作成者 |
|
本文言語 |
|
出版者 |
|
|
発行日 |
|
収録物名 |
|
巻 |
|
号 |
|
開始ページ |
|
終了ページ |
|
出版タイプ |
|
アクセス権 |
|
Crossref DOI |
|
関連DOI |
|
|
関連URI |
|
|
関連情報 |
|
|
概要 |
An information theoretic view is introduced into the discussion of knowledge representation. Information is defined to a description as the difference between the entropies the system contains before ...and after the description is given. Properties that should be attributed to knowledge representation are discussed taking this view into account. Then, it is shown that an extension of first order predicate logic has these properties and is therefore suited for knowledge representation. A formula in this logic involves data structures in such a way that the data structures can be manipulated without having any influence upon this formula and, nevertheless, the formula evaluates correctly the data structures it refers to.続きを見る
|