Abstract: "This paper describes a learning method for building knowledge bases. There are two types of knowledge acquisition systems which extract knowledge from human experts, interactive and non-interactive. This paper describes a non-interactive knowledge acquisition system which acquires a...

Ausführliche Beschreibung

Bibliographische Detailangaben

1. Verfasser:	Taki, Hirokazu (VerfasserIn)
Format:	Buch
Sprache:	Japanese English
Veröffentlicht:	Tokyo, Japan 1988
Schriftenreihe:	Shin-Sedai-Konpyūta-Gijutsu-Kaihatsu-Kikō <Tōkyō>: ICOT technical memorandum 512
Schlagworte:	Künstliche Intelligenz Artificial intelligence Expert systems (Computer science) Machine learning
Zusammenfassung:	Abstract: "This paper describes a learning method for building knowledge bases. There are two types of knowledge acquisition systems which extract knowledge from human experts, interactive and non-interactive. This paper describes a non-interactive knowledge acquisition system which acquires a human expert's knowledge by observation. It learns the human expert's problem solving strategies and makes logical rules from temporal sequential data. The learning method of the knowledge acquisition system is interpretation based learning (IBL), which uses advance knowledge in the learning process. Advance knowledge of IBL consists of domain concepts, concept relations and interpretation knowledge, which translates observed data into internal concepts Although explanation based learning (EBL) also uses advance knowledge, which consists of domain theory and operationality criteria, it learns knowledge using the domain theory, but IBL learns the domain theory itself. IBL is a useful knowledge acquisition method when a domain theory has not been prepared.
Beschreibung:	8 S.