Evaluating discovered rules from association rules mining based on interestingness measures using fuzzy expert system

Autor: Ahmad Nadali, Hamid Eslami Nosratabadi, Mahdieh Khalilinezhad, Sanaz Pourdarab
Rok vydání: 2011
Předmět:
Zdroj: Fourth International Conference on the Applications of Digital Information and Web Technologies (ICADIWT 2011).
Popis: Association rule mining is one of the popular pattern discovery methods in Knowledge Discovery in Databases (KDD). In this regard, all the obtained results from association rules are not useful and do not have same values. The aim of this paper is evaluating the extracted rules from data mining for bank credit customers. Here, a Fuzzy Expert System has been designed with considering interestingness measures as Input variables and Interestingness Rule level as output. Then, the system has been developed with the use of FIS tool of MATLAB software. This system is able to assess all the results of association rules according to the situation of each criterion. Finally, the presented steps have been run in an Iranian Bank as empirical study.
Databáze: OpenAIRE