Entropy based rule derivation from data with uncertainty

Due to its advantages, fuzzy data model has been widely used to model and represent data with uncertainty. More and more applications show the needs to explore the data with uncertainty and to perform tasks of knowledge discovery in fuzzy database. This paper presents an attribute-oriented and probabilistic entropy based approach to knowledge discovery from uncertain data. The probabilistic entropy with the weighted values of membership functions is used to measure the possibility from fuzzy data sets. Also, it is employed to derive the rules that characterize these data sets.