LEARNING FUZZY RULES FROM FUZZY DECISION TREES
Karol Matiaško, Ján Boháčik, Vitaly Levashenko, Štefan Kovalík · 2006
Classification rules are an important tool for disc overing knowledge from databases. Integrating fuzzy logic algorithms into databases a llows us to reduce uncertainty which is connected with data in databases and to increase discovered knowledge’s accuracy. In this paper, we analyze some possible variants of making classification rules from a given fuzzy decision based on cumulative informatio n. We compare their classification accuracy with the accuracy which is reached by stat istical methods and other fuzzy classification rules.