WO2016146005A1 - Method and device for correcting attribute values of commodity background attribute - Google Patents

Abstract

A method and device for correcting attribute values of commodity background attribute, which belong to the field of computer communication technologies. The method comprises: acquiring a commodity identifier of each of a number N of the commodities (101); dividing a number N of the commodities into a number M of the shareable background attribute commodity subsets (102); counting the number of occurrences of each original attribute value of the same type of background attributes of all the commodities included in each shareable background attribute commodity subset (103); according to the number of occurrences of each original attribute value of the same type of background attributes of all the commodities included in each shareable background attribute commodity subset, determining corrected attribute values of the same type of background attributes of all the commodities included in each shareable background attribute commodity subset (104); and modifying each original attribute value of the same type of background attributes of all the commodities included in each shareable background attribute commodity subset, as the corrected attribute value (105).

Description

Method and apparatus for correcting attribute values of background attributes of goods

Cross reference related reference

The present application claims priority to Chinese Patent Application No. 201510119332.6, filed on March 18, 2015, entitled,,,,,,,,,,,,,,,,,, in.

Technical field

The present invention relates to the field of computer communication technologies, and in particular, to a method and apparatus for correcting attribute values of background attributes of commodities.

Background technique

With the development of computer communication technology, more and more merchants sell goods through the Internet. When selling goods through the Internet, the background attribute of the goods is important information describing the goods. The background attributes of the products affect the search, shopping guide, recommendation and other products presented to The effect of the consumer, so the background properties of the product are important. However, there are a lot of existing products, such as missing and filling in the attribute values of the background attribute, for example, the background attribute "style" of the female bag category (this attribute includes more than 30 kinds of shell bags, Cambridge bags, bowling bags, etc.) The attribute value) is an example. When the consumer searches through the keyword "seashell bag", or clicks on the "seashell bag" label in the shopping guide path, or expects the recommendation system to recommend more "seashell bags", if a product is substantially a shell bag but is missing the style The attribute value of the attribute, so that when the consumer searches through the keyword "seashell bag", the item will not be displayed to the consumer, causing a miss recall; if a product in the platform is substantially other styles (such as a tote bag) The attribute value of the style attribute is incorrectly filled into the shell package, which causes the consumer to search through the keyword “seashell bag”, or click on the “seashell bag” label in the shopping guide path, or expect the recommendation system to recommend more “shell bags” for it. At the time, the item was incorrectly presented to the consumer, causing a false recall. The missed recall caused by the attribute value of the background attribute will cause the impression that the platform product is not rich to the consumer; the wrong recall caused by the incorrect attribute value of the background attribute will cause the platform search, shopping guide or recommended product effect to be inaccurate to the consumer. impression. Therefore, it is necessary to periodically modify, correct, and so on the attribute values of the background attributes of the product.

The existing method for correcting the attribute value of the background attribute of the commodity is to manually discover the problem by means of manual sampling or user report, and then urge the merchant or the operation of the second to manually make corrections such as supplementation and correction.

However, the existing method of correcting the attribute value of the background attribute of the commodity mainly relies on manual completion, and the efficiency is very low.

Summary of the invention

In order to solve the problems of the prior art, the present invention provides a method and apparatus for correcting an attribute value of a background attribute of an item, which can automatically modify the attribute value of the background attribute of the item, without requiring manual completion, and can improve the modification efficiency. .

In order to solve the above problems, the present invention discloses a method for correcting an attribute value of a background attribute of an item, the method comprising:

Obtaining an identifier of each of the N items; wherein the N is a natural number;

According to the identifier of each of the commodities, the N items are divided into M shareable background attribute product subsets; wherein, the M is a natural number, and the M is smaller than the N;

Counting the number of occurrences of each original attribute value of the same type of background attribute of all of the items included in each of the shareable background attribute item subsets;

Determining, according to the number of occurrences of each of the original attribute values of the same type of background attributes of all the commodities included in each of the shareable background attribute item subsets, determining each of the shareable background attribute item subsets included Corrected attribute values for the same type of background attributes of all of the items;

Each of the original attribute values of the same type of background attribute of all of the items included in each of the subsets of the shareable background attribute items is modified to the corrected attribute value.

Further, the identifier of each of the commodities includes:

a link address of a picture corresponding to each of the items, a content signature of a picture corresponding to each of the items, or a item number of each of the items.

Further, the picture corresponding to each of the commodities includes:

a main display image corresponding to each of the commodities, a supplementary display image corresponding to each of the commodities, a style color display image corresponding to each of the commodities, or a detail display image corresponding to each of the commodities.

Further, according to the identifier of each of the commodities, dividing the N items into M subsets of shareable background attribute products, including:

Constructing a binary group for each of the N of the commodities, wherein the first element of the binary group is an identifier of each of the commodities, and other elements of the binary group are An identity of each of the items, and a background attribute of each of the items, and an original attribute value of the background attribute;

Sorting all of the two groups according to the first element, and grouping the two groups of the same element to form a set of M groups, wherein each of the The set of tuples represents a subset of the shareable background attribute items.

Further, each of the shareable background attribute items is determined according to the number of occurrences of each of the original attribute values of the same type of background attribute of all the commodities included in each of the shareable background attribute item subsets. Corrected attribute values for the same type of background attribute of all of the items included in the set, including:

Taking the first subset of the sharable background attribute items in the M subset of the shareable background attribute items as the current subset of the shareable background attribute items;

And the number of occurrences of each of the original attribute values of the same type of background attribute of all the items included in the current subset of the shareable background attribute items, and all the items included in the current subset of the shareable background attribute items Performing a proportional calculation on the total number of occurrences of all the original attribute values of the same type of background attribute of the item, and obtaining each of the same type of background attributes of all the items included in the current subset of the shareable background attribute items a distribution ratio of the original attribute values;

And comparing the distribution ratio of each of the original attribute values of the same type of background attributes of all the commodities included in the current subset of the shareable background attribute items to a preset modification ratio threshold;

If there is a distribution ratio of the original attribute value in the background attribute of the same type of all the commodities included in the current share of the shareable background attribute item, which is greater than a preset modification ratio threshold, determine a certain The original attribute value is a corrected attribute value of a background attribute corresponding to a certain original attribute value of all the commodities included in the current subset of the shareable background attribute.

Further, counting the number of occurrences of each original attribute value of the same type of background attribute of all the commodities included in each of the shareable background attribute items, including:

Taking the first subset of the sharable background attribute items in the M subset of the shareable background attribute items as the current subset of the shareable background attribute items;

Determining whether the number of the products included in the subset of the tradable background attribute items is greater than a preset threshold of the modified quantity;

If it is greater than the preset modified quantity threshold, counting the number of occurrences of each of the original attribute values of the same type of background attribute of all the commodities included in the current share of the shareable background attribute item;

Correspondingly, each of the shareable background attribute items is determined according to the number of occurrences of each of the original attribute values of the same type of background attribute of all the commodities included in each of the shareable background attribute item subsets Corrected attribute values for the same type of background attribute of all of the items included in the set, including:

Determining, according to the number of occurrences of each of the original attribute values of the same type of background attribute of all the commodities included in the current share of the shareable background attribute item, all the places included in the current share of the shareable background attribute a corrected attribute value of a background attribute of the same type of the item;

Correspondingly, modifying each of the original attribute values of the same type of background attributes of all the items included in each of the subsets of the shareable background attribute items into the corrected attribute values comprises:

Each of the original attribute values of the same type of background attribute of all of the items included in the subset of the currently shareable background attribute items is modified to a corrected attribute value.

