Integrating Machine Learning Into Tracer Study: The MACA (Machine Learning Aided Curriculum Adjustment) Model for Computer Science Curriculum Evaluation

Felicia, Jennifer (2026) Integrating Machine Learning Into Tracer Study: The MACA (Machine Learning Aided Curriculum Adjustment) Model for Computer Science Curriculum Evaluation. Tugas Akhir (S1) - thesis, Universitas Bakrie.

	Text (Cover) 00 Cover.pdf.pdf - Accepted Version Restricted to Registered users only Download (1MB) | Request a copy
	Text (Bab 1-3) 01 Bab I-III.pdf.pdf - Accepted Version Restricted to Registered users only Download (5MB) | Request a copy
	Text (Bab 4) 02 Bab IV.pdf.pdf - Accepted Version Restricted to Registered users only Download (5MB) | Request a copy
	Text (Bab 5) 03 Bab V.pdf.pdf - Accepted Version Restricted to Registered users only Download (278kB) | Request a copy
	Text (DAFTAR PUSTAKA) 04 DAFTAR PUSTAKA.pdf.pdf - Accepted Version Restricted to Registered users only Download (454kB) | Request a copy
	Text (Lampiran) 05 Lampiran.pdf.pdf - Accepted Version Restricted to Registered users only Download (2MB) | Request a copy

Abstract

Evaluating higher education curricula is essential to ensure that academic instruction matches the fast-changing demands of the IT job market. However, traditional methods of reviewing curricula in Outcome-Based Education (OBE) are typically done by hand, rely on personal judgment, and are hard to apply consistently across larger scales. To tackle this problem, this study introduces and assesses the Machine Learning Aided Curriculum Adjustment (MACA) framework, a fully automated Decision Support System that identifies skill gaps and checks how well a curriculum aligns with the APTIKOM Version 2.0 guidelines for Computer Science programs. MACA combines information from various sources such as job listings from industries (LinkedIn and JobStreet), graduate tracer studies, national competency frameworks like KKNI, SKKNI, and APTIKOM, as well as academic course outlines from universities. The framework uses a mix of sentence embeddings, clustering methods, supervised machine learning, and rule-based reasoning to map and analyze how competencies align across various levels of the curriculum. The results indicate that MACA is effective in identifying important gaps in the curriculum. Most importantly, it highlights a major issue in documentation, where course syllabus place a strong emphasis on hard skills but give less attention to the important soft skills that are needed in the IT industry. In addition, the framework is capable of identifying issues related to institutional compliance, including incomplete mandatory courses and violations of prerequisite sequence requirements. Technical validation shows that contextual sentence embeddings effectively link academic descriptions with industry terminology, surpassing the shortcomings of conventional keyword-based matching methods. Overall, MACA offers an objective and data-based approach that supports the head of the program and the academic team in enhancing the quality of computing education. Keywords: Curriculum Evaluation, Outcome-Based Education (OBE), Decision Support System, Machine Learning, Sentence Embedding, Skill Gap Analysis, APTIKOM Guidelines.

Item Type:	Thesis (Tugas Akhir (S1) - )
Uncontrolled Keywords:	Curriculum Evaluation, Outcome-Based Education (OBE), Decision Support System, Machine Learning, Sentence Embedding, Skill Gap Analysis, APTIKOM Guidelines.
Subjects:	Computer Science > Decision Support System (DSS) Thesis > Thesis (S1)
Divisions:	Fakultas Teknik dan Ilmu Komputer > Program Studi Informatika
Depositing User:	Jennife Felicia
Date Deposited:	16 Sep 2026 09:02
Last Modified:	16 Sep 2026 09:03
URI:	https://repository.bakrie.ac.id/id/eprint/14112

Actions (login required)

View Item