions. The additional data sources include an external data source that provides data ancillary to a particular monitored transaction, and/or an information technology (IT) system infrastructure equipment that provides data corresponding to IT system utilization. The IT system infrastructure equipment is selected from the group: file server, application server, firewall, router, intrusion detection equipment, and authentication server. In a related aspect, the additional data sources include an external data source that provides data ancillary to a particular monitored transaction.
In yet another aspect, the present invention relates to a computer-implemented method for monitoring transactions of an enterprise stored in one or more enterprise systems associated with an enterprise, the transactions corresponding to one or more business transactions of the enterprise, to determine possible lack of compliance of a business transaction with one or more predetermined policies of the enterprise. The method comprises the steps of extracting an initial selected subset of information about monitored transactions from an enterprise system; storing the initial subset of information in a monitoring database as a monitoring entity; thereafter, in response to changed data in the enterprise system, extracting a second selected subset of information about the changed data; storing the second subset of information in the monitoring database as a monitoring entity; executing one or more computer-executable policy statements from a knowledge base against the monitoring database; and, in response to the determination from the execution of a policy statement of an exception to a predetermined policy, providing an output corresponding to an exception indicating a possible lack of compliance with the predetermined policy.
In one embodiment, the selected subsets of information correspond to a predetermined selection of data fields from a predetermined selection of database tables storing business transactions. In another embodiment, the initial selected subset of information from the enterprise database comprises information corresponding to a complete set of business transactions current through a predetermined date. The second selected subset of information from the enterprise database comprises information corresponding to business transaction changes occurring subsequent to the predetermined date.
In yet another embodiment, the policy statements comprise computer-executable expressions of relationships between predetermined data items in the monitoring database, expressed in terms of an enterprise ontology, that resolve into indicators that determine exceptions.
In a further embodiment, the extracting steps are effected with reference to stored extractor information identifying one or more data sources, access information for the data sources, selected data tables in the data sources, and selected fields in the selected tables and with computer program modules for a master extractor, a log extractor, a programmatic extractor, a resync extractor, an external data source extractor, and an information technology (IT) system extractor.
In yet a further embodiment, the method further comprises the step of extracting information from one or more additional data sources that provide information supplemental to the monitored transactions. These additional data sources may include an external data source that provides data ancillary to a particular monitored transaction, and/or an information technology (IT) system infrastructure equipment that provides data corresponding to IT system utilization. The IT system infrastructure equipment is selected from the group: file server, application server, firewall, router, intrusion detection equipment, and authentication server. In a related aspect, the additional data sources include an external data source that provides data ancillary to a particular monitored transaction.
From the foregoing claimed aspects of the inventions, it will be appreciated that use of the invention can reduce the cost of compliance with regulatory, contractual, and business process compliance, including ongoing Sarbanes-Oxley compliance, by continuously monitoring key controls including that required for Section 404 certification by auditors. Use of systems and methods constructed in accordance with the invention addresses the tangible costs of controls testing and remediation along with the opportunity costs associated with the internal distractions of compliance. Use of the present invention catches errors, fraud and internal control issues early in the transaction process so that corrections can be made before time is wasted duplicating and reversing work, before money is lost, and before controls are deemed deficient. By identifying the root causes of control violations and errors in real time, the present invention allows an enterprise to improve the quality of its earnings, ensure accountability, enhance business processes, and remediate any weaknesses for regulatory compliance.
To meet the heightened concerns about inside threats from systems-based fraud, misuse, and errors, the present inventors pioneered the concept and technologies of automated transaction integrity monitoring (“TIM”), according to systems and methods of the present invention. Unlike existing perimeter security solutions or access control systems, systems constructed in accordance with aspects of the inventions identify transactions where authorized users perform suspicious or otherwise noncompliant transactions within business systems. A TIM system according to the invention(s) analyzes transactions across multiple business applications to detects, prevents, and deters financial loss from systems-based fraud, misuse, and errors.
Embodiments of the present invention combine advanced data acquisition, data analytics, case management, and evidentiary analysis functionality. The disclosed systems and methods collect data across multiple platforms (including IT infrastructure as well as external data sources such as publicly accessible databases), and perform multi-perspective analysis to identify fraud, misuse, and errors. The system is capable of detecting problem transactions in several ways, including the use of multiple indicators of problems, linkage to related entities, tracking ongoing exceptions to policies that have not been resolved, and helping identify patterns that are associated with errors, misuse, and fraud. The system then generates high-impact reports, provides integrated case management, and enables evidentiary analysis.
The benefits of such transaction integrity monitoring are clear. First, such transaction monitoring establishes a business environment that deters employees and other insiders from committing business hacks. Transaction integrity monitoring provides the benefits of rigorous internal controls without the overhead. Even if procedural rules are not 100 percent maintained or employees learn to game the system, risk managers will be satisfied with a solution that keeps pace with real-time business transactions. Finally, transaction integrity monitoring acts as the ultimate layer of security from outsiders who penetrate the network as authorized users.
The data extraction methodologies utilized in aspects of the inventions employ multi-pass and multi-system query technologies. The system collects data across multiple platforms to correlate information from ERP systems, legacy mainframe applications, network monitoring solutions, and external data sources as relevant to many different types of enterprise applications, such as (but not limited to) accounts payable, accounts receivable, general ledger, human resources and payroll, customer relationship management (CRM), inventory management, email, electronic document storage and retention, and contracts management.
A system constructed in accordance with aspects of the invention also provides end-to-end case management and advanced investigative link analysis for high quality cases with irrefutable evidence. The case management system supports the collection and management of case-specific exceptions, clues, interviews, e-mails, and reports, in a secure work area to which only authorized users and/or administrators have access. This secure work area greatly increases an investigator's ability to quickly and thoroughly resolve multiple cases without sacrificing the legally required integrity of the process.
The advanced evidentiary analysis tools significantly reduce the investigative and forensics analysis workload. Complex link analysis of the case related subjects, systems, and accounts takes a fraction of the time associated with manual research and analysis methods. Furthermore, use of the system can assist in the recovery of lost assets.
Preferably, systems constructed in accordance with the inventions are designed with security as an essential element of a hardened appliance. The cases management system provides a digitally secure and trusted “evidence locker” that stores transaction records, the reasoning behind evaluations and activities associated with the investigation process. Other security features provided in embodiments of the invention include (1) encryption and authentication of communication channels, (2) out-of-band configuration options to block its visibility on a network, (3) a hardened operating system, and (4) support for authenticated external/supplemental queries into enterprise business systems and queries to external data sources.
More particularly described, the claimed invention(s) are particularly useful in connection with methods and systems for the continuous monitoring of data corresponding to transactions that are computerized or that have related data recorded or processed on a computer, a system of computers, or a computer network. Data traffic from multiple and possibly heterogeneous sources is continuously and repeatedly monitored for policy exceptions that may be clues indicating attention or investigation is in order with regard to particular transactions. Clues are potentially indicative of errors and omissions promulgated by mere operator mistakes, system malfunctions, or computer software failures. Furthermore, clues are potentially indicative of misuses and intentional or unintentional failures in compliance with policies. In the worst circumstances, clues are indicative of fraudulent activities.
In one aspect, the invention provides powerful data collection functionalities and abstracts information about transactions to a human user. The disclosed system provides automated data collection capabilities for seamless collaboration between automated and human selected searches. Heterogeneous data sources are utilized and correlation pre-processing sorts multi-level likelihoods of problem detection to alert a human operator or trigger pre-selected actions such as the assembling of detected events into case folders.
In another aspect, the invention provides data analysis capabilities using incremental evidence gathering and heuristic and collaborative reasoning. Automated monitoring techniques include enterprise policy rule sets to continuously monitor a data source and periodically or occasionally query a database for patterns indicative of non-compliant behavior or fraud schemes.
In another aspect, the invention provides case management tools such as user interface controls to empower an investigator with regard to directed queries and to provide implementation control with regard to automated searches.
In another aspect, the invention provides compliance monitoring to provide for synchronization between enterprise policies and operational activities. Process errors and inefficiencies are detected and tracked through human-managed automated monitoring and problem-pattern recognition.
In another aspect, audit compliance is ensured with internal controls that results in reduced asset loss, increased operational effectiveness, and increased corporate and shareholder confidence. Corporate governance is enhanced while addressing regulatory requirements related to Sarbanes-Oxley. Performance indices such as cumulated impact are determined and reported continuously or periodically.
A further aspect of the present invention solves a need in today's security marketplace by providing continuous controls monitoring of 100% of business transactions and also provides a solution for management to determine that internal controls are operating effectively. Data is acquired in near real time and analyzed optionally in an independent system to assure the integrity of the data.
Another aspect of the invention provides methods and systems for a collaborative reasoning engine (CORE). The CORE is coupled to a knowledge base that stores computer-executable policy statements in the form of XML frames. These statements or frames are provided in sets, with an initial set of base frames that addresses many common enterprise compliance policy scenarios. The base frames can be customized or supplemented with custom frames to address the specific compliance needs of the particular enterprise. Because the frames are expressed in the readily-understandable and modifiable XML, the system is robust, flexible, and permits ready and rapid adaptation to new compliance situations.
The CORE is operative to apply a set of policy-expressing frames to determine whether or not a particular transaction is deemed policy noncompliant. Through a combination of multi-system query and link-analysis techniques, a determination is made regarding likelihood of whether an event is indicative of fraud, errors, or misuse. Acts of concealment and conversion that are designed to circumvent standard auditing techniques are detected.
A further aspect of the present invention provides rapid continuous automated overview of internal activity across critical business functions. A variety of automated cross-platform data monitoring and collection techniques are utilized and complimented by a hybrid of analytic support modules to audit activity within business applications and to detect inappropriate and suspicious insider threat activity. Various categories of “business-hacker” activities are addressed in real-time. Non-compliant business activities recognized include but are not limited to fraud and theft, misuse and abuse, errors and omissions, and inappropriate system and control overrides.
Yet another aspect of the present invention provides measures of compliance of business transactions with company internal controls. Corporate governance scorecards and reports are generated. When transactions are detected as out of compliance, multi-level or iterative querying determines whether these transactions are errors, overrides, or activities that require further investigation. Process efficiency, revenue recognition and enhancement toward the ability to meet Sarbanes-Oxley requirements and any level of external or internal auditing requirements are provided.
