DocumentCode
1681293
Title
Retention and forgetting of the memory in the hierarchical knowledge system
Author
Shim, Jeong-Yon ; Hwang, Chong-Sun
Author_Institution
Dept. of Comput. Software, YongIn SongDam Coll., South Korea
Volume
3
fYear
2002
fDate
6/24/1905 12:00:00 AM
Firstpage
2849
Lastpage
2853
Abstract
For the purpose of building the more adaptable intelligent knowledge system to the dynamic environment,it is necessary to construct the knowledge structure with the strategy for the efficient memory management first of all. The efficient memory management can make the retention of memory and knowledge retrieval easier. Specially the concept of forgetting solves the problem of knowledge retrieval from the complex and large database in memory. Accordingly in this paper we adopt the theory relating short-term memory, long-term memory and forgetting and propose the intelligent knowledge system with the hierarchical associative memory managed by storage management strategy. It has the function of adaptive learning, selective processing, perception, inference and data extraction. This model is applied to the medical area and tested with symptom data
Keywords
content-addressable storage; database management systems; information retrieval; knowledge based systems; storage management; adaptable intelligent knowledge system; adaptive learning; data extraction; efficient memory management; forgetting; hierarchical associative memory; hierarchical knowledge system; inference; knowledge retrieval; long-term memory; memory retention; perception; selective processing; short-term memory; storage management strategy; Associative memory; Buildings; Databases; Environmental management; Information retrieval; Intelligent structures; Intelligent systems; Knowledge based systems; Knowledge management; Memory management;
fLanguage
English
Publisher
ieee
Conference_Titel
Neural Networks, 2002. IJCNN '02. Proceedings of the 2002 International Joint Conference on
Conference_Location
Honolulu, HI
ISSN
1098-7576
Print_ISBN
0-7803-7278-6
Type
conf
DOI
10.1109/IJCNN.2002.1007600
Filename
1007600
Link To Document