Further, determining whether the number of the products included in the current subset of the shareable background attribute items is large After the preset modified quantity threshold, it also includes:

If it is less than or equal to the preset modification quantity threshold, determining whether the current shareable background attribute product subset is the Mth of the sharable background attribute commodity subset;

If it is not the Mth shareable background attribute product subset, the next subset of the shareable background attribute item of the current shareable background attribute item subset is used as the current shareable background attribute item subset And performing the step of determining whether the number of the commodities included in the subset of the shareable background attribute items is greater than a preset modification quantity threshold;

If it is the Mth of the shareable background attribute product subset, it ends.

Further, after obtaining the identifier of each of the N items, the method further includes:

Mapping an identifier of each of the N of the commodities to an integer;

Taking an integer corresponding to each of the commodities to a preset number of parallel computing computer stations P; wherein the P is a natural number;

Assigning each of the commodities to a parallel computing computer of the number corresponding to the remainder;

Correspondingly, according to the identifier of each of the commodities, dividing the N items into M subsets of shareable background attribute products, including:

Determining, by each of the parallel computing computers, the N items into the M subsets of the shareable background attribute products according to the identifier of each of the commodities in each of the parallel computing computers;

Correspondingly, counting the number of occurrences of each original attribute value of the same type of background attribute of all the items included in each of the shareable background attribute items, including:

Counting, by each of the parallel computing computers, each of the original attribute values of the same type of background attribute of all of the items included in each of the subset of the shareable background attribute items in each of the parallel operation computers The number of occurrences;

Correspondingly, each of the shareable background attribute items is determined according to the number of occurrences of each of the original attribute values of the same type of background attribute of all the commodities included in each of the shareable background attribute item subsets Corrected attribute values for the same type of background attribute of all of the items included in the set, including:

And each of said original attribute values of the same type of background attribute of all said items included in each of said plurality of said shareable background attribute items in each of said parallel computing computers is passed by each said parallel computing computer a number of occurrences of determining a corrected attribute value of a background attribute of the same type of all of the items included in each of the subset of the shareable background attribute items in each of the parallel computing computers;

Correspondingly, modifying each of the original attribute values of the same type of background attributes of all the items included in each of the subsets of the shareable background attribute items into the corrected attribute values comprises:

Each of the parallelizable computing computers to each of the parallel computing computers Each of the original attribute values of the same type of background attribute of all of the items included in the attribute item subset is modified to a corrected attribute value.

In order to solve the above problem, the present invention also discloses an apparatus for correcting an attribute value of a background attribute of an item, the apparatus comprising:

An obtaining module, configured to obtain an identifier of each of the N items, wherein the N is a natural number;

a dividing module, configured to divide the N items into M sharable background attribute product subsets according to an identifier of each of the commodities; wherein, the M is a natural number, and the M is smaller than the N;

a statistics module, configured to count the number of occurrences of each original attribute value of the same type of background attribute of all the commodities included in each of the shareable background attribute commodity subsets;

a determining module, configured to determine each of the sharable background attributes according to an occurrence number of each of the original attribute values of a background attribute of a same type of all the commodities included in each of the shareable background attribute item subsets Corrected attribute values for background attributes of the same type for all of the items included in the subset of items;

And a modifying module, configured to modify each of the original attribute values of the same type of background attributes of all the items included in each of the subsets of the shareable background attribute items to the corrected attribute values.

Further, the identifier of each of the commodities includes:

a link address of a picture corresponding to each of the items, a content signature of a picture corresponding to each of the items, or a item number of each of the items.

Further, the picture corresponding to each of the commodities includes:

a main display image corresponding to each of the commodities, a supplementary display image corresponding to each of the commodities, a style color display image corresponding to each of the commodities, or a detail display image corresponding to each of the commodities.

Further, the dividing module includes:

a building unit, configured to construct a dual group for each of the N of the commodities, wherein a first element of the binary group is an identifier of each of the commodities, the binary The other elements of the group are the identity of each of the items, and the background attribute of each of the items, the original attribute value of the background attribute;

a sorting unit, configured to sort all the two groups according to the first element, and group the two groups of the same element to form a set of M groups, wherein Each of the set of binary groups represents a subset of the shareable background attribute items.

Further, the determining module includes:

a first processing unit, configured to use the first subset of the sharable background attribute items in the subset of the sharable background attribute items as a subset of the current shareable background attribute items;

a calculating unit, configured to display the number of occurrences of each of the original attribute values of the same type of background attribute of all the commodities included in the subset of the currently shareable background attribute items, and the current shareable background attribute item Performing a proportional calculation on the total number of occurrences of all the original attribute values of the same type of background attributes of all the commodities included in the set, and obtaining the same type of all the commodities included in the current subset of the shareable background attribute items The distribution ratio of each of the original attribute values of the background attribute;

a comparing unit, configured to perform a distribution ratio of each of the original attribute values of the same type of background attributes of all the products included in the current subset of the shareable background attribute items, and a preset modification ratio threshold Comparison

The attribute value determining unit is configured to: if a distribution ratio of the original attribute value exists in the background attribute of the same type of all the commodities included in the current subset of the shareable background attribute items, which is greater than a preset modification And a proportional threshold, determining that the original attribute value is a corrected attribute value of a background attribute corresponding to a certain original attribute value of all the commodities included in the current subset of the shareable background attribute commodity.

Further, the statistics module includes:

a second processing unit, configured to use the first subset of the sharable background attribute items in the M subset of the shareable background attribute items as the current subset of the shareable background attribute items;

a first determining unit, configured to determine whether the number of the products included in the subset of the tradable background attribute items is greater than a preset threshold of the modified quantity;

a statistical unit, configured to: if greater than the preset modified quantity threshold, count the number of occurrences of each of the original attribute values of the same type of background attribute of all the commodities included in the current share of the shareable background attribute commodity;

Correspondingly, the determining module comprises:

a current determining unit, configured to determine, according to the number of occurrences of each of the original attribute values of the same type of background attribute of all the commodities included in the current share of the shareable background attribute item, the current shareable background attribute item Corrected attribute values for the same type of background attributes of all of the items included in the subset;

Correspondingly, the modifying module comprises:

And a current modification unit, configured to modify each of the original attribute values of the same type of background attributes of all the items included in the current subset of the shareable background attribute items to a corrected attribute value.

