Development of an Ensemble Multi-stage Machine for Prediction of Breast Cancer Survivability

Prediction of cancer survivability using the machine learning techniques has become a common approach in the recent years. In this regard, an important issue is that preparation of some features may require conducting difficult and costly experiments, while these features have less significant impacts on the final decision and can be ignored in the feature set. Therefore, developing a machine for prediction of survivability, which ignores these features for simple cases and yields an acceptable prediction accuracy, has turned into a challenge for the researchers. In this work, we have developed an ensemble multi-stage machine for the survivability prediction, which ignores difficult features for simple cases. This machine employs three basic learners, namely multi-layer perceptron (MLP), support vector machine (SVM), and decision tree (DT), in the first stage in order to predict the survivability using simple features. If the learners agree on the output, the machine makes the final decision in the first stage; otherwise, for difficult cases, where the output of learners is different, the machine makes decision in the second stage using SVM over all features. The developed model is evaluated using the Surveillance, Epidemiology, and End Result (SEER) database. The experimental results obtained reveal that the developed machine is capable of obtaining a considerable accuracy, while it ignores difficult features for most of the input samples.