An Empirical Study on feature selection for Data Classification

In the task of pattern classification features play a very important role. Hence, the selection of suitable features is necessary as most of the raw data might be redundant or irrelevant to the recognition of patterns. In some cases, the classifier cannot perform well because of the large number of redundant features. This paper investigates the performance of different feature selection algorithms for the task of data classification. Different features play different roles in classifying datasets. Unwanted features will result in error information during classification which will reduce classification precision. The most of traditional feature selection can remove these distractions to improve classification performance. As shown in the experimental results, after feature selection using the traditional methods to control false discovery rate, the classification performance of DT’s and NB classifiers were significantly improved.