Further, the statistics module further includes:

a second determining unit, configured to determine, if less than or equal to the preset modified quantity threshold, whether the current shareable background attribute product subset is the Mth shareable background attribute product subset;

a notification unit, configured to use, if not the Mth, the subset of the shareable background attribute items, the next subset of the shareable background attribute items of the current shareable background attribute item subset as the current shareable a background attribute item subset, but notifying the first determining unit to perform a step of determining whether the number of the items included in the current shareable background attribute item subset is greater than a preset modification quantity threshold;

The ending unit is configured to end if it is the Mth subset of the shareable background attribute items.

Further, the device further includes:

a mapping module, configured to map an identifier of each of the N items into an integer;

a remainder calculation module, configured to take an integer corresponding to each of the commodities to a preset number of parallel computing computer stations P; wherein, the P is a natural number;

An allocation module, configured to allocate each of the commodities to a parallel computing computer of a number corresponding to the remainder;

Correspondingly, the dividing module comprises: P dividing units, each of the dividing units being respectively disposed in each of the parallel computing computers;

And the P dividing units are configured to divide the N items into M subsets of the shareable background attribute products according to an identifier of each of the commodities in each of the parallel computing computers;

Correspondingly, the statistic module includes: P number of statistics units, each of the number of statistic units is respectively disposed in each of the parallel computing computers;

Each of the number of statistics units is configured to count each of the original attributes of the same type of background attributes of all of the items included in each of the subset of the shareable background attribute items in each of the parallel operation computers The number of occurrences of the attribute value;

Correspondingly, the determining module includes: P determining units, each of the determining units being respectively disposed in each of the parallel computing computers;

Each of the determining units is configured to each of the original attributes of a background attribute of the same type of all of the items included in each of the subset of the shareable background attribute items in each of the parallel computing computers a number of occurrences of the value, determining a corrected attribute value of a background attribute of the same type of all of the items included in each of the subset of the shareable background attribute items in each of the parallel computing computers;

Correspondingly, the modifying module includes: P modifying units, each of the modifying units being respectively disposed in each of the parallel computing computers;

Each of the modifying units is configured to respectively use each of the original attributes of the same type of background attribute of all of the items included in each of the subset of the shareable background attribute items in each of the parallel computing computers The value is modified to correct the attribute value.

Compared with the prior art, the present invention can obtain the following technical effects:

1) According to the identifier of each item, divide N items into M subsets of shareable background attribute products, according to each original of the same type of background attribute of all items included in each share of the shareable background attribute item The number of occurrences of the attribute value, the corrected attribute value of the same type of background attribute of all the items included in each of the shareable background attribute item subsets, and the background of the same type of all items included in each of the shareable background attribute item subsets Each original attribute value of the attribute is modified to correct the attribute value, and the attribute value of the background attribute of the item can be automatically performed. Modifications do not need to be done manually, which can improve the efficiency of modification.

2) Parallel modification by P parallel computing computer can greatly accelerate the operation and further improve the modification efficiency.

Of course, implementing any of the products of the present invention necessarily does not necessarily require all of the technical effects described above to be achieved at the same time.

DRAWINGS

The drawings described herein are intended to provide a further understanding of the invention, and are intended to be a part of the invention. In the drawing:

1 is a flow chart of a first method for correcting attribute values of background attributes of an item according to an embodiment of the present invention;

2 is a flow chart of a second method for correcting attribute values of background attributes of an item according to an embodiment of the present invention;

3 is a flowchart of a third method for correcting attribute values of background attributes of an item according to an embodiment of the present invention;

4 is a flowchart of a fourth method for correcting attribute values of background attributes of an item according to an embodiment of the present invention;

FIG. 5 is a schematic structural diagram of an apparatus for correcting attribute values of background attributes of an item according to an embodiment of the present invention; FIG.

FIG. 6 is a schematic structural diagram of an apparatus for correcting attribute values of background attributes of an item according to an embodiment of the present invention.

detailed description

The embodiments of the present invention will be described in detail below with reference to the accompanying drawings and embodiments, in which the present invention can be fully understood and implemented by the technical means of solving the technical problems and achieving the technical effects.

In a typical configuration, a computing device includes one or more processors (CPUs), input/output interfaces, network interfaces, and memory.

The memory may include non-persistent memory, random access memory (RAM), and/or non-volatile memory in a computer readable medium, such as read only memory (ROM) or flash memory. Memory is an example of a computer readable medium.

Computer readable media includes both permanent and non-persistent, removable and non-removable media. Information storage can be implemented by any method or technology. The information can be computer readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read only memory. (ROM), electrically erasable programmable read only memory (EEPROM), flash memory or other memory technology, compact disk read only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, Magnetic tape cartridges, magnetic tape storage or other magnetic storage devices or any other non-transportable media can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include non-transitory computer readable media, such as modulated data signals and carrier waves.

Certain terms are used throughout the description and claims to refer to particular components. Those skilled in the art will appreciate that hardware manufacturers may refer to the same component by different nouns. The present specification and the claims do not use the difference in the name as the means for distinguishing the components, but the difference in function of the components as the criterion for distinguishing. The word "comprising" as used throughout the specification and claims is an open term and should be interpreted as "including but not limited to". "Substantially" means that within the range of acceptable errors, those skilled in the art will be able to solve the technical problems within a certain error range, substantially achieving the technical effects. In addition, the term "coupled" is used herein to include any direct and indirect electrical coupling means. Therefore, if a first device is coupled to a second device, the first device can be directly electrically coupled to the second device, or electrically coupled indirectly through other devices or coupling means. Connected to the second device. The description of the present invention is intended to be illustrative of the preferred embodiments of the invention. The scope of the invention is defined by the appended claims.

It should also be noted that the terms "including", "comprising" or "comprising" or any other variations thereof are intended to encompass a non-exclusive inclusion, such that the item or system comprising a plurality of elements includes not only those elements but also Other elements, or elements that are inherent to such goods or systems. An element defined by the phrase "comprising a ..." does not exclude the presence of additional identical elements in the item or system including the element, without further limitation.

Description of the embodiment

The implementation of the method of the present invention is further illustrated by an embodiment below. As shown in FIG. 1 , a flowchart of a method for correcting attribute values of background attributes of an item according to an embodiment of the present invention includes:

S101: Acquire an identifier of each of the N items; wherein N is a natural number.

Specifically, the N items may be all the items in the one or more trading platforms, or may be all the items in the same category in the one or more trading platforms, and are not specifically limited. The method is applicable to any commodity.

The identifier of each product includes: a link address of a picture corresponding to each item, a content signature of a picture corresponding to each item (such as MD5 or other specially designed image signature, etc.), or a item number of each item. The picture corresponding to each item includes: the main display picture corresponding to each item (may be referred to as the main picture), the supplementary display picture corresponding to each item (possibly multiple), and the style color number display picture corresponding to each item (SKU Figure), or the details of each product display picture (details map) and so on.

