@article{Boonbangyang_Nonsiri_2021, title={Cancer Epitope Classification}, volume={11}, url={https://ph02.tci-thaijo.org/index.php/JIST/article/view/245671}, DOI={10.14456/jist.2021.18}, abstractNote={<p>Cancer is a leading cause of death in the world. In 2020 World Health Organization (WHO) reported that approximately 10 million deaths caused by cancer and will increase for the coming years. This research paper aims to study the prediction of cancer epitope using machine learning for classifying between cancer cell surface and epitope on healthy cell surface. The comparison between the different machine learning algorithms is presented. This work can help to training T-cell for recognizing cancer cell and release enzyme to kill cancer cell (Targeted Therapy). The experiment results shown that imbalance data the model from Support Vector Machine (SVM) calculated based on Dipeptide Composition (DPC) feature achieved the best accuracy of 79% Sensitivity 16% and Specificity 100% on test dataset. While balance data with SMOTE Random Forest (RF) calculated based on Dipeptide Composition (DPC) feature achieved the best accuracy of 80% Sensitivity 28% and Specificity 96% on the same test dataset. In conclusion, Support Vector Machine (SVM) and Random Forest (RF) calculated based on Dipeptide Composition (DPC) feature can employ these models for predicting the cancer epitope in imbalance dataset and balanced dataset.</p>}, number={2}, journal={JOURNAL OF INFORMATION SCIENCE AND TECHNOLOGY}, author={Boonbangyang, Manon and Nonsiri, Sarayut}, year={2021}, month={Dec.}, pages={72–83} }