From the foregoing, those skilled in the art will understand and appreciate that, with its transaction integrity monitoring, exception detection, analysis engine, and case management aspects, a system constructed in accordance with aspects of the inventions identifies fraud, misuse, and errors that directly affect the bottom line of an enterprise or the operations of an organization. The disclosed system combines the benefits of a systems auditor, fraud examiner, forensics auditor, and an information security specialist that is on duty “24×7” to monitor the effectiveness of internal controls. With its advanced analysis engine, the disclosed system identifies systems-based fraud, misuse, and errors in veritable real time. Rather than relying on periodic audits that sample transaction data, the disclosed system's transaction integrity monitoring identifies a problem the moment it occurs and prevents a perpetrator from covering his or her tracks. By identifying errors, misuse, and abuse in veritable real time, the disclosed system minimizes financial loss by allowing an organization to quickly and decisively respond. In many cases, use of the system allows an enterprise to close a hole before it can be exploited. Finally, with transaction integrity monitoring, the disclosed system empowers enterprises to “trust but verify” its financial transactions. The system allows an enterprise's management team to establish a “tone at the top” regarding expectations of conduct within the organization.
These and other aspects, features, and benefits of the present invention(s) will become apparent from the following detailed written description of the preferred embodiments taken in conjunction with the following drawings, although variations and modifications therein may be affected without departing from the spirit and scope of the novel concepts of the disclosure.
Prior to a detailed description of the invention(s), the following definitions are provided as an aid to understanding the subject matter and terminology of aspects of the present invention(s), and not necessarily limiting of the invention(s), which are expressed in the claims. Whether or not a term is capitalized is not considered definitive or limiting of the meaning of a term. As used in this document, a capitalized term shall have the same meaning as an uncapitalized term, unless the context of the usage specifically indicates that a more restrictive meaning for the capitalized term is intended. A capitalized term within the glossary usually indicates that the capitalized term has a separate definition within the glossary. However, the capitalization or lack thereof within the remainder of this document is not intended to be necessarily limiting unless the context clearly indicates that such limitation is intended.
Actor: an individual responsible for conducting business activity within an Enterprise, typically generating Business Transactions. The activities of Actors are monitored in accordance with the principles and operations of the present invention.
Administrator: a type of user of a system made in accordance with the invention that has special permissions or access to certain administrative or configuration functions, e.g. a knowledge engineer, trained technician, system operator, or other person who works with an Enterprise. Typically, such a person assists in system configuration, creates Base Frames, and modifies/customizes Frames to create Custom Frames for use in the present invention.
A/P: Accounts Payable, a financial system function.
A/R: Accounts Receivable, a financial system function.
Application: a computer program that operates on a computer system, e.g., but not limited to, a computer program of an ERP, CRM, or HR system operated by or on behalf of an Enterprise. Further examples include an accounts payable (A/P) program that is used by the Enterprise to pay its vendors, employees, etc.; customer resource management (CRM) and other customer support programs, employee information applications, accounts receivable programs, inventory management programs, enterprise data storage and management systems, email systems and servers, and virtually any other type of program that generates transactions.
Base Frame: a Frame that is basic, or universal, or generally applicable to a wide range of circumstances for a variety of different Enterprises. For example, there is a strong correlation of fraud when an employee of a company is listed as a vendor within the company's payables system; a Base Frame that determines if an employee identifies himself or herself as a vendor for receiving payment is generally believed to be applicable to a wide variety of Enterprises. A set of Base Frames is preferably provided with an initial configuration in preferred embodiments of the present invention.
Business Transaction: a Transaction reflecting or representing business activity of an Enterprise, typically represented by one or more data fields of information stored in a database of the Enterprise, e.g. an HR or ERP type database. Generally, an Actor generates Business Transactions.
Case: A collection or repository of information representing one or more Exceptions and other related information, to facilitate investigations, monitoring, compliance tracking, etc. A Case is useful in collecting a Chain of Evidence that might be useful in policy enforcement or discipline. Generally synonymous with Preliminary Inquiry or Investigation, which may be considered different levels of a Case. Also can be referred to as a Case Folder.
CEV: comprehensive exception view, a user interface view of a particular exception including information related to this exception such as exception ID, description, priority, potential impact, owner, related entities, etc.
Chain of Evidence: a collection of related information that is or may be used to demonstrate that the current records match the claimed reality, i.e. proof that evidence has not been intentionally or unintentionally modified or corrupted.
Changed Entity: An Entity that has changed, e.g. if data corresponding to a business transaction has been changed, such as the invoice number of a Transactional Entity corresponding to a real invoice, the data representing the change is a Changed Entity.
Clue: something that leads one toward the solution of a problem. A Clue can be one or more Indicators or one or more Exceptions. Indicators and Exceptions are species of Clue, but a Clue may consist of information other than Indicators or Exceptions.
Collaborative Reasoning Engine or CORE: a component of a system or method in accordance with the invention that executes Frames in accordance with the invention. It is a particular inventive species of a Compliance Rules Engine. Also Transaction Analysis Engine.
Compliance: the state of consistency and adherence to a policy, as reflected by one or more Frames.
Compliance Rules Engine: a system and/or software operative for executing one or more logical rules against a collection of data, to determine whether there is a violation of data rules. This is a generalized concept; see also Collaborative Rules Engine.
Compliance Policy Statement: an expression of a policy of an enterprise, typically in the form of a computer-executable series of statements and expressions, expressed in a computer-executable language such as XML frames. See also Frames.
Confidence: a function, generally mathematical in the present invention, of the quality and quantity of certain Indicators. In certain aspects of this invention, Confidence is a probability that an Exception or Case represents an actual impact or real-world event. Confidence can be express as a numerical value or a percentage; Confidence may be compared to a threshold value in order to establish an Indicator or trigger an Exception. Confidence can also be expressed in cumulative or summary terms such as high, medium or low.
CRM: customer relationship management; relates to aspects of interaction that an enterprise has with its customers. Many aspects of CRM are now computerized and generate various transactions, e.g. inquiries from prospective new customers, contact management functions, customer list maintenance, help desks and customer service representative monitoring, email organizers, web site interactivity, product returns and credits, couponing and rebates, online chat functions, instant messaging, etc.
Customer Control Objectives: narrative statements of the policies of an Enterprise with respect to some topic or management goal or objective. Such narrative statements may be represented by and maintained in a document or table (e.g. a database table), provided by the Enterprise. Customer Control Objectives are expressed formalistically in Frames or Compliance Policy Statements.
Custom Frame: a Frame that is especially created or modified for a particular Enterprise, to reflect circumstances or policies applicable to that particular Enterprise. A Custom Frame may be created by modification of a Base Frame and run in lieu of a particular Base Frame that is not applicable to that Enterprise, or a Custom Frame may be created from scratch.
DBMS: database management system.
Entity: something that has a separate and distinct existence or conceptual reality outside the present invention and a lifetime beyond a single business transaction. As used most often in this document, an Entity is the representation of data in a document or other aspect of an enterprise's business processes. An Entity has two main species: Transactional Entity and Support Entity. See also Changed Entity. An Entity can also be Static Entity or a Transient Entity. A Transactional Entity can exist in a Monitored Database or in a Monitoring Database.
Enterprise: an organization or business entity that utilizes the present invention; such an Enterprise will usually have one or more computer systems running one or more Applications, for which compliance monitoring in accordance with the invention is effected. An Enterprise can be a business, a government agency, a person, or virtually any other organization that conducts business transactions reflective of its business activity.
Enterprise Database: a database associated with an Enterprise, typically storing Transactions, Business Transactions, etc.
ERP System: Enterprise Resource Planning system, generally the software, system, and/or Applications responsible for planning and tracking the financial, logistical and human operations of an Enterprise.
Estimated Impact: a mathematical term, Confidence*Potential Impact.
Exception: an indication and representation of data corresponding to a possible violation of an Enterprise Policy. An Exception can occur from a single incident or action, or a collection of incident or action. In accordance with aspects of the invention, one or more Exceptions occur or are triggered as the result of the settings or values of one or more Indicators from the execution of one or more Frames, in response to determination by the logic of a Frame that something has occurred that is suspicious or noteworthy and might be indicative of a lack of compliance. There can be multiple related Exceptions corresponding to a policy violation. There can also be an aggregate Exception (a Super Exception) that itself consists of multiple Exceptions.
Exception Handling: the processing of one or more Exceptions. A User can flag or identify an Exception with a status such as detected, under review, dismissed, or fixed. Also relates to the storage and processing of Exceptions within database or store of Exceptions, such as a Case.
Extraction or Extract: a process of retrieving selected information from the databases of an Enterprise.
Frame: a computer-executable logical representation of a rule or set of rules, determined by a User (typically an Administrator type user) responsible for establishing compliance monitoring processes to implement a Policy, as applied to data or information reflecting one or more transactions or one or more data items of transactions. A specific implementation of a Compliance Policy Statement. In the preferred embodiments, a Frame is represented by an XML frame, although other computer-implemented mechanisms may be utilized. Typically, a Frame includes logic responsive to values of one or more Indicators to generate Exceptions. See also Base Frame and Custom Frame.
Fraudster: a user of a system who uses the system to perpetrate a fraud.
HR: Human Resources. When referring to a system, typically means a computer system or Application operative to maintain information about personnel within an organization (such as employees), for example, payroll information, health care insurance information, retirement benefits information, etc.
Impact: see Potential Impact.
Indicator: a signal, variable, marker or pattern of data that corresponds to, represents information about, and/or constitutes a component of an Exception. Typically, one or more Indicators make up an Exception. In aspects of this invention, a Frame contains computer-executable logic that processes data representing Indicators and generates Exceptions. An Indicator typically relates to a specific control activity and is designed to represent a specific control objective or Policy of an Enterprise. An Indicator may be a cumulated value, and can itself be determined from other Exceptions. An Indicator may also be an individual transaction or set of transactions that when detected and analyzed are indicative of misuse, abuse, or fraud, or other lack of compliance with a Policy.
Inquiry: see Preliminary Inquiry.
Investigation: an official and systematic process of determining the facts surrounding one or more Exceptions. In accordance with aspects of the invention, an Investigation can be stored in and represented by a Case.
