Feature selection has become an important task for effective application of data mining techniquesin real-world high dimensional datasets. It is a process that selects a subset of original features by removing irrelevant and redundant features on the basis of the evaluation criteria without loss of information content. A feature selection method helps to reduce computational complexity of learning algorithm, improve prediction performance, better data understanding and reduce data storage space. Feature selectionhas gained more popularity in data mining and machine learning applications. The general procedure of feature selection process and overview of filter, wrapper and embedded method present in literature form the subject matter of this paper.