A Hybrid Feature Selection Method for Improve the Accuracy of Medical Classification Process

Generally, medical dataset classification has become one of the biggest problems in data mining research. Every database a given number features but it is observed that some these can be redundant and harmful as well disrupt process this problem known high dimensionality problem. Dimensionality reduction preprocessing critical for increasing performance machine learning algorithms. Besides contribution feature subset selection gives significant improvement accuracy. In paper, we proposed new hybrid approach based on (GA assisted by KNN) to deal with issues biomedical classification. The method first applies combination between GA KNN find optimal where accuracy k-Nearest Neighbor (kNN) used fitness function GA. After selecting best-suggested features, Support Vector Machine (SVM) are classifiers. experiments five datasets UCI Learning Repository. It noted suggested technique performs admirably databases, achieving higher while using fewer features.