Knowledge Base: a collection of Frames representing the compliance policies of the Enterprise, e.g. Policy Frames, stored in a data store or database. In accordance with aspects of the invention, a collection of XML data files that comprises computer-executable Frames, as well as data or tables associated with Extraction and with mapping of extracted data into the Monitoring Database.
Linking: the notion of retrieving and processing a number of related Entities, such as Transactional Entities, to assist in a broader review of transactions associated with particular Subjects, Support Entities, etc. For example, if a particular transaction such as a payment to a particular vendor is discovered to be duplicative of another, earlier payment to the same vendor, linking would allow the retrieval of other payments to that vendor, or identification, retrieval, and display of other payments authorized or initiated by the party that created the duplicate payment(s), or retrieval of other transactions such as invoices or other documents related to that particular vendor.
Link Analysis: correlating different business transactions or exceptions for the purpose of discovering or demonstrating a pattern of activity.
Mapping: a process of correlating data items identified in one manner with data items identified in a different manner. For example, relating a data item from an ERP database identified in that database as CUSTOMER_NAME with a data item in another database (such as the Monitoring Database) identified as VENDOR.OST.
Monitoring Database: a database or DBMS that stores a selected subset of information derived from an Extraction of predetermined information relating to Transactions that are monitored in accordance with aspects of the invention. Typically the Monitoring Database is maintained separately and independently of any database that stores Transactions or information relating thereto. Entries in this database are referred to as Monitoring Entities.
Monitoring Entity: an entity comprising a data item in a Monitoring Database, typically comprising a selected subset of information from a Monitored Database. Monitoring Entities can be Transactional Entities or Support Entities.
Monitored Database: a database or DBMS that stores information relating to Transactions those are to be monitored in accordance with aspects of the invention. Also called a Transactions database, or Enterprise database, or Source database.
Monitored Entity: database entries or item in a Monitored Database; can be a Transactional Entity or a Support Entity.
Normalize: a process of transforming data items expressed in a first data item naming schema (e.g. of an enterprise database) into data items expressed in a different data item naming schema (e.g. associated with a monitoring database). May also involve combining one or more data items or fields in the first schema to a single data item or field in the second schema (or vice versa), reducing or expanding the characters count of the data item or fields, changing the units, changing the data type, etc. See also Mapping, Ontology. E.g. Data items may be normalized by mapping them into a different naming and data storage schema, in accordance with ontology.
Ontology: A collection of data and/or metadata, somewhat like a dictionary, that provides for creating relationships and/or interoperability between things that have different names, e.g. a data field in one database might have a field identifier CUSTOMER_NAME, while the same information in another database might have the field identifier PERSONNAME. Ontology would have a list of each item, CUSTOMER_NAME and PERSONNAME, with pointers to each other, thereby defining the relationship. Used in Mapping. An ontology may be, but is not necessarily, constructed with known ontology construction tools such as the Resource Description Framework (RDF), which is a general framework often used to describe a web site's metadata or the information about the information on the site.
Policy: a statement reflecting or representing the manner in which an Enterprise is to abide by rules or guidelines of behavior, sequence of operations, protocols, requirements for information, regulations, laws, or other indicators of actions.
Policy Frame: a Frame expressing or indicative of a Policy. May be expressed in XML, but can be expressed in other computer-executable form. In various aspects of the invention, it should be understood that to some extent all Frames are Policy Frames. Also, a Compliance Policy Statement.
Potential Impact: the potential loss (typically monetary) associated with an Exception or Case. Also referred to simply as Impact.
Preliminary Inquiry: a process of gathering information about one or more Exceptions to determine if a formal Investigation is needed or to be conducted.
Priority: the relative importance of an Exception or a Case. The default priority for a detected Exception may be based on the Estimated Impact. May be represented as “High”, “Medium” or “Low”, with numerical values, or with alphabetical values.
Private Flag: an indicator or flag that an Exception or Case should not be included in search results or summary reports unless specifically requested; provided for access control that is limited to authorized Users only.
Record: in database parlance, a single instance or data item, usually consisting of one or more fields of information, each field typically having a field identifier identifying what the information in that field represents. An array of records is often referred to as a table.
Source Database: another term for Monitored Database.
Staging Database: a special database that receives data from an extraction operation and holds the data temporarily prior to a mapping operation.
Subject: person(s) and/or system(s) associated with a particular Exception or Case managed by embodiments of the present invention. Is typically an Actor.
Super Exception: an aggregate Exception that comprises multiple Exceptions, i.e. the logic of a Frame may be responsive to the occurrence of one or more previously generated Exceptions, possibly from the execution of other Frames.
Support Entity: an Entity that is persistent over a relatively long period of time. An Entity typically relating to an Actor and/or the subject matter of a business activity, e.g. vendors, employees, customers, products, third party service providers, service personnel, etc. and their associated information is considered Support Entities. Support Entities typically are static entities that have a longer persistence than Transactional Entities. Support Entity is sometime referred as Static Entity, See also Transactional Entity.
Table: a particular collection of database records in a DBMS, having predetermined data fields. A typical relational DBMS may be viewed as a plurality of tables or grids of information, with each row in the table or grid corresponding to a record, and each column of the table corresponding to a particular data field or data type. The term is typically used in the conventional DMBS sense.
Transaction: a set of system actions that result in a completed business activity. For example, a transaction includes the actions associated with adding or deleting a new vendor within an A/P system, or changing the name of an existing vendor from one name to another, or creating a purchase order. A transaction can relate to a Support Entity or a Transactional Entity. See also Monitored Transaction.
Transaction Analysis Engine: another term for CORE.
Transactions Database: see Monitored Database; generally synonymous therewith.
Transactional Entity: an Entity that has a relatively short lifecycle compared to a Static Entity; an Entity relating to a transaction and its associated information, typically corresponding to activities of or in a business process, for example, purchase orders, vouchers, payments, shipping records, service requests, change orders, etc. and their associated information are considered Transactional Entities. Also can include other activity that is not strictly business-process related, e.g. information technology (IT) infrastructure information of a transactional nature such as is provided by computer, networking, and telecommunications equipment such as firewalls, routers, intrusion detection devices, user authentication systems, application servers, and other similar equipment. Is typically a Transient Entity. See also Support Entity, Monitoring Entity.
User: an individual or other entity that accesses or uses a compliance monitoring system constructed in accordance with aspects of the invention. Typically, a user is an administrator who works for the enterprise, such as a policy Administrator or person who receives reports indicative of Exceptions, Events, or other failures of compliance. Typically, a User is not an Actor or Subject or other person subject to an Investigation.
UI: User Interface. Typically means a software Application with which a User interacts for purposes of entering information, obtaining information, or causing functions of an associated system to execute.
Wariness: indicates a level of suspicion of an Exception or Entity. Can build up or accumulate as a function of Confidence, Potential Impact, and/or Priority, or of Indicators.
The embodiments of the present invention are preferably implemented as a special purpose or general-purpose computer including various computer hardware, as discussed in greater detail below. Embodiments within the scope of the present invention also include computer-readable media for carrying or having computer-executable instructions or data structures stored thereon. Such computer-readable media can be any available media which can be accessed by a general purpose or special purpose computer. By way of example, and not limitation, such computer-readable media can comprise physical storage media such as RAM, ROM, EEPROM, CD-ROM or other optical disk storage, magnetic disk storage or other magnetic storage devices, or any other medium which can be used to carry or store desired program code means in the form of computer-executable instructions or data structures and which can be accessed by a general purpose or special purpose computer.
When information is transferred or provided over a network or another communications connection (either hardwired, wireless, or a combination of hardwired or wireless) to a computer, the computer properly views the connection as a computer-readable medium. Thus, any such a connection is properly termed a computer-readable medium. Combinations of the above should also be included within the scope of computer-readable media. Computer-executable instructions comprise, for example, instructions and data which cause a general purpose computer, special purpose computer; or special purpose processing device to perform a certain function or group of functions.
Those skilled in the art will understand the features and aspects of a suitable computing environment in which aspects of the invention may be implemented. Although not required, the inventions will be described in the general context of computer-executable instructions, such as program modules, being executed by computers in networked environments. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform particular tasks or implement particular abstract data types. Computer-executable instructions, associated data structures, and program modules represent examples of the program code means for executing steps of the methods disclosed herein. The particular sequence of such executable instructions or associated data structures represent examples of corresponding acts for implementing the functions described in such steps.
Those skilled in the art will also appreciate that the invention may be practiced in network computing environments with many types of computer system configurations, including personal computers, hand-held devices, multi-processor systems, microprocessor-based or programmable consumer electronics, networked PCs, minicomputers, mainframe computers, and the like. The invention may also be practiced in distributed computing environments where tasks are performed by local and remote processing devices that are linked (either by hardwired links, wireless links, or by a combination of hardwired or wireless links) through a communications network. In a distributed computing environment, program modules may be located in both local and remote memory storage devices.
An exemplary system for implementing the inventions, which is not illustrated, includes a general purpose computing device in the form of a conventional computer, including a processing unit, a system memory, and a system bus that couples various system components including the system memory to the processing unit. The computer will typically include one or more magnetic hard disk drives (also called “data stores” or “data storage” or other names) for reading from and writing to. The drives and their associated computer-readable media provide nonvolatile storage of computer-executable instructions, data structures, program modules and other data for the computer. Although the exemplary environment described herein employs a magnetic hard disk, a removable magnetic disk, removable optical disks, other types of computer readable media for storing data can be used, including magnetic cassettes, flash memory cards, digital video disks, Bernoulli cartridges, RAMs, ROMs, and the like.
Computer program code that implements most of the functionality described herein typically comprises one or more program modules may be stored on the hard disk or other storage medium. This program code, as it known to those skilled in the art, usually includes an operating system, one or more application programs, other program modules, and program data. A user may enter commands and information into the computer through keyboard, pointing device, or other input devices (not shown), such as a microphone, joy stick, game pad, satellite dish, scanner, or the like. These and other input devices are often connected to the processing unit through known electrical, optical, or wireless connections.
The main computer that effects many aspects of the inventions will typically operate in a networked environment using logical connections to one or more remote computers or data sources, which are described further below. Remote computers may be another personal computer, a server, a router, a network PC, a peer device or other common network node, and typically include many or all of the elements described above relative to the main computer system in which the inventions are embodied. The logical connections between computers include a local area network (LAN) and a wide area network (WAN) that are presented here by way of example and not limitation. Such networking environments are commonplace in office-wide or enterprise-wide computer networks, intranets and the Internet.