It should be noted that if two products, for example, product A and product B, have some equality between the picture of item A and the picture of item B, for example, the picture of item A and the picture of item B originate from the same link address. (That is, product A and product B refer to the same picture), or if the picture of item A matches the content signature of the picture of item B (that is, the content of picture of item A and item B is the same), or such as A's item number and item B The item number is the same (that is, the item A and the item B have the same style), then the item A and the item B are likely to be the same item, that is, the background attributes of the item A and the item B should be consistent (shareable). .

According to the above description, it is obvious that the commodity A and the commodity A are themselves shareable with the background attribute relationship, so it can be said that the relationship is "reflexive"; if the commodity A is a shareable background attribute relationship for the commodity B, then Commodity B is also a shareable background attribute relationship for item A. It can be said that this relationship is "symmetric". If item A and item B refer to the same main display picture, and this main display picture is also commonly quoted by item B and item C as the main display picture, then item A and item C are also shared background attribute relationships, obviously this relationship It can be "delivered". According to the definition of discrete mathematics, the relationship that satisfies these three conditions—the sharable back-end attribute relationship between commodities is an “equivalent relationship”.

It should be noted that, in general, the main display image of the product must comply with strict specifications, and it must be fully displayed and cannot contain irrelevant information (that is, the main display image of the product is highly relevant to the product) The reliability of the shareable background attribute relationship between the products established by the repeated reference relationship of the main display image of the product is higher. Therefore, it is preferable to sign the link address of the main display image corresponding to the product or the content signature of the main display image corresponding to the product. As the identifier of the product.

S102: Divide the N items into M subsets of the shareable background attribute according to the identifier of each item; wherein M is a natural number and M is less than N.

Specifically, according to the identifier of each item, the N items are divided into M subsets of the shareable background attribute items, including:

Constructing a two-tuple for each of the N items, wherein the first element of the binary is the identifier of each item, the other elements of the group are the identity of each item, and each The background attribute of the item and the original attribute value of the background attribute.

For example, for any of the N items K, a binary group is constructed for it, denoted as PairK:keyK–nidK, <pid0, vidK, 0>, <pid1, vidK, 1>. In the binary group PairK, keyK is the identifier of the commodity K (for example, the link address of the main display image corresponding to the commodity K), and is the first element of the binary group; the other elements of the binary group are composed of the following members. : nidK (for the identity ID of the item K), <pid0, vidK, 0>, <pid1, vidK, 1>... (for the background attribute/attribute value pair of the item K (ie the background attribute of the item K, the item K of the item) The attribute value of the background attribute). <pid0, vidK, 0>, <pid1, vidK, 1>... for example, <style-shell>, <cloth length-short paragraph>, <shoe head-fish mouth> and so on.

Sort all the two groups according to the first element, and group the same two groups of the first element to form a set of M groups, where each set of two sets represents a shareable background attribute A subset of goods.

S103: Count the number of occurrences of each original attribute value of the same type of background attribute of all items included in each of the items in the subset of the shareable background attribute.

Specifically, because the items in the set of the two sets have the same identifier (for example, having the same main display picture), Therefore, the items in the set of two sets are likely to represent the same item, and the attribute values of the background attributes of the items in the set of sets should also be consistent. However, in a subset of shareable background property items, since these items come from different sellers, although their background attributes should be unified, the actual situation is often different. Therefore, it is necessary to count the number of occurrences of each original attribute value of the same type of background attribute of all the items included in each subset of the shareable background attribute items, and determine all the items included in each of the shareable background attribute item subsets according to the statistical result. The correct attribute value for the same type of background property.

S104: Determine, according to the number of occurrences of each original attribute value of the same type of background attribute of all items included in each of the shareable background attribute item subsets, determine the same type of all items included in each of the shareable background attribute item subsets The corrected attribute value of the background property.

Specifically, the original attribute value that has the most occurrences may be determined according to the number of occurrences, and the corrected attribute value of the same type of background attribute of all the items included in each of the items in the subset of the shareable background attribute.

However, although the original attribute value has the most occurrences, because the other sellers do not fill in the attribute value of the background attribute of the type, or the original attribute value of the most frequently occurring is the seller's error-filling, etc., the above various In this case, the value of the original attribute that has the most occurrences is not necessarily the value of the correction attribute. For stability considerations, it can be determined that when the distribution ratio of a certain original attribute value reaches a certain modification ratio threshold, it is determined that the original attribute value is the same type of background attribute of all the items included in each of the shareable background attribute commodity subsets. Correct the attribute value. If the number of occurrences of an original attribute value is the most, but the distribution ratio does not reach a certain revision ratio threshold, then the attribute of the background attribute of the type is considered to be complicated and no modification is made.

Specifically, for stability consideration, referring to FIG. 2, each shareable background attribute item is determined according to the number of occurrences of each original attribute value of the same type of background attribute of all items included in each of the shareable background attribute item subsets. Corrected attribute values for the same type of background attribute for all items included in the subset, including:

S104a: The first shareable background attribute item subset of the M shareable background attribute items subset is used as the current shareable background attribute item subset.

S104b: the number of occurrences of each original attribute value of the same type of background attribute of all items included in the current shareable background attribute item subset, and the same type of background attribute of all items included in the current shareable background attribute item subset The total number of occurrences of all the original attribute values is proportionally calculated, and the distribution ratio of each original attribute value of the same type of background attribute of all the items included in the current shareable background attribute item subset is obtained.

S104c: Compare the distribution ratio of each original attribute value of the same type of background attribute of all the items included in the current share of the currently shareable background attribute item, and the preset modification ratio threshold, if the current shareable background is obtained If there is a distribution ratio of a certain original attribute value in the background attribute of the same type of all items included in the attribute item subset, which is greater than the preset modification ratio threshold, S104d is performed; if the obtained current shareable background attribute item subset is included The distribution of some original attribute value does not exist in the background attribute of the same type for all items If the ratio is greater than the preset modification ratio threshold, then S104f is performed.

Specifically, the preset modification ratio threshold may be set according to actual application conditions, for example, a monitoring penalty product having higher reliability requirements, and the preset modification ratio threshold may be set to 75% or higher to reduce the inspection. The rate of failure; for products with high coverage/correction coverage requirements, the requirements can be relaxed appropriately, and the preset modification ratio threshold can be set to 30%.

S104d: Determine a certain attribute value as a correction attribute value of a background attribute corresponding to a certain original attribute value of all items included in the current shareable background attribute item subset.

It should be noted that when there are more than two original attribute values that are greater than the preset modification ratio threshold, the original attribute value with the largest distribution ratio may be selected as the correction attribute value.

S104e: Modify each original attribute value of the background attribute corresponding to a certain original attribute value of all items included in the current shareable background attribute item subset to a corrected attribute value, and then execute S104g.

S104f: Determine not to modify each original attribute value of the background attribute corresponding to a certain original attribute value of all the commodities included in the current shareable background attribute item subset, and then execute S104g.

