DocumentCode
2072156
Title
The toolkit for development of hybrid expert systems
Author
Gavrilov, Andrev ; Novitskaya, Julia
Author_Institution
Novosibirsk State Tech. Univ., Russia
Volume
1
fYear
2001
fDate
26 Jun-3 Jul 2001
Firstpage
73
Abstract
The architecture of the toolkit ESWin (Expert Systems under Windows) and hybrid expert systems supported by it are described. Knowledge representation by rules and frames is used. The possibility of keeping and extracting data from external databases by external programs is provided. Frames allow one to describe an application domain in terms of hierarchies of classes and owners. Frames consist of slots, which can be symbols, numerical or linguistic variables, or date and time. Frames can be joined with rules and procedures, processing determined events. Three types of frames are used: frame classes; frame examples, in which facts are described; and frame patterns, in which situations (events, examples, etc.) are described. Inverse fuzzy inference is used as the main method of knowledge processing. Fuzzy rules are assigned and processed, on the one hand, by using certain factors of conclusions of rules, and, on the other hand, by using linguistic variables. A user can enter both numerical and fuzzy symbol values of corresponding slots using linguistic variables. The ESWin software consists of three modules - a shell-interpreter, an editor-constructor of knowledge bases and a program for the viewing and diagnosis of knowledge bases
Keywords
expert system shells; frame based representation; fuzzy logic; inference mechanisms; software architecture; uncertainty handling; ESWin; application domain description; class-owner hierarchies; data extraction; expert system shell interpreter; external databases; external programs; frame classes; frame examples; frame patterns; frame-based knowledge representation; fuzzy rules; hybrid expert systems development toolkit; inverse fuzzy inference; knowledge base diagnostics; knowledge base editor-constructor; knowledge base viewing program; knowledge processing; linguistic variables; rule conclusions; rule-based knowledge representation; slots; software architecture; symbol values; Artificial intelligence; Automation; Buildings; Computer architecture; Data mining; Databases; Diagnostic expert systems; Expert systems; Knowledge engineering; Knowledge representation;
fLanguage
English
Publisher
ieee
Conference_Titel
Science and Technology, 2001. KORUS '01. Proceedings. The Fifth Russian-Korean International Symposium on
Conference_Location
Tomsk
Print_ISBN
0-7803-7008-2
Type
conf
DOI
10.1109/KORUS.2001.975059
Filename
975059
Link To Document