A Comparative Study between Data Mining Classification and Ensemble Techniques for Predicting Survivability of Breast Cancer Patients

Breast Cancer is the most common type of cancer prevalent among female cancer patients, while it also is the second most dreaded disease causing cancer death among women. A variety of data mining techniques, including Artificial Neural Networks (ANNs), Bayesian Networks (BNs), Support Vector Machines (SVMs) and Decision Trees (DTs) have widely been used in cancer research to facilitate the development of predictive models to create an effective decision making environment. This study proposes a criteria for prediction of survivability of breast cancer patients, based on the analysis performed using four data mining classification techniques, which include, Decision Tree, Multilayer Perceptron, Naïve Bayes and Random Tree and comparing the results with those of four ensemble techniques such as Adaboost M1, Bagging, Voting and Stacking. The dataset used in our experiment consists of 23 attributes containing 492 samples obtained from the Mizoram Cancer Institute of Aizawl, Mizoram, India. We are using data mining classifiers to predict the recurrence of breast cancer over a period of three years evaluated based on the comparison of their performance. Feature and attribute selections have been carried out to enhance the prediction accuracy of the computations.