S104g: Determine whether the currently shareable background attribute product subset is the Mth shareable background attribute product subset, if not, execute S104h; otherwise, end.

S104h: The next shareable background attribute product subset of the current shareable background attribute item subset is taken as the current shareable background attribute item subset, and then S104b is performed.

S105: Modify each original attribute value of the same type of background attribute of all items included in each subset of the shareable background attribute items to a corrected attribute value.

Specifically, each original attribute value of the same type of background attribute of all items included in each subset of the shareable background attribute items is modified to a corrected attribute value, specifically: if each of the shareable background attribute item subsets is included The original attribute value of the same type of background attribute of all items is the same as the corrected attribute value, and is retained; if the original attribute value and correction attribute of the same type of background attribute of all items included in each of the shareable background attribute items are included If the values are not the same, the correction is made; if the original attribute value of the same type of background attribute of all the items included in each of the shareable background attribute items is missing, it is added.

Preferably, referring to FIG. 3, in a preferred embodiment, S103-S105 may include:

S201: The first shareable background attribute product subset of the M shareable background attribute item subsets is used as the current shareable background attribute product subset.

S202: Determine whether the number of items included in the current shareable background attribute item set is greater than a preset modification quantity threshold, if it is greater than a preset modification quantity threshold, execute S203; if it is less than or equal to a preset modification quantity threshold, Execute S206.

Specifically, statistics are performed when a subset of the shareable background attribute items contains a certain number of items, for example: If there is only one item, there is no basis for modification and it is necessary.

The preset modification quantity threshold may be set according to the actual application status, for example, it may be set to 2, 20, and the like.

S203: Count the number of occurrences of each original attribute value of the same type of background attribute of all items included in the current attribute group of the background attribute.

S204: Determine, according to the number of occurrences of each original attribute value of the same type of background attribute of all items included in the current shareable background attribute item subset, determine a background attribute of the same type of all items included in the current shareable background attribute item subset. Corrected attribute value.

S205: Modify each original attribute value of the same type of background attribute of all items included in the current shareable background attribute item subset to a corrected attribute value.

S206: Determine whether the current share of the background attribute item is a subset of the M shareable background attribute product, if not the Mth shareable background attribute product subset, execute S207; if it is the Mth shareable background attribute The subset of goods ends.

S207: The next shareable background attribute item subset of the current shareable background attribute item subset is taken as the current shareable background attribute item subset, but S202 is performed.

For platform-level products, the amount of merchandise that needs to be processed may be in the order of more than 100 million. If it is only implemented on a single computer, the time consumption is still unacceptable. However, the method of the present embodiment can be highly parallel, so the processing speed is also fast for a large number of commodities. Suppose that there are N items, which are processed by running the computer in parallel through the P station. Specifically, referring to FIG. 4, in a preferred embodiment, S102-S105 may include:

S301: Map an identifier of each of the N items to an integer.

Specifically, the identifier key of each commodity can be regarded as a character string, and the key can be mapped to an integer by hash code. Assuming that key=c0c1...cL consists of L characters, then the corresponding integer (denoted as HCL) is calculated as:

HC0=0;

HCL=HCL-1*Z+cL.

Where Z is any prime number, usually set to 31; character c is represented by its ASCII code (integer).

It is not limited to mapping the identifier of the commodity to an integer by the above method, and may be implemented in any feasible manner, which is not specifically limited.

S302: Take the integer of each commodity to the preset parallel computing computer number P to take a remainder; wherein P is a natural number.

S303: Assign each item to the parallel computing computer of the number corresponding to the remainder.

Specifically, the numbers of the P parallel computing computers are 0-P-1, respectively.

For each item, the HC corresponding to its key is allocated to the remainder of P. In this way, all the items to be processed are distributed substantially evenly to the P parallel computing computer. It is equivalent to pre-cutting the remainder of the task by the remainder.

S304: N pieces of goods are collectively divided into M sharable background attribute commodity subsets by each parallel computing computer according to the identifier of each commodity in each parallel computing computer.

Specifically, N products are allocated to the P parallel computing computer according to the identifier of the product (the number of products in each parallel computing computer is less than N, and the sum of the number of products in the P parallel computing computer is N), for each The parallel parallel computing computer has the same method as a computer partitioning the subset of the background attribute goods, as follows: Each parallel computing computer constructs a binary group for each commodity in each parallel computing computer, wherein, binary The first element of the group is the identifier of each item, the other elements of the group are the identity of each item, and the background attribute of each item, the original attribute value of the background attribute; all the groups are The first element is sorted, and the same binary group of the first element is grouped together to form a plurality of binary group sets (the number of binary group sets obtained by each parallel computing computer is less than M, P station parallel operation The sum of the number of sets of binary groups obtained by the computer is M), wherein each set of two sets represents a subset of the shareable background attribute goods.

It should be noted that since the identifiers of the same product are the same and the HCs are the same, the same product will be assigned to the same shareable back-end product regardless of whether it is sorted in the whole or in the partial sort after the split. A subset of attributes. Therefore, the subset of shareable background product attributes obtained by task segmentation is identical in quantity and content to the subset of shareable background product attributes obtained after overall sorting on a single computer. That is to say, for each parallel computing computer, the process is the same as when using a computer. Task segmentation does not affect the correctness of the entire process.

S305: Count, by each parallel computing computer, the number of occurrences of each original attribute value of a background attribute of the same type of all items included in each of the items in the subset of shareable background attribute items in each parallel operation computer.

S306: Determine, by each parallel computing computer, the number of occurrences of each original attribute value of the same type of background attribute of all commodities included in each subset of the shareable background attribute items in each parallel operation computer, and determine each parallel operation Each of the computers can share the corrected attribute value of the same type of background attribute for all items included in the background attribute item subset.

S307: Modify, by each parallel computing computer, each original attribute value of a background attribute of the same type of all commodities included in each subset of the shareable background attribute items in each parallel operation computer as a correction attribute value.

Specifically, the parallel strategy can greatly accelerate the operation for the following reasons: 1) Convert the ensemble sorting into partial sorting on the P parallel computing computer, which not only utilizes the parallel computing power of the P parallel computing computer, but also complicates the calculation. The degree is greatly reduced. Assuming that there are N commodities in the complete set, the complexity of the optimal sorting algorithm for the corpus sorting is N·logN; and if it is converted to partial sorting of P parallel computing computers, it is assumed that the uniform segmentation is The complexity is P·(N/P)·log(N/P)=N·log(N/P)=N·(logN-logP), so the computational complexity of N·logP is saved compared to the full set ordering. 2) Parallel modification of N commodities by P parallel computing computer. Although each parallel computing computer does not reduce the computational complexity from the whole process, the P parallel computing computer modifies N commodities in parallel, and the total of N commodities. The modification time will be reduced to 1/P.

