Compact Discourse on Feature Selection

Abstract

Feature Selection is an emerging field of Machine Learning and Data Mining. Feature Selection helps to remove the inappropriate i.e. redundant and irreverent features from the dataset and develop an improve the classification accuracy. The paper has explored the different Feature Selection techniques like Filter method, Wrapper and Embedded method as basic methods.The Hybrid and Ensembled based Feature Selection have used in many papers and shown that the better Machine Learning Prediction accuracy.The different research papers have used the UCI Repository dataset for experiment purpose.Feature Selection is applicable to almost every field.