A Contemporary Machine Learning Method for Accurate Prediction of Cervical Cancer

With the advent of new technologies in medical field, huge amounts cancerous data have been collected and are readily accessible to research community. Over years, researchers employed advanced mining machine learning techniques develop better models that can analyze datasets extract conceived patterns, ideas, hidden knowledge. The mined information be used as a support decision making for diagnostic processes. These techniques, while being able predict future outcomes certain diseases effectively, discover identify patterns relationships between them from complex datasets. In this research, predictive model predicting outcome patients’ cervical cancer results has developed, given risk individual records preliminary screening tests. This work presents Decision tree (DT) classification algorithm shows advantage feature selection approaches prediction using recursive elimination technique dimensionality reduction improving accuracy, sensitivity, specificity model. dataset here suffers missing values is highly imbalanced. Therefore, combination under oversampling called SMOTETomek was employed. A comparative analysis proposed performed show effectiveness class imbalance based on classifier’s specificity. DT with selected features an accuracy 98%, sensitivity 100%, 97%. Tree classifier shown excellent performance handling assignment when reduced, problem addressed.