The method for correcting the attribute value of the background attribute of the product according to the embodiment, according to the identifier of each item, dividing the N items into M subsets of the shareable background attribute, according to each shareable background attribute item The number of occurrences of each original attribute value of the same type of background attribute of all items included in the set, and the corrected attribute value of the same type of background attribute of all items included in each subset of the shareable background attribute items, each of which can be Each original attribute value of the same type of background attribute of all items included in the shared background attribute commodity group is modified to the corrected attribute value, and the attribute value of the background attribute of the item can be automatically modified, without manual completion, and the modification can be improved. effectiveness. Parallel modification by P parallel computing computer can greatly accelerate the operation and further improve the modification efficiency.

As shown in FIG. 5, it is an apparatus structure diagram for correcting attribute values of background attributes of an item according to an embodiment of the present invention, and the apparatus includes:

The obtaining module 401 is configured to obtain an identifier of each of the N items, where N is a natural number;

The dividing module 402 is configured to divide the N items into M subsets of the sharable background attribute according to the identifier of each item; wherein, M is a natural number, and M is less than N;

a statistics module 403, configured to count the number of occurrences of each original attribute value of a background attribute of the same type of all items included in each subset of the shareable background attribute items;

a determining module 404, configured to determine, according to the number of occurrences of each of the original attribute values of the same type of background attribute of all items included in each of the shareable background attribute item subsets, each of the shareable background attribute item subsets is included Corrected attribute values for background attributes of the same type for all items;

The modifying module 405 is configured to modify each of the original attribute values of the same type of background attributes of all items included in each subset of the shareable background attribute items to a corrected attribute value.

Further, the identifier of each item includes:

The link address of the picture corresponding to each item, the content signature of the picture corresponding to each item, or the item number of each item.

Further, the pictures corresponding to each item include:

The main display image corresponding to each item, the supplementary display picture corresponding to each item, the style color number display picture corresponding to each item, or the detail display picture corresponding to each item.

Further, the dividing module 402 includes:

a building unit for constructing a binary group for each of the N items, wherein the first of the two groups The element is an identifier of each item, the other elements of the group are the identity of each item, and the background attribute of each item, the original attribute value of the background attribute;

a sorting unit for sorting all the two groups according to the first element, and grouping the same two groups of the first element to form a set of M sets, wherein each set of the two sets represents A subset of items that can share background attributes.

Further, the determining module includes:

a first processing unit, configured to use the first shareable background attribute product subset of the M shareable background attribute item subsets as a current shareable background attribute product subset;

a calculating unit, configured to use the same number of occurrences of each original attribute value of the same type of background attribute of all items included in the current shareable background attribute item subset, and the same type of all items included in the current shareable background attribute item subset The total number of occurrences of all the original attribute values of the background attribute is proportionally calculated, and the distribution ratio of each original attribute value of the same type of background attribute of all the items included in the current shareable background attribute item subset is obtained;

a comparison unit, configured to compare a distribution ratio of each original attribute value of a background attribute of the same type of all items included in the obtained current shareable background attribute item subset with a preset modification ratio threshold;

The attribute value determining unit is configured to determine, if the distribution ratio of the original attribute value exists in the background attribute of the same type of all the items included in the current shareable background attribute item set, which is greater than the preset modification ratio threshold, determine a certain The original attribute value is the corrected attribute value of the background attribute corresponding to a certain original attribute value of all the items included in the current shareable background attribute item subset.

Further, the statistics module 403 includes:

a second processing unit, configured to use the first shareable background attribute product subset of the M shareable background attribute item subsets as a current shareable background attribute product subset;

a first determining unit, configured to determine whether the number of items included in the current shareable background attribute item subset is greater than a preset modification quantity threshold;

a statistical unit, configured to count, if greater than the preset modified quantity threshold, the number of occurrences of each original attribute value of the same type of background attribute of all commodities included in the current shareable background attribute commodity subset;

Accordingly, the determining module 404 includes:

a current determining unit, configured to determine, according to the number of occurrences of each original attribute value of the same type of background attribute of all items included in the current shareable background attribute item subset, the same item of all items included in the current shareable background attribute item subset The corrected attribute value of the background property of the type;

Accordingly, the modification module 405 includes:

Current modification unit for the same type of all items included in the current shareable background attribute item subset Each original attribute value of the background attribute is modified to correct the attribute value.

Further, the statistics module 403 further includes:

a second determining unit, configured to determine, if less than or equal to the preset modified quantity threshold, whether the current shareable background attribute product subset is the Mth shareable background attribute commodity subset;

a notification unit, configured to: if not the Mth shareable background attribute product subset, the next shareable background attribute item subset of the current shareable background attribute item subset is used as the current shareable background attribute item subset, but the notification The first determining unit performs a step of determining whether the number of items included in the current shareable background attribute item subset is greater than a preset modification quantity threshold;

The ending unit is used to end if it is the Mth shareable background attribute product subset.

Further, referring to FIG. 6, the apparatus further includes:

a mapping module 406, configured to map an identifier of each of the N items into an integer;

a remainder calculation module 407, configured to take an integer corresponding to each commodity to a preset number of parallel computing computer stations P; wherein, P is a natural number;

An allocation module 408, configured to allocate each item to the parallel computing computer of the number corresponding to the remainder;

Correspondingly, the dividing module 402 includes: P dividing units 402a, each of which is respectively disposed in each parallel computing computer;

The P dividing units 402a are configured to divide the N items into M sub-shared background attribute commodity subsets according to the identifier of each commodity in each parallel computing computer;

Correspondingly, the statistics module 403 includes: P number of statistics units 403a, each of which is set in each parallel computing computer;

Each number of times statistics unit 403a is configured to count the number of occurrences of each original attribute value of the same type of background attribute of all items included in each of the shareable background attribute commodity subsets in each parallel operation computer;

Correspondingly, the determining module 404 includes: P determining units 404a, each determining unit is respectively disposed in each parallel computing computer;

Each determining unit is configured to determine each parallel operation according to the number of occurrences of each original attribute value of the same type of background attribute of all the items included in each of the shareable background attribute items in each parallel computing computer Each of the computers can share a corrected attribute value of the same type of background attribute of all items included in the background attribute item subset;

Correspondingly, the modifying module 405 includes: P modifying units 405a, each of which is separately disposed in each parallel computing computer;

Each modification unit 405a is respectively configured to modify each original attribute value of the same type of background attribute of all items included in each of the shareable background attribute item subsets in each parallel operation computer to a correction attribute value.

The apparatus for correcting the attribute value of the background attribute of the product according to the embodiment, according to the identifier of each item, dividing the N items into M subsets of the shareable background attribute, according to each shareable background attribute item The number of occurrences of each original attribute value of the same type of background attribute of all items included in the set, and the corrected attribute value of the same type of background attribute of all items included in each subset of the shareable background attribute items, each of which can be Each original attribute value of the same type of background attribute of all items included in the shared background attribute commodity group is modified to the corrected attribute value, and the attribute value of the background attribute of the item can be automatically modified, without manual completion, and the modification can be improved. effectiveness. Parallel modification by P parallel computing computer can greatly accelerate the operation and further improve the modification efficiency.

