An intelligent consultant system for chess
An intelligent consultant system for chess
Inside Case-Based Explanation
Computers, Chess, and Cognition
Computers, Chess, and Cognition
Fuzzy Logic
A self-organizing retrieval system for graphs (organic, machine, chemistry, learning, partial-ordering)
Human Problem Solving
| Hi-index | 0.00 |
Ever since Shannon [Shannon 50] published his proposal for a chess playing program, most programs have followed the brute force approach to chess, which relies on searching a large number of possible chess positions in order to produce a move that is appropriate for a given chess position. Programs that have dominated the computer chess scene through the years, such as those described in [Marsland et al. 90], rely primarily on fast search-based algorithms and/or special purpose chess hardware rather than on an intensive application of knowledge. A few systems have used a knowledge-based approach to deal with chess positions. Unfortunately, these programs [Wilkins 80], [Pitrat 77] have been able to deal only with very limited subsets of the game. The problem of teaching chess has also been rarely explored. In addition to the canned tutorials found with several commercial chess programs, the intelligent tutoring system UMRAO [Gadwal 90], and the Chessmaster's natural language advisor, while limited, are perhaps the most representative examples of the application of artificial intelligence (AI) techniques for teaching chess. Despite the limited effort in this direction, cognitive psychology research suggests the importance of different factors, such as inexact pattern recognition [De Groot 78], [Newell et al. 72], and high-level knowledge [Cooke et al. 93] which have been successfully handled by AI techniques, such as case-based reasoning (CBR) [Schank 89], and fuzzy logic in the creation of learning environments in other fields, [Schank et al. 94], [Edelson 92], [McNeill et al. 94]. ICONCHESS combines some of these techniques in a learning environment for chess middlegames.