Determining Student Graduation Patterns Based on Semester Package Service Types at Universitas Terbuka Using the Apriori Algorithm

Lora Ersitha (Universitas Bina Darma)
Yesi Novaria Kunang (Universitas Bina Darma)
Usman Ependi (Universitas Bina Darma)
Ari Muzakir (Universitas Bina Darma)

Formosa Journal of Computer and Information Science
Vol. 5 No. 2 (2026): August 2026

This study identifies graduation patterns associated with semester package services at Universitas Terbuka (UT) Palembang using the Apriori association rule algorithm. The dataset consisted of 1,134 alumni of the Bachelor of Management program who graduated during 2020–2025. The attributes were transformed into transaction items representing semester package service, study completion status, graduation predicate, and learning management system (LMS) usage. Apriori was implemented in RapidMiner with minimum support of 10% and minimum confidence of 70%, and lift was used to assess association strength. Eight rules satisfied the thresholds. The most representative service-related pattern was Timely, SIPAS Semi, and Very Satisfactory, with support 15%, confidence 81%, and lift 1.13. SIPAS non-TTM had slightly higher support as a single service pattern, but its combined Timely rule produced lower confidence and lift than SIPAS Semi.