Diagnosis of Cervical Cancer with Oversampled Unscaled and Scaled Data Using Machine Learning Classifiers
Nitin Kumar Chauhan, Krishna Pal Singh · 2022 IEEE Delhi Section Conference (DELCON) · 2022
Cervical cancer is a gynecological disease and one of the primary causes of death in females from cancer in developing countries. The recovery rate of this cervix infection can be escalated by premature detection and screening. Automated detection of cancerous cells is proving more efficient and robust in contrast to the medical examination. Machine Learning algorithms are the key to artificial intelligence (AI) in medical images processing. These methods are pretty much helpful in the detection of cancer cells at an initial stage. In this paper, we analyze the performance comparison of eight popular machine learning models as Support Vector Machine (SVM), Naive Bayes (NB), Linear Discriminant Analysis (LDA), Logistic Regression (LR), K-Nearest Neighbor (KNN), Decision Tree (DT), Multi-Layer Perceptron (MLP), and Random Forest (RF) on original and scaled data to detect the cervical cancer of malignant category. It is obtained from the analysis that the RF classifier performs superbly for the oversampled cervical cancer patient data with a standard scaler.