Safari, Erfia Nadia (2026) Rancang Bangun Chatbot Akademik Menggunakan Arsitektur Retrieval-Augmented Generation (RAG) (Studi Kasus: Prodi TIF UNIPMA). S1 thesis, Universitas PGRI Madiun.
![[img]](https://eprint.unipma.ac.id/style/images/fileicons/text.png) |
Text
Halaman Depan.pdf Download (637kB) |
|
Text
ABSTRAK.pdf Download (134kB) |
|
Text
BAB I.pdf Download (166kB) |
|
Text
BAB II.pdf Download (845kB) |
|
Text
BAB III.pdf Restricted to Repository staff only Download (306kB) | Request a copy |
|
Text
BAB IV.pdf Restricted to Repository staff only Download (2MB) | Request a copy |
|
Text
BAB V.pdf Download (152kB) |
|
Text
DAFTAR PUSTAKA.pdf Download (165kB) |
|
Text
LAMPIRAN.pdf Restricted to Repository staff only Download (2MB) | Request a copy |
Abstract
The increasing number of students has led to a growing demand for academic information services, while manual information delivery by administrative staff has been unable to provide fast and efficient responses. This study aims to design and develop an academic chatbot for the Informatics Engineering Study Program at Universitas PGRI Madiun using the Retrieval Augmented Generation (RAG) architecture and to evaluate its accuracy and usability. The system was developed using the Extreme Programming methodology. It utilizes Gemini Embedding 001 for the embedding process, ChromaDB as a local vector database, and Gemini 2.5 Flash Lite as the generative model. System evaluation was conducted using Retrieval-Augmented Generation Assessment (RAGAS) and User Acceptance Testing (UAT). The test results showed that a chunk size of 512 with an overlap of 30 provided the best performance, achieving a Semantic Similarity score of 0.9325, Answer Relevancy of 0.7804, Faithfulness of 0.9762, Context Recall of 1, Factual Correctness Precision of 0.7500, Factual Correctness F1 of 0.8229, and Factual Correctness Recall of 0.9529. UAT involving 85 respondents achieved a score of 88.9%, categorized as Very Good. The results demonstrate that the chatbot is capable of providing relevant, accurate, and user friendly academic information.
| Item Type: | Thesis/Skripsi/Tugas Akhir (S1) |
|---|---|
| Kata Kunci: | Chatbot, Retrieval-Augmented Generation, Gemini, ChromaDB |
| Subjects: | Q Science > QA Mathematics > QA75 Electronic computers. Computer science Q Science > QA Mathematics > QA76 Computer software |
| Divisions: | Fakultas Teknik > Teknik Informatika |
| Depositing User: | SAFARI NADIA ERFIA |
| Date Deposited: | 22 Sep 2026 07:56 |
| Last Modified: | 22 Sep 2026 07:56 |
| URI: | http://eprint.unipma.ac.id/id/eprint/9808 |
Actions (login required)
 |
View Item |