When used in a LAN networking environment, the main computer system implementing aspects of the invention is connected to the local network through a network interface or adapter. When used in a WAN networking environment, the computer may include a modem, a wireless link, or other means for establishing communications over the wide area network, such as the Internet. In a networked environment, program modules depicted relative to the computer, or portions thereof, may be stored in a remote memory storage device. It will be appreciated that the network connections described or shown are exemplary and other means of establishing communications over wide area networks or the Internet may be used.
Referring now to the drawings, in which like numerals indicate like elements or steps throughout the several drawing figures,
Briefly summarized, the TIM system 100 is connected for computer-to-computer communications with an enterprise resource planning (ERP) system 110 which in turn is connected to and stores data in one or more data sources such as ERP database systems 120, e.g. databases 121 a, 121 b, . . . 121 n. Users 101 of the TIM system interact with the system via a user interface (UI) comprising a personal computer or terminal and associated display for configuring the system, constructing and maintaining the information such as policy statements, ontology mappings, extraction requirements, etc. (collectively referred to as “knowledge maintenance”), and receiving analysis and reports in a manner that will be described later.
ERP systems 110 with which the embodiments of the present invention are operative include both disparate heterogeneous and stand alone computer systems that run individual ERP applications on behalf of an enterprise, such as account payables systems, human resources systems, accounts receivable, general ledger, inventory management, and like. However, it will be understood that integrated ERP applications (see
Data generated from enterprise transactions is stored in ERP database system 120, such as databases 121 a, 121 b, . . . 121 n. Such databases are considered monitored databases in accordance with aspects of the invention. Transactions generated by the various applications in the ERP system 110 are stored in these databases, and information from such databases is extracted, processed, and analyzed in accordance with aspects of the invention. Information stored in the monitored databases 120 is considered monitored entities, which can take the form of support entities or transactional entities, as defined herein. As described elsewhere herein, support entities are typically static entities that have a relatively long persistence within an enterprise such as employee names, customer identification, vendor name, etc. Transactional entities, on the other hand, are typically transient entities that exist in a plurality relative to one or more support entities, e.g. a vendor of an enterprise may generate numerous invoices and receive numerous payments over a period of time.
It should be understood that electronic transaction can be considered a transactional entity and processed in accordance with the invention, and considered as a part of assessing policy compliance. For example, the creation of a vendor in an accounts payable system is a transaction, as is creation of a new employee in an HR system; both of these become support entities after their transaction is completed.
In accordance with aspects of the inventions, monitored entities are extracted from the monitored databases 120 and processed in accordance with the inventions as will be described. Such monitoring entities correspond to transactions generated within a data source such as ERP system 110. Transactions within an ERP system 110 are created and/or handled by an actor 102, such as an employee, executive, consultant, other authorized individual, or other person that interacts with the ERP system via a computer terminal such as that shown at 111. In accordance with the aspects of the inventions, an actor 102 who is responsible for a transaction within the system 110 but violates policies—and thereby constitute exceptions in accordance with the techniques described herein—may become a subject of an investigation. As will be discussed in greater detail, transactions associated with various actors, or subjects, may be cumulated and viewed in accordance with aspects of the case management features as described herein. For example, if a particular actor 102 works in an enterprise's accounts payable function and operates an A/P program module such as 114 a, he or she may generate one or more payments to various vendors, e.g., vendor 1, vendor 2, etc. Under certain circumstances, rules reflective of enterprise policy can be constructed to determine if improprieties occur in the identification of vendors, or providing unauthorized or excessive payments to vendors, or changing a vendor name to a different entity and changing it back, and other known fraudulent schemes.
Still referring to
An extractor 140 is operative to interface with the various data sources such as monitored databases 120 and retrieve, be provided, or otherwise obtain data from such data sources and monitored databases. Extracted data from extractor 140 is stored in a staging database 155, which temporarily stores data so that the TIM system can operate out of band with respect to enterprise applications and thereby minimize performance degradation of the monitored ERP systems. The extractor is responsive to and/or configured by extraction data (stored as extraction files a knowledge base 165) to determine an appropriate extraction methodology for a particular type of ERP database from which data is extracted and monitored.
The extractor 140 comprises several subcomponents 141. A programmatic extractor 141 a is operative to provide information from an ERP system (such as provided by the known ERP system provider SAP, located in Walldorf, Germany), that provides or exports data automatically rather by extraction. A master extractor 141 b is operative for an initial data load to extract an entire predetermined data set, typically a limited set (subset) of information or “snapshot” of data within an enterprise at a predetermined point in time, and to store that initial and “master” data set in the staging database 155 or in the monitoring database 175. A resync extractor 141 c is operative to synchronize data in the monitoring database 175 to a master set of information in the monitored databases 120 so as to minimize the likelihood that a lack of synchronization between data sets will create issues. A log extractor 141 d is responsive to the provision of audit data logs by certain types of ERP database systems, in the form of records indicating the addition or change of records within particular ERP data. An environmental source extractor 141 e is operative to obtain data from an enterprise's environment 133 (e.g. internal systems such as its information technology (IT) infrastructure). Finally, an external source extractor 141 f is operative to access and retrieve information from external data sources 132 via an external network 131 such as the Internet. As will be described in greater detail, access to information in external data stores such as proprietary databases, publicly accessible databases, and the like that may be required in enterprise policies to provide certain integrity checks.
The environmental source extractor 141 e and external source extractor 141 f obtain data from additional data sources 130. Such additional sources are typically supplemental and additional to primary sources of transaction data such as ERP systems. In accordance with an aspect of the invention, these additional data sources include the environmental sources 133 and external data sources 132. The environmental data sources, such as the enterprise IT infrastructure, can include devices such as server logs 134, intrusion detection systems (IDS) 135, firewalls 136, and routers 137, as well as telephone systems, caller ID logs, physical security access devices, and other types of devices (not shown) that generate electronic information indicative of use or activity.
A mapper 150 is operative to retrieve data from the staging database 155 and normalize, transform or map that information into a predetermine format to comprise monitoring entities, which are then stored in a monitoring database 175. The monitoring database 175 stores monitoring entities, both support and transactional, identified by table and field names in accordance with mapping data stored in a knowledge base 165. The mapping data (e.g. in the form of mapping files and ontology files) establishes relationships between monitoring entities stored in the monitoring database and monitored entities from the ERP databases. A principal function of the mapper 150 is to transform data from various and disparate (and possibly heterogeneous) data sources into a shared schema or ontology, so that an analysis engine can examine and correlate data across the disparate systems and facilitate the preparation of policy statements that consider information from different data sources.
A knowledge base 165 stores information required by the extractor 140 (extraction data in the form of extractor files), information required by the mapper 150 (mapping data in the form of mapping files and ontology files), and a plurality of computer-executable policy statements or frames 167 that constitutes the rules and/or logic for determining exceptions. A collaborative rules engine (CORE) 160, also called a transaction analysis engine, is operative in accordance with aspects of the invention to execute policy statements, which constitutes one or more logical rules and/or expressions, against the monitoring database 175 to determine whether there is a violation of policies or rules (i.e. exceptions), in a manner that will be described in greater detail below. Output from the CORE 160 in the form of exceptions resulting from execution of frames 167 is stored as one or more exceptions in an exceptions database 185. These exceptions, as will be described, constitute indications of violations of enterprise policy that are reflected by and maintained in the frames 167.
By “computer executable policy statement,” it is generally meant that the policy statements are expressed in a computer-readable form, such as an XML frame containing data, commands, and logical expressions that resolve to an outcome such as an exception, etc. Typically a policy statement can be “executed” in the sense that an interpreter can parse the file and determine what computer-operations should be effected. The policy statement is therefore a form of computer program or data for use in a computer program. A number of equivalents to the XML expression will be apparent to those skilled in the art, for example, a computer program in a conventional programming language such as C++, various types of data structures, scripting languages, object modules, rules statements, and other forms of expression that can be interpreted and executed by a computer.
An analysis & reporting server 180 provides a user interface (UI) to users, e.g., users 101, for purposes of receiving reports regarding exceptions and implications thereof, as well as providing a user interface to a case management component 190 also as will be described in greater detail below. Users 101 can be of different types, having different authorizations for different purposes. For example, certain users such as user 101 a may have access to receive reports and manage certain cases or learn of certain exceptions, while another user 101 b may be considered an administrator, with authorizations to set up other users, access and edit enterprise policy statements (or enter them initially), configure the mapping data and extraction data, etc.
The foregoing discussion of
As in a more distributed, disparate, heterogeneous environment like that of
Those skilled in the art will understand and appreciate that the aspects of the present invention described herein that are shown in
From the foregoing, it will be appreciated that the principles of the present invention are applicable in both distributed and integrated environments, in that monitored entities are extracted and mapped to constitute one or more monitoring entities stored in the monitoring database as described herein, so as to allow the detection of exceptions that not in compliance with policies or procedures of the enterprise.
Certain functions of the present invention may be dispersed and separated, and need not be in the same physical location. For example, the extractor 140 and mapper 150 may be dispersed and located at various remote sites and in different configurations. Specifically for example, the extractor 140 a and mapper 150 a in connection with ERP 1 can be physically located at the ERP site 301 a, instead of locally proximate to the TIM system 100. Similarly, the extractor and mapper for an ERP system 301 b can be located within or proximate to the TIM system 100, such as shown at 140 b, 150 b. Likewise, the functions of extraction and mapping can be separated by locating the extractor 140 c at a remote ERP site 301 c, with the mapper 150 c located proximate to the TIM system 100. Those skilled in the art will understand and appreciate that these principle functions of extraction and mapping are readily dispersible, for convenience, and need not be physically located at or near the other components of the TIM system.
It will be understood from the outset that the examples shown in
In this example, the first enterprise computing system ERP 1 has a database 121 a storing a data structure or table 402 representing an exemplary vendor table for a typical A/P system, having a plurality of data records 410 a, each record having columns or field identifiers such as name, address, vendor ID, bank account number, contact, etc. In many ERP systems, a vendor would be considered a relatively static entity, because vendor information tends to persist over an extended period of time.
