Teknik Kernel K-Means Clustering dan Support Vector Machine pada Sistem Evaluasi Akademik Mahasiswa

Abstract

The evaluation of the progress of students' academic achievement at a college in general is still performed manually based on CGP A with a view of the Student Study Result Card each semester. Further the evaluation of the number of credits achieved by students in a set period of time. This method is not effective because it cannot evaluate the students' academic achievement early inclination is whether to increase or decrease for each semester. The authors used a technique Kernel KMeans Clustering (KKMC) and Support Vector Machine (SVM) to construct a model of monitoring and evaluation of student academic achievement, both of the methods chosen because it is based on the kernel method, techniques are still relatively new in the educational data mining, pattern recogpition, bioinformatics, image processing and machine learning for data processing reliability in many dimensions. The study produced a model of monitoring and evaluation of student academic achievement. Data obtained from the University's academic database Almuslim Bireuen and survey the students to see the intrinsic factors that affect academic achievement is family support, motivation, and interest in student learning. The information generated by the monitoring and evaluation model can be used by management to enhance the quality of decision making by the leadership of the college.