The device corresponds to the foregoing method flow description, and the deficiencies refer to the description of the above method flow, and will not be further described.

The above description illustrates and describes several preferred embodiments of the present invention, but as described above, it should be understood that the invention is not limited to the forms disclosed herein, and should not be construed as Other combinations, modifications, and environments are possible and can be modified by the above teachings or related art or knowledge within the scope of the inventive concept described herein. All changes and modifications made by those skilled in the art are intended to be within the scope of the appended claims.

Claims (16)

1. A method for correcting attribute values of a background attribute of an item, the method comprising:

   Obtaining an identifier of each of the N items; wherein the N is a natural number;

   According to the identifier of each of the commodities, the N items are divided into M shareable background attribute product subsets; wherein, the M is a natural number, and the M is smaller than the N;

   Counting the number of occurrences of each original attribute value of the same type of background attribute of all of the items included in each of the shareable background attribute item subsets;

   Determining, according to the number of occurrences of each of the original attribute values of the same type of background attributes of all the commodities included in each of the shareable background attribute item subsets, determining each of the shareable background attribute item subsets included Corrected attribute values for the same type of background attributes of all of the items;

   Each of the original attribute values of the same type of background attribute of all of the items included in each of the subsets of the shareable background attribute items is modified to the corrected attribute value.
2. The method of claim 1 wherein the identifier of each of said merchandise comprises:

   a link address of a picture corresponding to each of the items, a content signature of a picture corresponding to each of the items, or a item number of each of the items.
3. The method of claim 2 wherein the picture corresponding to each of said items comprises:

   a main display image corresponding to each of the commodities, a supplementary display image corresponding to each of the commodities, a style color display image corresponding to each of the commodities, or a detail display image corresponding to each of the commodities.
4. The method of claim 1, wherein dividing the N items into M subsets of shareable background attributes according to the identifier of each of the items comprises:

   Constructing a binary group for each of the N of the commodities, wherein the first element of the binary group is an identifier of each of the commodities, and other elements of the binary group are An identity of each of the items, and a background attribute of each of the items, and an original attribute value of the background attribute;

   Sorting all of the two groups according to the first element, and grouping the two groups of the same element to form a set of M groups, wherein each of the The set of tuples represents a subset of the shareable background attribute items.
5. The method of claim 1 wherein determining the number of occurrences of each of said original attribute values for a background attribute of the same type of all of said items of said plurality of said items in said subset of shareable background attributes Corrected attribute values of the same type of background attributes of all of the items included in each of the shareable background attribute merchandise subsets, including:

   Taking the first subset of the sharable background attribute items in the M subset of the shareable background attribute items as the current subset of the shareable background attribute items;

   And the number of occurrences of each of the original attribute values of the same type of background attribute of all the items included in the current subset of the shareable background attribute items, and all the items included in the current subset of the shareable background attribute items Performing a proportional calculation on the total number of occurrences of all the original attribute values of the same type of background attribute of the item, and obtaining each of the same type of background attributes of all the items included in the current subset of the shareable background attribute items a distribution ratio of the original attribute values;

   And comparing the distribution ratio of each of the original attribute values of the same type of background attributes of all the commodities included in the current subset of the shareable background attribute items to a preset modification ratio threshold;

   If there is a distribution ratio of the original attribute value in the background attribute of the same type of all the commodities included in the current share of the shareable background attribute item, which is greater than a preset modification ratio threshold, determine a certain The original attribute value is a corrected attribute value of a background attribute corresponding to a certain original attribute value of all the commodities included in the current subset of the shareable background attribute.
6. The method of claim 1, wherein counting the number of occurrences of each of the original attribute values of the same type of background attribute of all of the items included in each of the shareable background attribute items includes:

   Taking the first subset of the sharable background attribute items in the M subset of the shareable background attribute items as the current subset of the shareable background attribute items;

   Determining whether the number of the products included in the subset of the tradable background attribute items is greater than a preset threshold of the modified quantity;

   If it is greater than the preset modified quantity threshold, counting the number of occurrences of each of the original attribute values of the same type of background attribute of all the commodities included in the current share of the shareable background attribute item;

   Correspondingly, each of the shareable background attribute items is determined according to the number of occurrences of each of the original attribute values of the same type of background attribute of all the commodities included in each of the shareable background attribute item subsets Corrected attribute values for the same type of background attribute of all of the items included in the set, including:

   Determining, according to the number of occurrences of each of the original attribute values of the same type of background attribute of all the commodities included in the current share of the shareable background attribute item, all the places included in the current share of the shareable background attribute a corrected attribute value of a background attribute of the same type of the item;

   Correspondingly, modifying each of the original attribute values of the same type of background attributes of all the items included in each of the subsets of the shareable background attribute items into the corrected attribute values comprises:

   Each of the original attribute values of the same type of background attribute of all of the items included in the subset of the currently shareable background attribute items is modified to a corrected attribute value.
7. The method of claim 6, wherein the determining whether the number of the products included in the subset of the shareable background attribute items is greater than a preset number of modification thresholds further includes:

   If the preset modification quantity threshold is less than or equal to, the judging whether the current shareable background attribute product subset is Is the Mth subset of the shareable background attribute products;

   If it is not the Mth shareable background attribute product subset, the next subset of the shareable background attribute item of the current shareable background attribute item subset is used as the current shareable background attribute item subset And performing the step of determining whether the number of the commodities included in the subset of the shareable background attribute items is greater than a preset modification quantity threshold;

   If it is the Mth of the shareable background attribute product subset, it ends.
8. The method according to any one of claims 1 to 7, wherein after obtaining the identifier of each of the N items, the method further comprises:

   Mapping an identifier of each of the N of the commodities to an integer;

   Taking an integer corresponding to each of the commodities to a preset number of parallel computing computer stations P; wherein the P is a natural number;

   Assigning each of the commodities to a parallel computing computer of the number corresponding to the remainder;

   Correspondingly, according to the identifier of each of the commodities, dividing the N items into M subsets of shareable background attribute products, including:

   Determining, by each of the parallel computing computers, the N items into the M subsets of the shareable background attribute products according to the identifier of each of the commodities in each of the parallel computing computers;

   Correspondingly, counting the number of occurrences of each original attribute value of the same type of background attribute of all the items included in each of the shareable background attribute items, including:

   Counting, by each of the parallel computing computers, each of the original attribute values of the same type of background attribute of all of the items included in each of the subset of the shareable background attribute items in each of the parallel operation computers The number of occurrences;

   Correspondingly, each of the shareable background attribute items is determined according to the number of occurrences of each of the original attribute values of the same type of background attribute of all the commodities included in each of the shareable background attribute item subsets Corrected attribute values for the same type of background attribute of all of the items included in the set, including:

   And each of said original attribute values of the same type of background attribute of all said items included in each of said plurality of said shareable background attribute items in each of said parallel computing computers is passed by each said parallel computing computer a number of occurrences of determining a corrected attribute value of a background attribute of the same type of all of the items included in each of the subset of the shareable background attribute items in each of the parallel computing computers;