The second and different enterprise computing system ERP2 has an associated database 121 b relates to the human resources function, which stores a data structure or table 404 comprising of a plurality of data records 410 b that identify employees of the enterprise. Each record of such an employee table 404 has field identifiers such as name, address, bank account number, social security number (SSN), and other related information. Typically, employees are also considered static entities, because data records associated with employees tend to persist within the records of the organization over an extended period of time.
The example of
Invoices are an example of transactions that are transient entities, as the term in understood in the present invention, because such entities typically tend to exist in greater numbers relative to static entities, are related to one or more particular static or support entities, and there tend to be a plurality of such entities respect to a single static entity, e.g. a single vendor may have issued multiple invoices to the enterprise over a predetermined period of time.
It will be understand that the present invention is operative with both static and/or support entities, as well as transient and/or transactional entities, in that different instances or versions of each of these entities may exist over time. In accordance with the invention and as shown by way of example in
Either static or transient entities, but especially static entities, can exist over time as different versions of the same conceptual entity. Conceptually, a particular vendor is always the same person, but that vendor's name may change, its address may change, its bank account may change, its telephone number may change, but the conceptual entity remains the same. Over time, the data representing this entity may change. Some ERP systems do not provide for robust version change tracking, or provide adequate controls or supervision over changes to entities. Unauthorized or hidden changes to information about static entities is known as a classic pattern of fraud for many organizations and enterprises, and a typical subject for an enterprise policy that detects certain types of changes to static entities, or requires supervisory permission and an auditable record.
It will be understood that the creation of any entity, whether static or transient, is the creation of a transactional entity. Creation of a vendor or new employee (both relatively static entities) involves a transaction that is initially a transactional entity (the data involved in the creation); that entity may later be considered a support entity. Issuance of an invoice or purchase order involves a transaction that is a transactional entity. A change to a vendor or employee also involves a transaction that is a transactional entity, as does a change to an invoice or payment. Any entity becomes a support entity and can be referred to or utilized in policy statements.
For example, an entity such as a vendor can originally be created with a first address, thereafter changed to a different address at a later point in time, and thereafter changed again to provide a different bank account number for receiving payments, and thereafter changed yet again to undergo a name change from a corporation to an LLC. Throughout such changes, of which three have been given as an example, the basic entity remains conceptually, logically, and legally the same, yet certain information relating to the entity was changed over time.
Similarly with respect to transactional entities such as invoices, an invoice will typically have a predetermined and fixed invoice number, but could have different and changeable information such as date, amount, items identified, and the like.
In accordance with aspects of the inventions, the state of a given entity is captured in an initial or master extraction so as to provide an initial data set representative of data in the enterprise at an initial point in time (e.g. a snapshot). Thereafter, each change to a certain entities, whether static entities or transient entities, as required by enterprise policy statements, is captured by operation of the invention and stored as monitoring entities in the monitoring database and analyzed in accordance with policies of the enterprise, as expressed in the policy statements that execute on the monitoring database, for purposes of detecting exceptions in the form of anomalies, duplications, duplicate payments, violations of company policy, and the like.
Still referring to
Subsequent extractions of information from a table in a monitored database, for example, the vendor table 402, is represented at different points in time by different data entries and stored in the monitoring database 175, for example in a vendor table 430 that represents different versions of the same entity (Vendor 1) at different points in time. Considering the example of the modified vendor discussed above, which occurred at different points in time t1, t2, t3, etc., the subsequent extraction comprises a plurality of records relating to vendor 1 at different points in time. For example, the change to the address of vendor 1 subsequent to the master extraction at time t2 appears as a separate entry in the monitoring database, and the change in bank account number of vendor 1 that occurred at time t3 is yet another entry in the monitoring database 175. It will thus be understood and appreciated that operation of the present invention allows the capture of the state and/or status and/or versions of a particular entity, such as particular vendor, as it might exist in various points in time during the life of that entity. Operation of the policy statements against the various entities at different points in time allows the comparison of fields of information for different “versions” of the same entity, which thereby facilitates the determination of improper actions with respect to particular entities.
Although the foregoing example of a vendor change is described in connection with a static entity such as a vendor and a vendor table, it will be understood and appreciated that the same principles apply to more transient transactional entities such as invoices, checks, vouchers, insurance claims, refunds, returns, and virtually any other type of transaction maintained within the data processing systems of an enterprise.
In accordance with aspects of the invention, transactional entities that originate within enterprise systems (such as vendors, invoices, etc.) have a virtual counterpart as transactional entities in the monitoring database. For example, support entities such as vendors in source table 402 are mapped into a corresponding table(s) such as vendor table(s) 420, 430 in the monitoring database 175, and entities such as invoices in source table 406 are mapped into a corresponding table such as invoice table 440 in the monitoring database 175.
With the foregoing principles relating to changes to static entities over time, turn next to
In the example of
As an example of the operation of the invention, consider that the enterprise has a policy that indicates an exception in the case of a vendor address change and bank account change followed by a changeback within a short period of time. Such an exception would be determined and reported in accordance with aspects of the invention. Although this example is given in the context of a financial type transaction, it should be understood that invention has applicability to other types of transactions including security applications, IT infrastructure, personnel management, health care, banking and financial institutions, logistics, and virtually any time of activity that involves an electronic record or transaction.
Still referring to
The initial metadata of actor, timestamp, and revision number is associated with initial selected data 515, to create the initial record for vendor Widget Company during the master extraction. It will of course be appreciated that subsequent changes to various information associated with the vendor Widget Company includes different metadata identifying the actor, timestamp, and an assigned revision number, to each subsequent change, for record keeping purposes.
Those skilled in the art will understand that the system 100 will require initial configuration with predetermined information so as to allow their various functions to execute. A system constructed in accordance with the present invention(s) is configured by an administrator or other authorized user, prior to operation. Such configuration involves establishment and expression of the enterprise policies in one or more enterprise policy statements, determination of the manner of extraction of data from monitored databases and providing extraction data, and determination of the mapping or normalization of data in the monitored databases to the enterprise ontology, so that extracted data can be stored in the monitoring database for out-of-band operations and exception detection and management.
The process 700 includes several subsidiary program functions, including ConfigureExtractor, ConfigureMapper, ConfigureCore, and ConfigureWorkbench. These program functions or routines provide for configuration of the extractor 140, mapper 150, CORE 160, and aspects of the analysis and reporting functions 180, respectively.
The ConfigureExtractor routine includes steps for specifying enterprise systems or other data sources from which data is acquired or provided, as well as specifying primary key fields, field identifiers, filters, context, and parameters of accessing remote data sources. From
The ConfigureMapper routine includes steps for specifying parameters and aspects of the staging database 155, the monitoring database 175, the ontology files, entities involved or identified in an ontology, mappings of data items for entities, and other contexts and parameters. Configuration of the mapper as utilized in the present invention comprises identification of particular data tables and fields of data provided by an extraction from a monitored database, and maintenance of the relationship between such particular tables and fields of the monitored database with corresponding tables and fields of the monitoring database. Such information is reflected by and stored in the enterprise ontology. The ontology in particular represents the mapping of fields from monitoring databases into particular selected fields of the monitoring database.
The ConfigureCore routine includes steps for defining a set of policy statements or frames, including identifying the transaction involved in the policy, any required support entities, any indicators of the policy, and other frame or policy parameters. This routine enables a user to access to stored policy statements (as expressed in XML frames in the disclosed embodiment) so as to create new statements, update existing policy statements, activate or deactivate particular statements, change the sequence of statement (frame) execution, and provide any other required administrative functions for determining or modifying the logic or expressions associated with frame execution.
The ConfigureWorkbench routine includes steps for configuring access to the monitoring database (to obtain related entities), setting usernames and permissions, and configuring any reporting functions of the system. This routine enables a user to input or correct information associated with the analysis server, such as the configuration and contents of reports, alarms, status, cases, and other information associated with the provision of information relating to exceptions as determined by operation of the TIM system 100. Further details are provided below as to specific user interfaces relating to case management and reporting functions.
The process 700 operates in a continuous loop to monitor for user input, receive authorization information in the form of user name and password, test for a condition for actuation of one or more of the configuration functions, and permit access to the desired configuration function through a user interface (not shown) applicable to the informational requirements of the functions being configured.
Turn next to
A system constructed as described herein pulls data from various and sometimes heterogeneous sources, such as ERP systems, network logs, and other enterprise systems, hereinafter referred to as “source” systems, transforms the information via the enterprise ontology, and populates the monitoring database, also called the “target” system or database.
It will be appreciated that data corresponding to monitoring entities must be pulled frequently, preferably continuously, but not necessarily in real time. This minimizes the chance that fraudster can cover tracks by changing data back. Near real time data extraction is preferably in the range of several minutes, although this time period is arbitrary. In cases where data is retrieved by way of a detailed log (e.g., an audit log), frequency of extraction is less important since all changes to the ERP databases are logged anyway.
As discussed above, if updates to entity information are pulled via a change log, preferred embodiments provide for initial population of the target database utilizing the mapping transformations required by the enterprise ontology. Furthermore, because of the possibility that the source and target data will get out of synchronization if a change log is exclusively used, the resync extractor 141 c provides for resyncing the data between a source database and the target database. Those skilled in the art will understand that re-syncing is a snapshot mechanism, so frequency is an important parameter. Non-log changes are believed to occur in batch jobs, so that daily or nightly resyncing is preferred.
Applications of the present invention will draw data from enterprise source systems comprising ERP system manufactured by various vendors. It is important not to disrupt the mission-critical operational system of an enterprise. Preferably, therefore, data extraction is usually in the form of simple queries, without additional constraints, joins, or follow-on queries.
In some aspects of the invention, target entity tables in the staging database or in the monitoring database are not necessarily exactly copies of the source tables. Table names, fields, data types, and even values may have to be transformed according to the system ontology. In accordance with aspects of the invention, each update in a source entity that is monitored should result in creation of a new “revision” created in the target or monitoring database. This advantageously allows the TIM system to retain and reason about past updates. Further details about this mapping are provided elsewhere.
The programmatic extractor 141 a is operative in connection with data sources that do not provide for direct query access to data contained in the databases, for example certain SAP ERP systems. Further details of the programmatic extractor are provided below.
