<![CDATA[JPTIIK (Journal of Information Technology and Computer Science Development) is a platform that presents student journals of FILKOM UB. In this platform, journals have not been classified based on the thesis theme which refers to the 2018 FILKOM thesis guidebook, especially for the Information Systems department. From these problems, it was decided to classify topics in the majors of SI, FILKOM, UB based on the title, abstract, and a combination of abstract titles. There are 3 thesis themes used in the classification process, namely development, data and information management, and IS governance and management. The data collected was 300 with a comparison of 125 development, 100 SI governance and management, and 75 management data and information. This classification will compare the K-Nearest Neighbor and Support Vector Machine methods and will compare the classification results based on the title, abstract, and abstract title. Tests with a value of K=9 for the KNN method, a value of C=10 and iteration=50 on the title and abstract, and iteration=150 for the abstract title on SVM got the best accuracy value. The results of the classification based on the title and abstract of the SVM method get the highest accuracy value compared to the KNN method with the classification results in the title getting an accuracy value of 97.08%, precision 97.81%, recall 96.91%, and f-measure 97.11% while for Abstract titles get 97.08% accuracy, 97.93% precision, 96.67% recall, and 97.07% f-measure.]]>
Copyrights © 2022
