Penentuan Peminatan Mahasiswa Prodi Informatika di Universitas Amikom Yogyakarta Menggunakan SVM

Abstract

Universitas Amikom Yogyakarta has 16 courses, one of its courses in the Bachelor of Informatics Study Program, which is currently the number of students most commonly compared to other courses. By the time the student is in the final semester, the student is crucified completing the thesis. Thesis interest is still a difficult thing to be mapped by individual students or courses. The interest of the thesis will be in 3 classes, namely programming, networking, and Multimedia. SVM (Support Vector Machine) is one of the algorithms used in classifications. In this research, SVM was used to classify the interest of student thesis based on Holland's interest test assessment of 6 features (realistic, investigative, artistic, social, Enterprising and Conventional) which would then be used in Training data using the Support Vector Machine algorithm. The results of this study involved 150 datasets that resulted in a 66% accuracy rate using the 5-Fold Cross-Validation.

Keywords :

Classification, interest, SVM