Evaluasi Rekomendasi Dosen Pembimbing Berdasarkan Topik Penelitian Mahasiswa Menggunakan SBERT

Darmawan, Alfirda Moqta and , Yusuf Sulistyo Nugroho, S.T., M.Eng. (2026) Evaluasi Rekomendasi Dosen Pembimbing Berdasarkan Topik Penelitian Mahasiswa Menggunakan SBERT. Skripsi thesis, Universitas Muhammadiyah Surakarta.

BACA FULLTEXT
[img] PDF (Naskah Publikasi)
NASKAHPUBLIKASI_Alfirda.pdf
Download (571kB)
[img] PDF (Surat Pernyataan Publikasi)
SURATPERNYATAAN.pdf
Restricted to Repository staff only
Download (247kB)

Abstract

The selection of final project supervisors in the Informatics Engineering Department at Muhammadiyah University of Surakarta is still done manually. Students generally choose lecturers based on limited information without considering the suitability of the lecturer's expertise, which can lead to a mismatch between the final project topic and the supervising lecturer's expertise, resulting in a suboptimal final project. Therefore, this research needs to be conducted to analyze the semantic similarity between faculty publications and student thesis descriptions. The CBF approach was used in this study, utilizing SBERT to represent faculty publication texts and student final project descriptions into vector embeddings and cosine similarity to calculate similarity scores between vectors. Evaluation results showed that the method used produced an accuracy rate of 90%, with 9 out of 10 test data deemed suitable, indicating that SBERT and cosine similarity are proven to be quite effective in matching faculty publications with student final project descriptions. Further research could utilize data that more accurately reflects the actual expertise of lecturers, such as their areas of interest and mentoring history, so that evaluation does not solely rely on publications as the ground truth and recommendation results can be more accurate

Item Type: Thesis (Skripsi)
Uncontrolled Keywords: content-based filtering, SBERT, cosine similarity, semantic meaning,publication
Subjects: T Technology > Technical Information
Divisions: Fakultas Komunikasi dan Informatika > S1 Teknik Informatika
Depositing User: ALFIRDA MOQTA DARMAWAN
Date Deposited: 21 Feb 2026 03:43
Last Modified: 21 Feb 2026 03:43
URI: http://eprints.ums.ac.id/id/eprint/143105

Actions (login required)

View Item View Item
UMS ETD-db is powered by EPrints 3 which is developed by the School of Electronics and Computer Science at the University of Southampton. More information and software credits.