Comprehensibility Improvement of Tabular Knowledge Bases

sugiuraQbtl.cl.nec.co.jp This paper discusses the important. issue of knowl-edge base comprehensibility and describes a tech-nique for comprehensibility improvement. Com-prehensibility is often measured by simplicity of concept description. Even in the simplest form, however, there will be a number of different DNF (Disjunctive Normal Form) descriptions possible to represent the same concept, and each of these will have a different degree of comprehensibility. In other words, simplification does not necessarily guarantee improved comprehensibility. In this pa-per, the authors introduce three new comprehen-sibility criteria, similarity, continuity, and confor-mity, for use with tabular knowledge bases. In addition, they propose an algorithm to convert a decision table with poor comprehensibility to one with high comprehensibility, while preserv-ing logical equivalency. In experiments, the algo-rithm generated either the same or similar tables to those generated by humans