Decision tree generation using fuzzy measure theory
Kang Chang, Chien‐Hsing Wu · Journal of the Chinese Institute of Industrial Engineers · 1997
The methodology of generating rules for knowledge base development requires the understanding and regulation of several complex tasks. While ID3 algorithm is used to induce a decision tree from a set of examples, conversion either from a linguistic value to a numeric value or vice versa is necessary due to the requirement of information consistency. The conversion technique used is the key factor that determines the quality of the final decision trees. Uncertainty always exists in the real world. Fuzzy measure theory has been widely applied to help model the uncertain information in a realistic manner. This paper addresses the process that fuzzy measure theory is utilized in ID3 algorithm to enhance the accuracy of the generated decision trees. A model that employs fuzzy measure theory and ID3 algorithm to help generate rules is also delineated.