Cervical Cancer Classification using Machine Learning with Feature Importance and Model Explainability
Mahmudul Hasan, Priyanka Roy, Adiba Mahjabin Nitu
Abstract
Cervical cancer is still one of the most common gynecological cancers in the world. The rate at which cervical cancer develops is the fourth highest among common female disorders. It's one of the diseases that threaten women's health worldwide, and early symptoms are notoriously hard to see. In the fields of gynecology and computer science, there has been a dearth of studies focusing on the diagnosis of cervical cancer based on machine learning. Early-stage prediction of cervical cancer can be a great solution as it aware women to control their lifestyles also. This study classifies cervical cancer from secondary data using machine learning algorithms. Also, the top features responsible for cervical cancer are found out in this study. Support Vector Machine, Random Forest, and Logistic Regression are used as classifiers, and Boruta feature selection technique is used to find the best features to train the model. Two model explainable tools Explain like I'm 5 (eli5) and SHapley Additive exPlanations (SHAP) are used to rank the top feature, and their effect on the model are also analyzed. RF performs better than other classifiers using the Synthetic Minority Oversampling Technique with Tomek links (SMOTETomek) data balancing technique that greatly impacts model accuracy. It shows 99.85% accuracy with 100% precision, recall and f1 score. This proposed paradigm will help the medical domain people to predict the early stage of cervical cancer more accurately and can explain the behind information.