Fuzzy interpretation of induction results
Xindong Wu, Petter Måhlén · 1995
When applying rules induced from training ex-amples to a test example, there are three possi-ble cases which demand different actions: (1) no match, (2) single match, and (3) multiple match. Existing techniques for dealing with the fist and third cases are exclusively based on probability estimation. However, when there are continuous attributes in the example space, and if these at-trihrrtnn hilvn haan Awrc=.ti.caA intn Lqkprv & he “-.l-..u ” _.... ” I--aa-.,-.--------“-fore induction, fuzzy interpretation of the discre-tised intervals at deduction time could be very valuable. This paper introduces the idea of us-ing fuzzy borders for interpretation of discretised intervals at deduction time, and outlines the re-sults we have obtained with the HCV (Version 2.0) software.