Predictive Machine Learning Approaches for Cervical Cancer Detection: An Analytical Comparison

Cancer is the illness that impacts people the most. Previous attempts for evaluating cancer disorders aided in determining what progressions should be performed on people with a high risk, decreasing their dangers. These cells may eventually result in cervical cancer. The proposed study made use of the dataset collected from Kaggle for cervical cancer prediction. During the evaluation of the collected data, there weren't any missing data values. This research has provided a prediction model that can assist in the identification or awareness of whether a patient has cervical cancer or not. The performance of the predictive system is measured using classification accuracy, F1-measure, recall, precision, Relative absolute error (RAE), and Root means square error (RMSE) analysis. On the basis of obtained results from various machine-learning algorithms, an analytical comparison has been done. The proposed model predicts cancer among patients using various machine learning classifiers wherein the Decision Tree exhibits the highest accuracy of 96.8%. Further, it is discovered that the Decision Tree scored better results in comparison to different machine-learning algorithms achieving the lowest RMSE and RAE of 0.15, and 41.6 respectively.