arpit2412 / adaboost-decisionstumps

AdaBoost-DecisionStumps

ABSTRACT

Cancer is one of the major causes of human death in the world and breast cancer is the most common form of cancer in women. It may be a deadly disease but if diagnosed early on, it can be treated. Machine learning is growing at a very fast pace which has led to a time where we can even use it to predict cancer. This field has expanded by combining different methods to maximise the accuracy of prediction. Boosting is a machine learning model where we combine many weak classifiers to obtain one high-performance prediction model. AdaBoost is the most popular boosting algorithm. Apart from AdaBoost, Support Vector Machines (SVMs) is another successful classification method which is essentially similar as they both try to maximize the minimal margin on a training set. We have tried to implement a learning algorithm on the Wisconsin Diagnostic Breast Cancer (WDBC) by measuring their classification test accuracy. For the implementation of our AdaBoost model, the first 300 samples were used for training and the rest for testing. As a result of the experiment, it was observed that AdaBoost is the most accurate model with error as low as 0. 1269.Support vector machine did a pretty decent job with an accuracy of 0.902 when we used a non-linear kernel.