On the Computational Measurement of Intelligence Factors

摘要: In this paper we develop a computational framework for the measurement of different factors or abilities which are usually found in intelligent behaviours. For this, first scale measuring complexity an instance problem, depending on descriptional (Levin LT variant) ‘explanation’ answer to problem. We centre establishment either deductive and inductive abilities, show that their evaluation settings special cases general framework. Some classical dependencies between them shown way separate these is developed. Finally, some variants previous other possible ones be taken into account investigated. end, application measurements AI progress discussed.