Deep Learning on Curriculum Study Pattern by Selective Cross Join in Advising Students’ Study Path

Tekad Matulatan (Computer Science Department Universitas Maritim Raja Ali Haji Tanjung Pinang)
Muhammad Resha (Electrical Engineering Department Universitas Hasanuddin Makassar)

Proceeding of the Electrical Engineering Computer Science and Informatics
Vol 4: EECSI 2017

Advising engineering students in their study path need to understand the curriculum structure, student capabilities and challenge that commonly appear in courses. This paper offered the simple method to help student advisor in analyzing student performance in their study path based on academic progress record of the student it-self and pattern that been built from other students that have taken the courses. Using selective cross join for each  possible permutation of pair courses with respect to courses’ grade to create knowledge base. This knowledge base will be used to construct complex tree of any possible study path that might be taken by student to reach the end of study including course that must be retaken. Finding the best suggestion for study path using Monte Carlo tree search style