C4.5: programs for machine learning
C4.5: programs for machine learning
From rough set theory to evidence theory
Advances in the Dempster-Shafer theory of evidence
Fuzzy Sets and Systems - Special issue: fuzzy sets: where do we stand? Where do we go?
Information Sciences: an International Journal
Probabilistic approach to rough sets
International Journal of Approximate Reasoning
Hi-index | 0.00 |
One of the most important problems on rule induction methods is that they cannot extract rules, which plausibly represent experts' decision processes. It is because rule induction methods induce probabilistic rules that discriminates between a target concept and other concepts, assuming that all the concepts are on the same level. However, medical experts assume that all the concepts of diseases are belonging to the different level of hierarchy. In this paper, the characteristics of experts' rules are closely examined and a new approach to extract plausible rules is introduced, which consists of the following three procedures. First, the characterization of decision attributes (given classes) is extracted from databases and the concept hierarchy for given classes is calculated. Second, based on the hierarchy, rules for each hierarchical level are induced from data. Then, for each given class, rules for all the hierarchical levels are integrated into one rule. The proposed method was evaluated on a medical database, the experimental results of which show that induced rules correctly represent experts' decision processes.