Applying Machine Learning Techniques for Early Prediction of Breast Cancer

Abstract: Women are the major strengths for world and family. Around 15 percent of the women are affected by breast cancer in various stages of their lives. It has become very common to find women with breast cancer nowadays. Though there are various treatments for breast cancer, the treatment gets effective only if it is identified at an earlier stage. It will be better if the cancer is diagnosed at the initial stage. Machine Learning provides a way in which the disease can be predicted at an early stage. There are various classification algorithms which can be used to classify the type of cancer mainly in terms of medical diagnosis. Using Breast Cancer dataset, classification algorithms like Linear Discriminant Analysis, Logistic Regression, KNN, SVM and Naïve Bayes are applied, and the data is analyzed for prediction. The classification algorithm determines the accuracy, recall, precision and F1 Score. SVM and Logistic regression has better performance than all other classification algorithms.

Keywords: Classification, Machine Learning, KNN, Logistic Regression, SVM, Recall.