Publication

Abstract : Cervical cancer, a dangerous disease arising in the cervix, the lower part of the uterus connected to the vagina, typically occurs by persistent infection with high-risk forms of human papillomavirus, also known as HPV, a sexually transmitted virus. This type of cancer affects women of all ages and has a major impact on morbidity and mortality rates, which has significant implications for world health. Investigating the details of the cause, risk factors, and preventive measures offers vital insights intothe continuous endeavors to tackle this huge health burden. In this study, we propose an ensemble approach for predicting cervical cancer, which identifies cancer indicators. This dataset was obtained from the UCI Repository which is used to classify the samplesthrough various classifiers like Logistic Regression (LR), Random Forest Classifier, k-nearest Neighbors Classifier (KNN), and Support Vector Classifier (SVM). To balance the classes, we used techniques such as Synthetic Minority Oversampling Technique (SMOTE) and Adaptive Synthetic Sampling (ADASYN). We interpret the results through certain Evaluation measures. The "Dx: Cancer" class is an imbalanced class with only 18 samples classified as cancer and 840 as not cancer, which translates to 2.1 percentclassified as cancer and 97.9 percent classified as not cancer. This metric measures the proportion of actual positives identified correctly. In terms of evaluating the model’s performance, this metric is fundamental. The F1 score is defined as the harmonic mean of precision and recall. Therefore, a high F1 score indicates both high precision and recall, as well as low and medium scoresif one score is high and the other is low. The Logistic Regression model and Support Vector Classifier performed equally well witha recall score of 99.4 percent.