Projection of Malignant Tumor of the Cervix using Machine Learning

Abstract

Cervical cancer is the second most common form of gynecologic cancer in less developed countries, after breast cancer. The Pap-Smear examination is now becoming as one of the most important screening methodologies in the speedy identification of this form of carcinoma, and among all strategies, the diagnostic test is the one that is most widely used in cervical cancer diagnosis. Machine Learning has the ability to provide accurate prognosis by using machine algorithm to perform classification, prediction, and estimation to achieve a high prediction rate. The Ensemble approach incorporates three machine learning techniques: K-Nearest Neighbor (KNN), Support Vector Machine (SVM) and Random Forest. With the precision percentage of 97.83 percent, the last technique provides more accurate results. To summarize, machine learning has the potential to achieve high diagnosis accuracy, while still being effective.