Comparison of Naive Bayes and Support Vector Machine with Grey Wolf Optimization Feature Selection for Cervical Cancer Data Classification

The cervix is part of the female reproductive system, and it can be affected by cancer disease. On a global scale, cervical cancer causes death and malignancy in women. However, it has a considerable chance of recovery when detected early. Along with the development of technology in various fields such as the medical field, the early detection of cervical cancer can be conducted through the use of machine learning classification methods. Therefore, this research aims to solve cervical cancer classification problems using Naive Bayes (NB) and Support Vector Machine (SVM) with Grey Wolf Optimization (GWO) feature selection. GWO feature is a wrapper feature selection method that was used to eliminate irrelevant features in classifying cervical cancer allowing NB and SVM to classify accurately. Thus, these methods are referred to as NB-GWO and SVM-GWO. The cervical cancer dataset used was numerical data from MRI obtained from Dr. Cipto Mangunkusumo Hospital. Furthermore, performance results from both methods were compared to find out which one is better in classifying cervical cancer data. The results indicated the highest average accuracy of NB-GWO and SVM-GWO with 96.30% and 95.37% respectively. Therefore, NB-GWO performed better in classifying cervical cancer data compared to SVM-GWO.