IMPLEMENTASI RETRIEVAL AUGMENTED GENERATION UNTUK EVALUASI PROPOSAL TUGAS AKHIR MAHASISWA

Fanani, Ikhsan

Jurnal Teknologi Komputer dan Informatika
Vol 3 No 2 (2025): Jurnal Teknologi Komputer dan Informatika (TEKOMIN)

This research aims to develop a thesis title evaluation system based on Retrieval Augmented Generation (RAG), utilizing the thesis repository from the past year as a knowledge source. The system is developed by integrating a retrieval component that employs semantic embedding techniques to identify similar titles from the repository and a generative component that provides evaluation and improvement recommendations. The process includes preprocessing data from the thesis repository, implementing a sentence-transformers model to create a vector database, and integrating it with a Large Language Model (LLM). The test results on 20 new titles showed that the RAG system achieved an answer correctness score of 80%. The implementation also succeeded in automating and improving the objectivity of the evaluation process.