Effect of Shallow Learning Techniques and Feature Selection for Breast Cancer Detection

Breast cancer is among the most prevalent cancers in women and one of the highest reason for women’s fatality rates. Most of the works in breast cancer detection are done either using deep learning models or heavily concentrated neural networks that require high computational machine and large datasets. This paper incorporates four shallow machine learning models, i.e., Naive Bayes (Both Gaussian and Categorical), K-Nearest Neighbor (KNN), and Decision Tree (DT) for detecting breast cancer. A state-of-art breast cancer dataset named–Breast Cancer Wisconsin has been incorporated for model training, testing, and validation. Data pre-processing and feature engineering has been done to make the raw data meaningful and robust for efficient model classification. An extensive data exploration and visualization is carried out for added understanding and correlation. On testing the different models, Gaussian Naive Bayes and Decision Tree managed to achieve the best accuracies with a highest accuracy of 97.36% and 96.61%, respectively.