<![CDATA[Human Papillomavirus is a virus that causes various types of diseases such as warts, infertility, miscarriage, vaginosis, and others. However, HPV status in tumors is a factor that helps in surviving and developing to survive in getting better response to radiotherapy and tumor control compared to tumors without HPV. Factors used to understand the problem or not. HPV does not only depend on status, age, age, tumor differences, sex and treatment strategies. But, also age, less exposure to tobacco and alcohol, as well as factors related to tumors. Classification and feature selection will be carried out to study features with significant weights used for the classification of HPV use in tumors. Algorithm flow in this research is by selecting features using the relief method, then classification using the naive bayes method is to predict the probability of class classification used in nominal and numeric type datasets. In this study, the appropriate features were obtained, namely, N_Category, T_Category, Tumor_side, Smoking_status_at_diagnosis, Tumor_substite, AJCC_Stage, and Age_at_diagnosis features. The best accuracy value is 90.97% by testing the number of features using 5 times, for each fold 25 test data and 98 training data are used. Meanwhile, the accuracy of testing the balanced data is 85% using 20 balanced data with 4 test data and 16 training data.]]>
Copyrights © 2020
