Using the Breast Cancer Wisconsin (Diagnostic) Database, we can create a classifier that can help diagnose patients and predict the likelihood of a breast cancer. A few machine learning techniques will be explored. In this exercise, Support Vector Machine is being implemented with 99% accuracy. From the dataset, we will analysis and build a model to predict if a given set of symptoms lead to breast cancer. This is a binary classification problem, and a few algorithms are appropriate for use. Since we do not know which one will perform the best at the point, we will do a quick test on the few appropriate algorithms with default setting to get an early indication of how each of them perform. We will use 10 fold cross validation for each testing.
Log in or sign up for Devpost to join the conversation.