The master extractor 141 b is operative for an initial or master data extraction, at initial start up of a TIM system 100 constructed in accordance with the present invention. According to certain aspects, an entire initial dataset, which may consist of entire data tables representative of a “snapshot” of the state of the enterprise data at initial point in time, is obtained by the master of extractor 141 b and provided to the staging database 155. Those skilled in the art will understand and appreciate that the master extractor 141 b is preferably a high-speed data interface that retrieves information at the maximum rate possible from connected ERP systems with immediate storage in the staging database 155, without any further data processing operations, to minimize the data processing load on the ERP system. Subsequent operations of mapping may therefore be done “out of band” relative to the operation of the ERP systems, to minimize the impact on time responsiveness and other operations. Further details are provided below.
The resync extractor 141 c provides for a synchronization operation between the monitoring database and any associated monitored (source) databases, typically on a periodic basis, to compensate for loss of data synchronization that can occur over time, e.g. by reliance on log files. Further details are provided below.
The log extractor 141 d provides for data extraction based on information provided by a data source in the form of a log or audit file, which typically contains selected information identifying a change to an entity. Further details are provided below.
The environmental source extractor 141 e provides for data extraction from data sources associated with the system environment of an enterprise, e.g. its IT infrastructure such as firewalls, intrusion detection devices, routers, servers, etc. Such information is typically additional or supplemental to entity information, and can provide corroborating evidence for instances of errors, fraud, or misuse. For example, a repeated pattern of failed username/password attempts to an enterprise system might be indicative of a hacking attempt. If a user is finally successful in gaining access to a system (i.e. guessing the password), and thereafter perpetrates a fraudulent payment request in an A/P module using the same username, the log file of failed access attempts indicates that it is likely that the fraudulent payment transaction was likely generated by the same person that hacked the password after several failed attempts.
The external source extractor 141 f is an interface to external data sources, e.g. public and/or proprietary data sources, that might provide supplemental information utilized in implementing certain enterprise policies, e.g. verification of a driver's license number, or a telephone number in an online telephone directory, or other similar data source that is not necessarily a direct part of the enterprise's computing infrastructure.
The disclosed embodiments of the TIM system 100 and methods thereof are operative with various different classes of data sources. One type of data source is a synchronous polling type database, such as PeopleSoft. In the case of this type of data source, the extractor preferably makes a JDBC connection to a database table and pulls its contents into the staging and/or monitoring database for further processing.
A second type of data source is a synchronous polling type database with an RFC/API. For this type of data source, the extractor uses an external system API and protocol to poll for new information that can be retrieved, transformed, and loaded into the staging and/or monitoring database. This type of data source includes SAP R3, and constitutes a programmatic type extraction.
A third type of data source is an asynchronous event broker type system. For this type of data source, the extractor must register with a data broker that sends events asynchronously as transactions as state changes occur in the enterprise system. Examples of this type data source include PeopleSoft EIP and SAP Netweaver type event brokers.
A fourth type of data source is an asynchronous file sending type system, such as SAP R3 ABAP cluster table. For this type of data source, the extractor initiates a program or script on an external machine that provides the actual extraction from the data source. Upon completing the extraction the external program sends the data to the extractor. This type of data source also constitutes a programmatic type extraction.
Each data extractor 141 makes reference to extractor data in the form of an extractor file stored in the knowledge base, for information specific to the data source from which data is extracted. In this regard,
From the foregoing, it will be understood and appreciated that aspects of the invention involve an initial data load to obtain an initial set of data that is analyzed for exceptions, followed by a “detect changes” type operation to minimize the load on the ERP systems by extracting changes to certain entities, as reflected by changed data.
As mentioned above, a programmatic extractor 141 a is utilized for enterprise databases that do not permit direct extraction. Those skilled in the art will understand that certain types of enterprise systems do not permit direct access to data tables or information stored therein. For example, certain SAP products do not provide an interface for external queries to data stored within its databases. However, the SAP product provides a scripting language that allows users to write programs to cause the export of selected data for external use. The SAP system and other similar systems output information in response to internal execution of such a script or program. The programmatic extractor 141 a therefore represents the combination of (a) a scripting element operative internally to systems such as SAP that do not provide direct data querying, for internal retrieval of selected data and exporting such selected data, (b) a communications interface or file transfer mechanism for communicating data exported, and (c) a software component in the extractor 140 responsive to exported data from a scripting type export operation, for receiving this data and transmitting the data to the staging database or for other utilization, e.g. in the case of staging database bypass for small amounts of data that do not require staging to minimize performance degradation.
The system SAP R3 does not provide a data extraction interface that allows direct integration with an application program interface (API) that would permit direct data extraction. Nor is it possible to integrate directly with the underlying SAP database due to the presence of proprietary cluster tables. As with other types of ERP systems, SAP R3 contains many tables that will contain information comprising entity data. With some systems, a synchronous database query protocol can be used to periodically replicate tables of interest from SAP R3 into the monitoring database. It will then be possible to reuse existing algorithms to handle merge and loading behaviors.
However, many high volume data entities are stored in SAP R3 as “cluster” tables. These kinds of tables essentially store information as a column set for a table, in a single compressed proprietary data block inside a column in the cluster table. Other columns in that roll contain metadata such as key fields and values so that the “cluster” of data can be accessed and read or written as needed. The cluster table storage method used by SAP presents an issue for data extraction because there is no published specification for the compression algorithm or the data format used in the cluster. In such cases, and with other similarly constructed systems, information stored in cluster data can be made in R3 open SQL interface. An open SQL interface is available to ABAP programs as an internal API.
Accordingly, it will be understood that programmatic extractor 141 a is invoked externally, but essentially comprises an independent process effected by a script or code 1210 that runs within a clustered type environment such as that provided by SAP, and returns a reply containing requested information to the programmatic extractor 141 a. In this manner, the programmatic extractor 141 a provides a synchronous source interaction with data storage systems of this nature.
Other conditions for triggering a master extraction can occur, for example, in the event that a system administrator determines that the monitoring database is grossly out of sync with a monitored database and elects to “start over” with a new master data extraction. Other equivalent conditions for a master data extraction will occur to those skilled in the art.
A log file might, or might not, provide the actual information that is desired for storage in the monitoring database. If so, no further data extraction operations would be required. If not, the log extractor 141 d executes instructions to query the relevant database and obtain the needed data.
As known to those skilled in the art and as illustrated in
A re-sync or resynchronization operation essentially ensures consistency of data between a data source table and a corresponding table in the staging database. Such an operation is typically run periodically so as to ensure that the source tables and staging database table reflect the same information.
The resync process 1700 includes steps for determining new and modified rows in a table of a data source, and creating new rows in the staging database for any new records added that were not previously detected or identified in a log file. The process typically applies to each data source table that has a counterpart in the staging database. A procedure DetermineNewAndModifiedRows is operative to compare the contents of the tables in the data source table and the corresponding staging database table and determine if any new rows have been added or any fields of existing rows were changed. If so, the staging database table is updated.
Although not directly illustrated, extraction processes similar to those described above are effected to obtain additional data from additional data sources such as external data sources 132 and environmental data sources 133 (
With the foregoing extraction processes and structure having been described, turn next to
For example, the vendor account creation transaction 1802 generates an Account Created transactional monitoring entity 1813, bearing revision 1, with a timestamp as shown, with the actor indicated as John Doe. This is an example of a static monitoring entity in that the creation of a vendor account within an ERP system will tend to persist for an extended period of time. The subsequent transactions with monitoring entity names PO Issued 1814, Received Purchase 1816, Invoice Received 1818, and Payment Issued 1820 are considered transient or transactional entities. In accordance with aspects of the invention, each of these transactional entities is recorded in the monitoring database as monitoring entities, with appropriate metadata.
According to an aspect of the invention, information required by the mapper to conduct its mapping and transformation operations is provided in an ontology file and a mapper file retrieved from the knowledge base. An exemplary ontology and mapper files are discussed elsewhere herein.
According to another aspect of the invention, information required by the mapper for its operations is provided in an ontology target table 1912 that provides information that identifies field names and parameters of the data, after mapping, as the data is stored in the monitoring database. The ontology target table 1912 also identifies what metadata 1915 is associated with each entity provided by an extraction. Examples of metadata include but are not limited to a revision identifier (Revision_ID), a unique entity identifier (Entity_ID), an entity version (Entity_Version), a person or actor associated with the transaction (Actor_ID), and a timestamp or time identifying the last update to the entity (Update_Time). Other types of metadata can also be utilized for other purposes, for example, a unique object identifier could be generated and assigned to the entity as it is reflected in the monitoring database.
The mapper carries out the following basic steps: (1) map an ERP source table name to a target table name in monitoring database; (2) map a source table field name to a target table field name in the monitoring database; (3) if required, map a predetermined set of source field values into single target field value (e.g., if multiple address lines are provided in a source table that are mapped to a single data item or field in the target table); (4) collapse any required “child” tables into a single parent entity and a target table in the monitoring database; (5) populate metadata fields in the target table including revision identifier, entity identifier, entity version, actor, and update time; (6) for each new entity revision for an entity preexisting in the monitoring database, duplicate the existing monitoring entity and create a new revision, i.e., in the monitoring database table.
If a change log approach is used for information from the ERP databases, which often records individual field updates in ERP databases as separate entries, changes to a particular entity that occurred within the same transaction (as determined by the change time and actor ID) are preferably combined into a new entity revision.
The mapper 150 may further require a query to a source database in the event that certain information needed to create a monitoring entity is not immediately available from an entry provided. For example, an address record update for a static entity such as a vendor or employee might not contain a reference to the particular address information that constituted the address change. If the ERP system is so constructed, this information must be obtained via a query from the source database. It will be understood that information could potentially be obtained from the target or monitoring database, obviating a copy of additional information from the ERP system.
It will be appreciated that the mapping operations conducted by the mapper 150 could, were such operations not segregated to separate computing processes, create an additional load on the ERP systems data processors. In accordance with the aspects of the invention, the staging database 155 provides a cache of information from the source databases, thereby facilitating mass copy operations from the ERP system. Such a caching architecture is more computationally efficient than conducting mapping operations in real time as data is received, and reduces the data processing load on ERP systems. This is believed to provide a significant architectural advantage for embodiments of the present invention. It will therefore be appreciated that this architecture minimizes the complexity and computational load on any component that runs remotely from the TIM system 100. Furthermore, the partition of functionality into logical steps of extraction, caching, and mapping provides for an architectural separation of functions and asynchronous operation to improve performance.
