A Web Based Fuzzy Data Mining Using Combs Inference Method And Decision Predictor

Fuzzy logic has become a very popular method of reasoning a system with approximate input system instead of a precise one. When qualitative variables are used to determine the decisions then we have to create some specific functions where the membership values of the input can be any number between 0 to 1 instead of 1 or 0 which is used in binary logic. When number of input attribute increases it the combinatorial rules increases exponentially, and diminishes performance of the system. The problem is generally known as “combinatorial rule explosion”. The Information Technology Department of Minnesota State University, Mankato has been developing a system to analyze historical data and mining. The research paper presents a methodology to reduce the number of rules used in the application and creating a data prediction system using partial incomplete data set