A Two-phase Feature Selection Method using both Filter and Wrapper

Feature selection is an integral step of data mining process to find an optimal subset of features. After examine the problems with both the filter and wrapper approach to feature selection, we propose a two-phase feature selection algorithm of filter and wrapper that can take advantage of both approaches. It begins by running GFSIC(fi1ter approach) to remove irrelevant features, then it runs SBFCV(wrapper approach) to remove redundant or useless features. Analysis and experimental studies show that the effectiveness and scalability of the proposed algorithm. The generalization of neural network is improved when the algorithm is used to preprocess the training data by eliminating the irrelevant and useless features from neural network’s consideration.