It will thus be appreciated that an asynchronous architecture with staging database caching allows the processes of extraction and mapping to optimize the rate at which data is provided from ERP systems, and that mapping operations can occur out-of-band (i.e. offline), at times when ERP systems have reduced workload, to improve performance.
The process 2000 comprises steps for creating tables in the monitoring database, retrieving data from the staging database, conducting transformations and renaming of selected fields in accordance with the enterprise ontology, and storing the transformed data in monitoring entity data records in the monitoring database. A process CreateTargetTables reads an ontology file corresponding to the particular data source that is providing entities for processing, creates a table in the monitoring database, creates meta data for the table such as pointers to related entities and predetermined text that displays in connection with the entity.
A process Mapper comprises steps for opening a connection to the staging database 155, opening a connection to the monitoring database 175, reading a corresponding mapper file to obtain information required for the data renaming and transformation (which may include base mappings provided with the system as well as custom mapping resulting from a customization operation to a base mapping), and validating the mappings to ensure consistency with the ontology. Then for each table in the monitoring database that is to receive transformed/mapped data, steps are taken to query/join source staging tables to retrieve particular data items or fields, perform any necessary table or field transformations, compute any additional fields, mark the previous version (revision) of the entity as “old”, save the new fields as a new current version (revision) in the monitoring database, and set a “new” flag to indicate that a new monitoring entity is ready and available for processing by a policy statement.
The mapping table or enterprise ontology, it will be recalled from earlier discussion, stores information and establishes the mapping between predetermined tables, fields and parameters of a source (monitored) database and corresponding tables, fields and parameters in the monitoring database. The mapping, transformation, and renaming of data from the source data format (monitored database) to the target data format (monitoring database) is also referred to as normalizing.
The mapper is responsive to an ontology source table 2111 to select only the needed data items from the source database in table 2101 and store those items as identified by the field names shown in the target table 2121, in the monitoring database 175. Also shown in
In addition to the selected subset of information from the source table 2101, and as was previously described, predetermined metadata is added to each entity created and stored, e.g. Revision_ID, Entity_ID, Entity_Version, Actor_ID, Update_Time. It will be further understood that the mapper 150 is responsive to a stored ontology mapping predetermined source table data fields to predetermined target table data fields and including predetermined metadata, for purposes of creating monitoring entities in the monitoring database, on which the policy statements are executed in order to identify a possible policy of violations.
All of the specifications in the various knowledge base files can be modified and overridden element by element by more customized knowledge files as described below. In addition, individual files include a context mechanism allowing individual tags with in the knowledge files to be filtered on a run context, which is specified when the system is installed and run for an enterprise. These are only used of the context specified matches that for the specific installation. Finally, individual parameters can be set in the knowledge files, values for which are also set as installation for run time. This provides a third means of customizing knowledge files, including extractions, mappings, and frames.
The CORE process 2500 comprises steps for retrieving computer-executable policy statements (frames) from the knowledge base 165 and executing them on data (monitoring entities) in the monitoring database 175. The preferred data structure for the computer-executable policy statements or frames is described in greater detail elsewhere. These steps include opening a connection to the monitoring database, retrieving and reading any base frames that are configured to run, retrieving and reading any custom frames that are configured to run, and validating the frames for syntax. Then, for each frame so retrieved, the system computes any required statistics tables, and for each new version (revision) of a monitoring entity, and for all applicable and required support entities, evaluates all indicators in the frame. Such operation may require retrieving additional information from additional data sources, as discussed elsewhere. If the evaluation of any indicator produces a confidence level that exceeds a predetermined threshold, a new exception is created. Exceptions can be based on the absence of specific entities as well as on specific patterns of data in existing entities. Exceptions can also be based on the results of previous exceptions, thereby proving a form of chaining reasoning wherein subsequent transactions can be used to build on the conclusions related to previous transactions. This new exception is added to the exception database, in a format (data model) described elsewhere herein. Any required potential impact and probabilities are computed. A natural language description of the exception is generated based on a template in the frame, all basis revisions are determined and saved, and a wariness value for each entity underlying the exception is updated based on exception. Basis revisions are the entities (actually a specific revision of the entities) upon which an exception is based. This includes the single transactional entity (i.e. the new revision of an entity that has changed) and any supporting entities (revisions thereof) required to match the pattern in the frame. These are used to provide a comprehensive exception view to the user in the UI (see below). Additional computation is performed to identify transitive links between entities and exceptions, which are used in the link analysis functionality in the UI.
The right hand side of
In accordance with aspects of the invention, a frame can be constructed to track the logical business activity steps and impose the discipline (i.e. enterprise policy) of a particular sequence in a business process of an enterprise, and to indicate a policy exception in the event that a portion of a business transaction sequence is out of order. Those skilled in the art will understand that information pertaining to each of these particular activities involved of the process must be reflected as transactional entities in a system of the present invention, and that appropriate metadata such as timestamps associated with the particular transactions must be inspected so that a determination of sequence can be established. For the foregoing, those skilled in the art will understand that many other business activity and transaction policies can be implemented by utilizing the techniques and methodologies of the present invention.
Note the group of transactions shown at 2720, which is provided as an example of an overridden transaction. The expected sequence of business transactions would be creation of a vendor account 2701, purchase order processing 2702, purchase receiving 2706, invoice and vouchering 2707, and payment issued 2708. An inappropriate or invalid set of transactions might occur if a certain actor within the enterprise changed the vendor's bank account number at 2721, issued a payment at 2722, and then changed the vendor's bank account number at 2723 back to the original number that was input during the vendor account creation. This is an example of a typical fraud where an insider employee changes bank account numbers or other payment type information of a vendor so as to misdirect or misappropriate funds from the enterprise, and then attempts to “cover his tracks” by changing the bank account or other payment information back to the initial invalid setting.
According to aspects of the invention, each version of particular entities, and in this case the static entity of vendor with corresponding bank account number, is preserved as an historical record. A frame expressing the exemplary policy statement that “any changes of bank account numbers of vendors followed by changeback within a predetermined time period constitutes an exception to be investigated” can be easily written and utilized in embodiments of the invention. Such a frame or a group of frames expressing related policies can be executed to inspect a pattern or series of changes to particular entities, with the purpose of determining if policy exceptions have occurred.
In the example described, the enterprise has a policy that such an exception should be indicated. The right hand side of
In the example of
In accordance with aspects of the invention, the illustrative exceptions 2800 are shown with an identifier and/or description of the nature of the exception, the time, actor identification, and status. Further explanation of the nature of indicators, status, and other aspects of stored exceptions are described elsewhere herein.
In accordance with aspects of the invention, frames in the present invention are implemented as XML code, which those skilled in the art will understand is a computer programming expression methodology that is computer-executable by a number of different commercially-available XML processing engines that can be utilized in constructing embodiments of the present invention. As will be known to those skilled in the art, XML stands for eXtensible Markup Language and bears certain similarities to the well known hypertext markup language (HTML). The XML language is designed to describe data itself and attributes thereof, and has a known structure having “tags”, document type definition portions (DTD), and an XML schema. Each logical expression, statement, attribute, or other identifier in an XML document is delimited by a starting tag in the format of “<tag>” and concludes with a closing tag in the format of “</tag>”. The information between the starting tag and closing tag sets forth a statement, attribute, description, computer command, formula, or other information that can be parsed by an XML interpreter and executed by a computer.
As those skilled in the art will understand how to construct statements and frames using XML, no further discussion of the particular implementation methodology using XML will be provided herein.
It should, however, be understood that other computer-implemented mechanisms, languages, scripting methodologies, program modules, and/or devices could be employed in constructing embodiments of the present invention to express enterprise policies in other forms of policy statements, access information in the monitoring database, apply the rules of such policy statements, determine exceptions, etc, so as to identify and determine exceptions and store them in the exception database.
The exemplary frame of
In accordance with aspects of the inventions, each frame such as the exemplary frame 2900 processes data identified in terms of the enterprise ontology, as stored in the knowledge base, with respect to entities that are monitored in the present invention, and determines indicators based on the processing of data on the monitoring database. In the example of
The logical expression that would detect whether the vendor telephone number is the same as the employee telephone number is set forth in the <expression> tag within the <indicator> tag. In response to the detection that a vendor telephone number is the same as an employee telephone number, a new exception is generated. According to aspects of the invention described elsewhere, the new exception comprises a new data record for storage in the exceptions database 185. The CORE process generates the exception and provides information needed to create the exception record including an exception ID, a name, the identities of all related transactional and support entities, a confidence calculation (if employed). Specific details of the exception are provide in connection with
In connection with executing the exemplary frame 2900, an impact parameter may be provided. In accordance with aspects of the invention, the impact parameter is a predetermined amount that reflects an anticipated or expect financial impact or effect of the indicator. This impact may be the particular hard dollar amount associated with the transaction, if a transaction has an amount, or may be an arbitrary value. The impact value facilitates prioritizing the exception. The exemplary impact in
Each frame typically includes one or more transaction tags. These identify particular transactional entities involved in the frame execution. In the example of
Each frame typically includes one or more entity tags, e.g. the entity tag in
Each frame typically involves one or more indicator tags, e.g.
The indicator typically includes one or more logical expressions, identified by the <expr> tag, which contains a logical expression of a condition that if satisfied indicates an exception. The exemplary expression
The indicator typically includes a confidence value, identified by a tag <conf>, which is an arbitrary number or probability that the associated indicator is revealing a compliance policy violation. The example in
A frame typically includes one or more <cev> tags, which represents the term “comprehensive exception view”. This information is provided with an exception as information associated with the exception and how to display that information to a user. In particular, the <cev> tag provides information that enables the frame of
In accordance with other aspects of the invention, a plurality of different indicators and frames can execute to determine more complex scenarios, to thereby obtain a greater certainty in establishing a compliance policy violation. In the example of
For example, if the vendor telephone number and employee telephone number are found to be the same (with confidence value of 0.20), and the respective tax ID (e.g. social security numbers) are found to be the same (with confidence value of 0.20), and addresses are found to be the same (with confidence value of 0.20), and a number of payments have been found to be the same (with confidence value of 0.20), a cumulative confidence factor of 0.80 may result (assuming simple cumulation). By comparing the cumulative confidence level to a predetermined value such as 0.75, as reflected by another frame that executes in conjunction with the frame 2300, a more complex logical expression may be constructed so as to provide for a number of different logical, financial, and other checks to reflect and represent enterprise policy.
