An Explainable Artificial Intelligence Model to Detect Breast Cancer

The death rate of women has been increasing rapidly over the years due to carcinoma of the female breast. Every year, more than two million female get news that they have breast cancer, according to cancer research. Thus, early diagnosis and treatment can only reduce the death rate, which leads to a long life. It is crucial to classify tumors correctly; in that case, the machine-learning technique plays a vital role. Explainable AI is aiding in obtaining trust in the ML model by clarifying the decision-making process. This work has applied six different classifiers to the Wisconsin Breast Cancer Diagnostic dataset for breast cancer detection. Additionally, in order to evaluate the model's forecast and comprehend the significance of the features in its result, this study employed SHAP and LIME analysis to the logistic regression model. The best model was selected after evaluating and comparing the results of different performance matrices. Out of all the classifiers, Support Vector Machine and Logistic Regression had the best accuracy (98%).