Feature Subset selection in Medical Data Mining using cascaded GA & CFS: A filter approach

Medical data mining has enormous potential for exploring the hidden patterns in the data sets of the medical domain. These patterns can be utilized for clinical diagnosis. Data preprocessing is a significant step in the knowledge discovery process, since quality decisions must be based on quality data. Feature subset selection is one of data preprocessing step, which is of immense importance in the field of data mining. The increased dimensionality of data makes testing and training of general classification method difficult. Feature subset selection reduces the number of attributes appearing in the discovered patterns, helping to make the patterns easier to understand. Further it enhances the classification accuracy and learning runtime this paper presents the development of a filter approach for feature subset selection experimented on the three medical dataset namely Pima Diabetic dataset, Heart Statlog Dataset and Breast cancer dataset available at UCI Machine Learning Repository. The proposed model consists of two stages. In the first stage, a filter approach with genetic algorithm (GA) and Correlation based feature selection has been used in a cascaded fashion. GA rendered global search of attributes with fitness evaluation effected by CFS. The second stage a fine tuned classification is done using three different classifiers namely Naïve Bayes, Bayesian and Radial basis function. Experimental results signify that the feature subset identified by the proposed filter GA+CFS when given as input to Naïve Bayes, Bayesian and Radial basis function classifiers, showed enhanced classification accuracy.