Performance Analysis of a new Filter and Wrapper Sequence for the Survivability Prediction of Breast Cancer Patients

Feature selection is an essential preprocessing step for removing redundant or irrelevant features from multidimensional data to improve predictive performance. Currently, medical clinical datasets are increasingly large and not every feature helps in the necessary predictions. So, techniques used determine relevant set that can performance of a learning algorithm. This study presents analysis new filter wrapper sequence involving intersection methods, Mutual Information Chi-Square followed by one methods: Sequential Forward Selection Backward obtain more informative improved prediction survivability breast cancer patients dataset, SEER. The improvement due this terms Accuracy, False Positive Rate, Negative Rate Area under Receiver Operating Characteristics curve tested using Machine algorithms: Logistic Regression, K-Nearest Neighbour, Decision Tree, Random Forest, Support Vector Multilayer Perceptron. supports over SEER dataset.