Effectiveness of Classifiers for the Credit Data Set : An Analysis

Abstract

In today's world data mining is becoming an important area in terms of all business applications especially in the banking sector. In developing countries like India, bankers should be vigilant to fraudsters because they will create more problems to the banking organization. Application of data mining techniques helps the banks to look for hidden patterns in a group and discover unknown relationship in the data. Feature selection is a method used in data mining to select the most appropriate attributes for defining a relationship in a data set. It is very effective to build models based on these data mining techniques. There are several types of classifiers in data mining that helps to classify the records into two major groups based on the list of attributes. The proposed work is a comparative study of different types classifiers and evaluating the accuracy of the classifiers before and after applying the feature selection. After evaluating the results of experiment, it is easy to predict that feature selection is an important and necessary step during the process of data mining. From the results we can see that the performance metrics we obtained in different classifiers after applying feature selection is equal or better than that of before applying feature selection.