Comparative Performance Evaluation of Breast Cancer Detection Techniques

Breast cancer has been an area of concern these days. Many machine learning algorithms have promised solution for classification of benign and malignant conditions. In this paper, different algorithms such as support vector machine (SVM), K-nearest neighbor (KNN), decision tree, random forest, and hybrid techniques have been thoroughly investigated. Simulation results show that decision tree and random forest lead to over-fitting of data, whereas SVM and KNN performed well. The hybrid model suggested in this paper has inculcated the benefits of combining different models. The study helps to contribute medical professionals to make better and informed results.