Breast cancer survivability rate prediction using Neural network, Decision tree, K-nearest Neighbours and Naïve Bayes
Breast cancer is one of the most common forms of cancer. It is the most common among women only after skin cancer. The problem is of predicting the survivability rate of breast cancer based on available medical diagnosis, accurately, on which other work already exists. Existing work on breast cancer is on predicting and classifying breast cancer based on medical data with many different classification algorithms, prediction algorithms and different sources for datasets having been applied on this problem with various levels of success. Our work is on using 4 such algorithms: Naïve Bayes, KNN, decision tree, and a neural network, to predict survivability. We aim to improve upon the exiting work by combining various elements of feature extraction, dimensional reduction, optimization and shuffling of data to give classification models with high accuracy. We then also aim to have a comparison of the accuracies of each of the models in a comprehensive study of these models and which of them is best suited for the problem statement.