Cervical Cancer Classification: Optimizing Accuracy, Precision, and Recall using SMOTE Preprocessing and t-SNE Feature Extraction

Machine Learning(ML) algorithms are used in cervical cancer categorization to determine whether a person has cervical cancer based on pertinent information from medical records. This procedure is critical in healthcare for prior treatment. For this kind of categorization problem, many ML approaches such as RF(Random Forest), SVM(Support Vector Machine), and LR(Logistic Regression) can be used. We employed ML techniques to increase the accuracy, precision, and recall of cervical cancer diagnosis in this study. SMOTE preprocessing was used to resolve data imbalance by producing synthetic samples for the minority class. Furthermore, t-SNE feature extraction was used to capture complex structures in the data. SVM consistently outperforms RF and Logistic Regression when combined with SMOTE preprocessing and t-SNE feature extraction, demonstrating improved accuracy, precision, and recall rates. This research highlights the efficacy of these strategies in enhancing cervical cancer categorization, indicating the possibilities for precise and reliable forecasting in medical investigations. The proposed SVM gave accuracy of 92.60%, precision of 0.91 and recall of 0.90 respectively. The tool used for execution is python.