As described elsewhere herein, the confidence values are preferably combined in a statistically appropriate manner rather than simple accumulation as described above. See the discussion in connection with
In accordance with aspects of the invention,
In like manner, a first custom frame 3112 possesses a <frameoff> tag to indicate that this frame should not run, while a second custom frame 3014 possesses a <frame> tag to indicate this frame should run.
At runtime, and as shown at 3140, the preferred CORE process 160 retrieves a set of frames from the knowledge base 165 and executes the frames in a predetermined sequence. In the event that a particular frame should not execute, it would possess a <frameoff> tag. Thus, it will be seen in the runtime sequence at 3140 that base frame 1 3002 executes, custom frame 2 3114 executes, base frame 3 3106 executes, etc., while all frames possessing a <frameoff> tag are not executed. In this manner, a predetermined set of base frames may be called and executed, may be selectively turned off, and may be modified so as to reflect particular circumstances of a particular enterprise and execute in place of a different base frame, or new frames may be created, as desired by a system administrator.
Consider that a monitoring entity 3202 for a payment was generated in monitored system. Assume that the vendor associated with the payment is a preexisting support entity having a vendor record 3204, and that this entity (ID=5678) possesses an initial wariness value of 300 (expressed in arbitrary terms, but perhaps dollars). A payment (transaction 3202) is issued to this vendor in the amount of $1000, which triggers the execution of a frame 3206 that is intended to test for some condition relating payments to vendors (not specifically illustrated). Within this frame 3206, there exist two indicators related to the vendor (the logic for which is not shown but assumed), and their confidence values are c1=0.2 and c2=0.3, as shown in the abbreviated frame, respectively. The confidence C of this two-indicator exception may be calculated as shown in step 3208:
The priority of the exception is calculated as follows:
where in this case the impact is the payment amount ($1000) of the triggering payment entity 3202. The exception has a priority value of 440, which in this example results from the application of the confidence level (which may be considered a probability) to the nominal impact of the transaction (which in this case is the amount of the transaction, $1000). Since the exception is related to the vendor, the wariness of this particular vendor is increased by 440 after this exception is generated. Therefore, the wariness of the vendor is now increased to 740, the sum of the initial wariness and the priority, as shown in 3210.
From the foregoing, those skilled in the art now understand how to determine various types of indicators, exceptions, wariness, confidence, impact, and priority in accordance with the present invention for many different types of transactions, not just financial transactions.
Assume now that the “Vendor—Employee Similar” frame determined that some aspects of the newly created vendor and an employee of the enterprise were similar, for example, the addresses of the two entities are the same. According to the logic of the frame, this creates an exception 3414 (step 3), “Vendor—Employee Similar”. As discussed above in connection with frames, this exception will have an associated confidence level.
Assume next that a purchase order is generated at 3404. This creates a transactional entity, which is related to (connected) to the vendor created at step 1. Assume another frame that is designed to examine purchase orders to determine if any vendors associated with a purchase order already are connected to any exceptions (step 4). According to the logic of this second frame (or another indicator in the first frame), another exception 3416 is created (step 5). This exception is described as “Suspected Ghost Vendor,” and will also have an associated confidence level.
Assume next that a frame or indicator determines that no other employees within the company other than a single employee have ever bought from the vendor 3408, i.e. there is a sole buyer 3410 in the system's database associated with this particular vendor. The fact that only a single person within the organization has ever bought from this particular vendor indicates that the vendor is very new, very small, or very suspicious. Assume that an indicator of “sole buyer” increases the confidence level associated with the exception 3416 of “Suspected Ghost Vendor” (step 6). In accordance with aspects of the invention, the exception 4016 receives an incremental boost to its existing confidence level.
Assume next that a frame or indicator is operative upon additional data from additional data sources, for example, the system authentication function of the enterprise's IT infrastructure. As shown in
The foregoing has illustrated a number of different aspects of indicators, frame execution, increase of confidence level, exceptions, transactional and support entities, and the like, and demonstrates to those skilled in the art that systems and methods of the present invention may be adapted to address myriads of enterprise compliance policy situations.
As shown in
From the discussion above, it will be understood that the CORE 160 is operative to executive frames from the knowledge base on the monitoring database 175 and to generate one or more exceptions that are representative of the violation or possible violation of enterprise policies. Exceptions take the form of data items generated by the CORE 160. Principally, each exception data item includes information reflecting the nature of the exception as determined by particular frame that generated the exception, time information, actor identification information, and a status indicator. The case management system 190 provides analysis of exceptions and reporting of exceptions, so as to facilitate the provision of information to users regarding generated exceptions, to provide a mechanism for storing collections of exceptions, and to manage investigations in the form of a case. A case typically comprises a plurality of exceptions with status relating to such exceptions, to permit the enterprise to monitor and control its business processes.
In accordance with aspects of the invention, the case management system 190 is a computer-implemented process that retrieves information from the exceptions database 185 and the monitoring database 175 and presents the information to users, e.g., users 101 as shown in
There are three major aspects to the analysis component of the case management system of the invention. The first is a set of detailed displays of exceptions including an exception summary list, and for each exception an automatically generated natural language description, several detailed views of the entities underlying the exception, and a display for various other parameters describing the exception. The second is a similar display for entities, which displays a summary list of entities optionally sorted by their composite wariness score, and a detailed view for an entity including a display for all the exceptions that the entity supports as well as a display for various attributes of the entity. The third is a user-configurable summary display that displays aggregate statistics for exceptions and entities, graphs, and alerts.
The Entities tab in the lower display region is shown selected, which causes display of information associated with the entities that were involved in triggering the exception of duplicate vouchers with the same amount. A description field 3515 shows information associated with the exception, and identifies the indicators that triggered the exception, e.g. that “Exactly two VoucherLines were entered for the same Vendor for the same amount within 14 days.” This information is provided from the description information in the exception data record, as described above.
Note that a popup control (effective on a right click in embodiments of the invention) is shown at 3506. Assume that the cursor was positioned over the Voucher ID data field. Assume that the user selects the command “Show Related Entity” at 3508. In accordance with aspects of the invention, this will cause display of entities that are related to the particular voucher, for example, an associated purchase order (PO).
Turn in this regard to
As another example of link analysis provided in aspects of the invention,
In the screen 4200, a number of particular transactions relating to exceptions are displayed, to illustrate certain linking functions of certain aspects of the present invention. According to such aspects of the invention, a collection of transactions related to particular actors or to particular static entities are collected and displayed in a timeline chronology. In the example of
For example, the transaction 4201 relates to the creation of a vendor identified as AA, while transaction 4203 represents the creation of a second vendor identified as ABBC. In the example illustrated, assumed that the two transactions 4201, 4203 resulted in the utilization of the same address for two different vendors. Assume further that a policy frame is provided to determine whether duplicate addresses exists for different vendors, and generate an exception as a result the processing of an appropriate frame. This would be indicated at as the exception 3505 in the chronology, “Opportunity Exception (Duplicated Address)”, together with a date and time of detection, and displayed with the filled-in circle icon.
Advantageously, the different display techniques allow users to readily identify and analyze exceptions and their related transactions, and graphically display the relationships between various occurrences of transactions and exceptions over time.
Another example of a detected exception is shown at 4213, identified as “Ghost Vendor Scheme Detected.” A predetermined display indicator 4215 is provided to indicate that supervisory personnel or auditor have been notified regarding the exceptions.
According to aspects of the invention, the Impacts column is cumulated to form a total, as indicated in by the Total display area 4306.
Selector buttons are also provided so as to allow cumulating the calculated impact(s) for a predetermined time intervals, e.g., Date, Week, Month, Year. The Month selector is activated at 4302, indicating that the cumulated total impact of the exceptions assigned to this particular user represents the impact for the particular month in question.
In order to illustrate certain principles involves in the disclosed embodiments and aspects of the present invention turn to
The various steps of interest in the exception processing are identified as numbers within circles, reflecting steps 1-9 of a process for identifying a Ghost Vendor exception and establishing a case to handle the determination of the exception. At step 1, assume that an employee (actor) 102 of the enterprise changes the bank account number of a particular vendor ID 1234 from 5678 to account number 8899. The data relating to this change will be stored in the ERP database 120, and in particular in the vendor table 4815.
At step 2, a transaction containing the information relating to this change is acquired by the extractor 140 in accordance with aspects of the present invention. Assume that information indicating changes to entities is provided in the form of an audit log, stored in an audit log table 4817 in the ERP database. An audit or transaction log data item 4801 is provided to the data extractor 140 via the audit/transaction log. At step 3 the extractor 40 queries the ERP database 120 to retrieve other information (related entity information) regarding Vendor 1234, in accordance with the extractor information. At step 4, other information from the query back to the ERP database is provided, including the new bank account number for Vendor 1234.
At step 5, a corresponding monitoring entity 4803 is created in the monitoring database 175. At step 6, a frame 4805 reflecting an enterprise policy is retrieved from the knowledge base 165 and provided to the CORE 160. Assume that the frame 4805 expresses the policy that an exception should be indicated if a vendor bank account number is the same as an employee bank account number. At step 6, the CORE 160 determines from execution of the frame 4805 that an exception has occurred, and stores an exception entry 4810 in the exceptions database 185.
At step 7, the case management process 190 retrieves (or is provided) the exception and generates a case report indicating that a Ghost Vendor exception has been determined, with provision of information relating to impact and confidence level. At step 8, information relating to the Ghost Vendor exception is displayed to a user 101 via the case management/analysis and report UI 180. At step 9, optionally, the extractor retrieves additional information relating to vendor 1234 from the ERP database in a subsequent extraction operation (e.g. from access to external data sources 131), so as to obtain further details about vendor 1234 or about the employee, for use in connection with a case or investigation.
The foregoing description of the exemplary embodiments of the inventions has been presented only for the purposes of illustration and description and is not intended to be exhaustive or to limit the invention to the precise forms disclosed. Many modifications and variations are possible in light of the above teachings.