Breast Cancer Classification Using Machine and Deep Learning
Dhruv Kolhatkar, Geervani Reddy, Tanvi Raut, Om Jain, Jayashree V. Bagade · 2023
One of the diseases that causes the highest figure of deaths among the female population is breast cancer. Using Malignant and Benign tumors as the basis of classification we can detect the disease. The diagnosing process is required to differentiate; therefore, the process can be automated to recognize tumors. Various research has been conducted on breast cancer classification by applying different machine and deep learning algorithms to yield maximum efficiency and accuracy. In this paper, the algorithms SVM, Logistic Regression, KNN, Random Forest, Naive Bayes classifier and ANN have been used to classify Breast Cancer using Wisconsin Breast Cancer (Diagnostic) dataset. The performance of these classification algorithms has been compared to analyze which algorithm gives maximum accuracy. It was found that an ANN provided the most accurate results, with an accuracy of 98.7 %. The loss function used was binary cross entropy and the metric for backpropagation is accuracy. Standard batch size of 32 was used. Experimentation revealed that epochs less than or more than 100 gave diminishing results, hence the model was run for 100 epochs.