NEURAL NETWORK ALGORITHM FOR CLASSIFICATION OF STUDENT GRADUATION IN FACULTY OF ECONOMICS, UNIVERSITY OF GARUT
- Prosiding Internasional
- Fikri Fahru Roji, Agna Hilyah, Ridwan Setiawan, Diqy Fakhrun Shiddieq
- Proceedings of International Conference on Sustainable Collaboration in Business, Technology, Information and Innovation (SCBTII) Third Edition, December 2022 Vol 3. Year 2022 e-ISSN: 2621-3192
Abstrak
The academic performance is one of aspect which has remained the benchmark of the success in learning activities at the university. The indicator of academic performance in the university is the students able to complete their studies on time. Unfortunately, the problem regarding academic performance was associated with the completion time of student studies in Faculty of Economics, Garut University. In this research explore the model that able to classify the graduation of student through the data mining classification technique using Neural Network Algorithm. The classification conducted by evaluating the academic performance based on Semester Performance Index (IPS) during first years, Semester Credit Unit (SK S) at second years and use the demographics of students as attributes that will be used in the dataset. Based on the examinations that conducted by using k-fold cross validation, there are 12 attributes that influence the graduation of students. Moreover, based on the evaluation of accuracy and recall score, with some modified number of hidden layers, and number of hidden nodes and removing some features will indicated the high value. The model that represented able to applied for classify the active students from 2012 to 2017 that indicate accuracy value of 82.96% and recall value was 66.54%. Keywords: Data Mining; Student Graduation; Classifications; Neural Network; K-Fold Cross Validation; Confusion Matrix.
Lihat Dokumen