Title :
Knowledge processing system using multi-module associative memory
Author :
Hattori, Motonobu
Author_Institution :
Dept. of Electr. Eng. & Comput. Sci., Yamanashi Univ., Kofu, Japan
Abstract :
A knowledge processing system using multi-module associative memory for many-to-many associations (MMA)2 is proposed and simulated. In order to improve the fault tolerant performance, we employ Hopfield associative memories in the (MMA)2. Then the (MMA) 2 is applied to knowledge processing in which the knowledge is represented in a form of semantic network. The features of the proposed (MMA)2 are: 1) it can deal with characteristics inheritance, exception processing and plural queries in knowledge processing; 2) it is robust for noisy inputs; and 3) it can guarantee the recall of all training data
Keywords :
Hopfield neural nets; content-addressable storage; exception handling; fault tolerant computing; inheritance; knowledge based systems; semantic networks; Hopfield associative memories; characteristics inheritance; exception processing; fault tolerance; knowledge processing system; knowledge representation; many-to-many associations; multiple module associative memory; semantic network; Associative memory; Biological neural networks; Computational modeling; Computer simulation; Fault tolerance; Hebbian theory; Humans; Knowledge engineering; Robustness; Training data;
Conference_Titel :
Neural Networks Proceedings, 1998. IEEE World Congress on Computational Intelligence. The 1998 IEEE International Joint Conference on
Conference_Location :
Anchorage, AK
Print_ISBN :
0-7803-4859-1
DOI :
10.1109/IJCNN.1998.682323
