作者: X.-J.M. Zhou , T.S. Dillon
关键词: Theoretical computer science 、 Decision list 、 Decision table 、 Decision tree learning 、 Decision tree 、 Probabilistic logic 、 Binary tree 、 Computer science 、 Machine learning 、 Boolean function 、 Decision theory 、 Artificial intelligence
摘要: The authors develop a theory for general decision tree induction based on both the logical structure of concepts and probability distribution examples. discrete function is common analytic representation trees tables (rules). One most important classes functions disjunctive normal forms (DNF). Disjunctiveness has great effect accuracy speed concept learning. A developed Shannon's expansion DNF. function-equivalence, structural manipulations, irreducible DNFs are studied. For optimizing in context induction, functional criteria investigated.