« PreviousContinue »
United States Patent im
Maeda et al.
US005802509A Patent Number: Date of Patent:
 RULE GENERATION SYSTEM AND METHOD OF GENERATING RULE
 Inventors: Akira Maeda; Hitoshi Ashida, both of Yokohama; Toshihide Ichimori. Kawasaki; Chiaki Hirai, Tokyo; Yori Takahashi. Yokohama, all of Japan
 Assignee: Hitachi, Ltd., Tokyo, Japan
 Appl. No.: 893,422
 Filed: Jul. 11,1997
Related U.S. Application Data
 Continuation of Set No. 306,111, Sep. 14,1994, abandoned.  Foreign Application Priority Data
Sep. 21, 1993 [JP] Japan 5-257778
Dec. 24,1993 [JP] Japan 5-327352
 Int CL6 GOfiF 15/18
 U.S. Q. 706/59
 Field of Search 395/75, 51, 72,
395/68, 22; 364/157, 474.23
 References Cited
U.S. PATENT DOCUMENTS 5,179,634 1/1993 Matstmaga et al 395/75
5222,197 6/1993 Tengetal 395/68
5,229,950 7/1993 Nfiwa 364/47453
5,241,620 8/1993 Rnggiero 395/22
5,295,061 3/1994 Katayamaetal 364/157
5,422,984 6/1995 Iokibeetal 395/51
5,440,672 8/1995 Araki et al 395/51
5,504,840 4/1996 Hiji et al 395/77
Kim et al, Automatic generation of membership function and fuzzy rule using inductive reasoning, JFIS 93, pp. 93-96, Dec. 3, 1993.
Silva et al., Induction of fuzzy production rules, Proceedings of the twentieth International symposium on multipe-valued logic, pp. 270-278, May 25, 1990.
Higgins et al., Learning fuzzy rule based neural network for function approximation, DCNN, pp. 251-256, Jun. 11,1992. Higgins et al., Fuzzy rule -based networks for control, IEEE transactions on fuzzy systems, pp. 82-88, Feb. 1994. Sebag et al., Learning membership functions from examples, IEEE computer society press, pp. 169-173, Apr. 28, 1993.
Krishnapuram et al., Compact fuzzy rule base generation methods for computer vision, Second IEEE International conference on fuzzy systems, pp. 809—814, vol. 2, Apr. 1, 1993.
Quinlan, J. Ross. "Learning Efficient Classification Procedures and Their Application to Chess End Games," Machine-Learning, Springer-Verlag, 1983, pp. 463-482 (English).
Hayashi, Isao, et al. "Learning Control of Inverted Pendulum System Using Artificial Neural Network Driven Fuzzy Reasoning," 5th Fuzzy Symposium, Jun. 2-3, 1989, pp. 183-188. (Japanese; English Abstract).
(List continued on next page.)
Primary Examiner—Allen R. MacDonald
Assistant Examiner—Sanjiv Shah
Attorney Agent, or Firm—Fay, Sharpe, Beall, Fagan,
Minnich & McKee
A rule generation apparatus includes a label presenter in which when producing from training data including a set of specific values related to input and output variables rules representing input/output relationships between the input and output variables, numeric data of the training data is converted into categorical data expressed by symbols to generate an instance table, an RI device for extracting rules from the instance table, and a rule converter for converting the extracted rules into fuzzy rules. When the training data is divided to be distributively stored in a plurality of server processors, label assignment is conducted by each server processor such that a client processor later combines instance tables with each other to achieve rule induction and conversion.
14 Claims, 27 Drawing Sheets
Enbutsu, Ichiro, et al. "Exlractionof Explicit Knowledge From an Artificial Neural Network," IEEE Tokyo Section Denshi Tokyo, No. 30, 1991, pp. 101-104. (English).
Function of ES/TOOL/W-RI, Hitachi User Manual, 1990, pp. 33-53. (Japanese).
Maeda, Akira. et al. "A Fast Learning Algorithm for Fuzzy Membership Functions," 91-MIC-66-5,pp. 1-8. (Japanese; English Abstract).