Feature selection using principal component analysis and genetic algorithm

Feature engineering is the way toward utilizing domain knowledge of the records to build features that in turn assist Machine Learning (ML) algorithms to provide efficient results. It is crucial to the utilization of ML and is both difficult and costly. The next buzz word after big data is feature engineering, which involves both feature selection and feature extraction. Feature Selection (FS also called attribute selection) is a procedure of selecting a subset of pertinent features for use in model building. It is an optimization problem. In our case, we have used principal component analysis for feature transformation followed by genetic algorithm to select optimal feature set and in the last, decision tree as a classifier. The proposed approach shows that use of principal component analysis before genetic algorithms improves the accuracy of the model with less number of features.