2003 | OriginalPaper | Chapter
Characteristics of Accuracy and Coverage in Rule Induction
Author : Shusaku Tsumoto
Published in: Rough Sets, Fuzzy Sets, Data Mining, and Granular Computing
Publisher: Springer Berlin Heidelberg
Included in: Professional Book Archive
Activate our intelligent search to find suitable subject content or patents.
Select sections of text to find matching patents with Artificial Intelligence. powered by
Select sections of text to find additional relevant content using AI-assisted search. powered by
Rough set analysis are closely related with accuracy and coverage. However, there have been few studies on the formal characteristics of accuracy and coverage for rule induction have never been discussed until Tsumoto showed several characteristics of accuracy and coverage. In this paper, the following characteristics of accuracy and coverage are further investigated: (1) The higher the accuracy of the conjunctive formula become, the lower the effect on the conjunction will become. (2) Coverage will decrease more rapidly than accuracy. (3) The change of coverage becomes very small when the length of the conjunctive formula becomes larger. (4) The discussions above are corresponding to those on sensitivity and specificity. (5) When we focus on accurate classification, the classification efficiency, which is the product of sensitivity and specificity will become lower.