   Correspondingly, modifying each of the original attribute values of the same type of background attributes of all the items included in each of the subsets of the shareable background attribute items into the corrected attribute values comprises:

   Each of the original attribute values of the same type of background attribute of all of the items included in each of the subset of the shareable background attribute items in each of the parallel computing computers by each of the parallel computing computers Modified to correct the attribute value.
9. An apparatus for correcting an attribute value of a background attribute of a commodity, wherein the apparatus comprises:

   An obtaining module, configured to obtain an identifier of each of the N items, wherein the N is a natural number;

   a dividing module, configured to divide the N items into M sharable background attribute product subsets according to an identifier of each of the commodities; wherein, the M is a natural number, and the M is smaller than the N;

   a statistics module, configured to count the number of occurrences of each original attribute value of the same type of background attribute of all the commodities included in each of the shareable background attribute commodity subsets;

   a determining module, configured to determine each of the sharable background attributes according to an occurrence number of each of the original attribute values of a background attribute of a same type of all the commodities included in each of the shareable background attribute item subsets Corrected attribute values for background attributes of the same type for all of the items included in the subset of items;

   And a modifying module, configured to modify each of the original attribute values of the same type of background attributes of all the items included in each of the subsets of the shareable background attribute items to the corrected attribute values.
10. The apparatus of claim 9 wherein the identifier of each of said merchandise comprises:

    a link address of a picture corresponding to each of the items, a content signature of a picture corresponding to each of the items, or a item number of each of the items.
11. The device according to claim 10, wherein each of the pictures corresponding to the item comprises:

    a main display image corresponding to each of the commodities, a supplementary display image corresponding to each of the commodities, a style color display image corresponding to each of the commodities, or a detail display image corresponding to each of the commodities.
12. The device of claim 9, wherein the dividing module comprises:

    a building unit, configured to construct a dual group for each of the N of the commodities, wherein a first element of the binary group is an identifier of each of the commodities, the binary The other elements of the group are the identity of each of the items, and the background attribute of each of the items, the original attribute value of the background attribute;

    a sorting unit, configured to sort all the two groups according to the first element, and group the two groups of the same element to form a set of M groups, wherein Each of the set of binary groups represents a subset of the shareable background attribute items.
13. The apparatus of claim 9, wherein the determining module comprises:

    a first processing unit, configured to use the first subset of the sharable background attribute items in the subset of the sharable background attribute items as a subset of the current shareable background attribute items;

    a calculating unit, configured to display the number of occurrences of each of the original attribute values of the same type of background attribute of all the commodities included in the subset of the currently shareable background attribute items, and the current shareable background attribute item Performing a proportional calculation on the total number of occurrences of all the original attribute values of the same type of background attribute of all the commodities included in the set, and obtaining the same of all the commodities included in the current subset of the shareable background attribute The distribution ratio of each of the original attribute values of the background attribute of the type;

    a comparing unit, configured to perform a distribution ratio of each of the original attribute values of the same type of background attributes of all the products included in the current subset of the shareable background attribute items, and a preset modification ratio threshold Comparison

    The attribute value determining unit is configured to: if a distribution ratio of the original attribute value exists in the background attribute of the same type of all the commodities included in the current subset of the shareable background attribute items, which is greater than a preset modification And a proportional threshold, determining that the original attribute value is a corrected attribute value of a background attribute corresponding to a certain original attribute value of all the commodities included in the current subset of the shareable background attribute commodity.
14. The device of claim 9, wherein the statistical module comprises:

    a second processing unit, configured to use the first subset of the sharable background attribute items in the M subset of the shareable background attribute items as the current subset of the shareable background attribute items;

    a first determining unit, configured to determine whether the number of the products included in the subset of the tradable background attribute items is greater than a preset threshold of the modified quantity;

    a statistical unit, configured to: if greater than the preset modified quantity threshold, count the number of occurrences of each of the original attribute values of the same type of background attribute of all the commodities included in the current share of the shareable background attribute commodity;

    Correspondingly, the determining module comprises:

    a current determining unit, configured to determine, according to the number of occurrences of each of the original attribute values of the same type of background attribute of all the commodities included in the current share of the shareable background attribute item, the current shareable background attribute item Corrected attribute values for the same type of background attributes of all of the items included in the subset;

    Correspondingly, the modifying module comprises:

    And a current modification unit, configured to modify each of the original attribute values of the same type of background attributes of all the items included in the current subset of the shareable background attribute items to a corrected attribute value.
15. The device of claim 14, wherein the statistics module further comprises:

    a second determining unit, configured to determine, if less than or equal to the preset modified quantity threshold, whether the current shareable background attribute product subset is the Mth shareable background attribute product subset;

    a notification unit, configured to use, if not the Mth, the subset of the shareable background attribute items, the next subset of the shareable background attribute items of the current shareable background attribute item subset as the current shareable a background attribute item subset, but notifying the first determining unit to perform a step of determining whether the number of the items included in the current shareable background attribute item subset is greater than a preset modification quantity threshold;

    The ending unit is configured to end if it is the Mth subset of the shareable background attribute items.
16. The device according to any one of claims 9-15, wherein the device further comprises:

    a mapping module, configured to map an identifier of each of the N items into an integer;

    a remainder calculation module, configured to take an integer corresponding to each of the commodities to a preset number of parallel computing computer stations P; wherein, the P is a natural number;

    An allocation module, configured to allocate each of the commodities to a parallel computing computer of a number corresponding to the remainder;

    Correspondingly, the dividing module comprises: P dividing units, each of the dividing units being respectively disposed in each of the parallel computing computers;

    And the P dividing units are configured to divide the N items into M subsets of the shareable background attribute products according to an identifier of each of the commodities in each of the parallel computing computers;

    Correspondingly, the statistic module includes: P number of statistics units, each of the number of statistic units is respectively disposed in each of the parallel computing computers;

    Each of the number of statistics units is configured to count each of the original attributes of the same type of background attributes of all of the items included in each of the subset of the shareable background attribute items in each of the parallel operation computers The number of occurrences of the attribute value;

    Correspondingly, the determining module includes: P determining units, each of the determining units being respectively disposed in each of the parallel computing computers;

    Each of the determining units is configured to each of the original attributes of a background attribute of the same type of all of the items included in each of the subset of the shareable background attribute items in each of the parallel computing computers a number of occurrences of the value, determining a corrected attribute value of a background attribute of the same type of all of the items included in each of the subset of the shareable background attribute items in each of the parallel computing computers;

    Correspondingly, the modifying module includes: P modifying units, each of the modifying units being respectively disposed in each of the parallel computing computers;

    Each of the modifying units is configured to respectively use each of the original attributes of the same type of background attribute of all of the items included in each of the subset of the shareable background attribute items in each of the parallel computing computers The value is modified to correct the attribute value.

Images