The Occam's Razor Principle for Data Mining Models Based on Degenerate Selfguessing Fuzzy Classification Algorithms

Abstract

Fuzzy classification models are one of the basic types of data mining models. The concepts of the simplicity and efficiency for fuzzy classifiers are introduced. We also introduced the concepts of consistent and degenerate selfguesssing fuzzy classifiers. The Occam’s Razor principle for data mining models based on fuzzy classification algorithms is formulated. The quality criterion for degenerate selfguessing fuzzy classifiers based on invariant simplicity measure is proved. The theorems on the conditions of improvement of degenerate selfguessing fuzzy classifiers are proved.