Louis, Ziven (2025) Implementasi dan analisis performa chatbot berbasis website untuk mendukung peternak ayam pemula di Indonesia menggunakan metode retrieval-augmented generation (rag) pada model llama 3.1. Bachelor thesis, Universitas Pelita Harapan.
![Title [thumbnail of Title]](http://repository.uph.edu/style/images/fileicons/text.png)
Title.pdf
Available under License Creative Commons Attribution Non-commercial Share Alike.
Download (158kB)
![Abstract [thumbnail of Abstract]](http://repository.uph.edu/style/images/fileicons/text.png)
Abstract.pdf
Available under License Creative Commons Attribution Non-commercial Share Alike.
Download (331kB)
![ToC [thumbnail of ToC]](http://repository.uph.edu/style/images/fileicons/text.png)
ToC.pdf
Available under License Creative Commons Attribution Non-commercial Share Alike.
Download (342kB)
![Chapter1 [thumbnail of Chapter1]](http://repository.uph.edu/style/images/fileicons/text.png)
Chapter1.pdf
Available under License Creative Commons Attribution Non-commercial Share Alike.
Download (406kB)
![Chapter2 [thumbnail of Chapter2]](http://repository.uph.edu/style/images/fileicons/text.png)
Chapter2.pdf
Restricted to Registered users only
Available under License Creative Commons Attribution Non-commercial Share Alike.
Download (1MB)
![Chapter3 [thumbnail of Chapter3]](http://repository.uph.edu/style/images/fileicons/text.png)
Chapter3.pdf
Restricted to Registered users only
Available under License Creative Commons Attribution Non-commercial Share Alike.
Download (2MB)
![Chapter4 [thumbnail of Chapter4]](http://repository.uph.edu/style/images/fileicons/text.png)
Chapter4.pdf
Restricted to Registered users only
Available under License Creative Commons Attribution Non-commercial Share Alike.
Download (1MB)
![Chapter5 [thumbnail of Chapter5]](http://repository.uph.edu/style/images/fileicons/text.png)
Chapter5.pdf
Restricted to Registered users only
Available under License Creative Commons Attribution Non-commercial Share Alike.
Download (310kB)
![Bibliography [thumbnail of Bibliography]](http://repository.uph.edu/style/images/fileicons/text.png)
Bibliography.pdf
Available under License Creative Commons Attribution Non-commercial Share Alike.
Download (293kB)
![Appendices [thumbnail of Appendices]](http://repository.uph.edu/style/images/fileicons/text.png)
Appendices.pdf
Restricted to Repository staff only
Available under License Creative Commons Attribution Non-commercial Share Alike.
Download (3MB)
Abstract
Dalam satu dekade terakhir, produksi daging dan telur ayam di Indonesia meningkat signifikan. Namun, pertumbuhan ini tidak diikuti oleh peningkatan jumlah peternak skala kecil. Banyak peternak pemula mengalami kegagalan akibat keterbatasan akses terhadap informasi mengenai manajemen kandang, nutrisi, dan kesehatan ayam. Kondisi ini dapat berdampak pada menurunnya keberlanjutan sektor peternakan rakyat. Penelitian ini bertujuan mengembangkan chatbot berbasis website untuk membantu peternak ayam pemula mendapatkan informasi relevan, akurat dan lengkap dengan menggunakan metode Retrieval-Augmented Generation (RAG). Metode ini memungkinkan chatbot mengambil referensi dari dokumen peternakan sebelum menghasilkan jawaban, sehingga menghasilkan jawaban yang lebih kontekstual dan informatif. Pengujian teknis mencakup black-box testing pada website untuk memastikan fungsionalitas berjalan tanpa bug kritis. Hasil evaluasi model RAG menggunakan metrik Contextual Precision, Contextual Recall, dan Faithfulness menunjukkan performa yang relatif lebih baik dibandingkan hasil dari pendekatan serupa dalam literatur sebelumnya. Efektivitas implementasi RAG juga divalidasi menggunakan metrik Answer Relevancy, BERTScore, dan BLEURT, dengan hasil kompetitif apabila dibandingkan dengan Llama 3.1 base dan GPT-4o. Human evaluation oleh 25 pengguna menilai kualitas jawaban berdasarkan relevansi (4,89), keakuratan (4,80), dan kelengkapan (4,68). Survei kepuasan terhadap 27 pengguna juga menunjukkan skor tinggi pada kualitas jawaban (4,73), niat penggunaan ulang (4,41), dan tampilan UI/UX (4,34).
/
Over the past decade, the production of chicken meat and eggs in Indonesia has increased significantly. However, this growth has not been accompanied by an increase in the number of small-scale farmers. Many novice farmers experience failure due to limited access to information regarding coop management, nutrition, and poultry health. This situation poses a threat to the sustainability of the local poultry farming sector. This study aims to develop a website-based chatbot to assist novice chicken farmers in obtaining relevant, accurate, and comprehensive information by utilizing the Retrieval-Augmented Generation (RAG) method. This method enables the chatbot to retrieve references from poultry-related documents before generating responses, resulting in more contextual and informative answers. Technical testing was conducted using black-box testing to ensure that the website functions without critical bugs. Evaluation of the RAG model using Contextual Precision, Contextual Recall, and Faithfulness metrics showed relatively better performance compared to similar approaches reported in previous literature. The effectiveness of the RAG implementation was also validated using Answer Relevancy, BERTScore, and BLEURT metrics, with competitive results when compared to LLaMA 3.1 base and GPT-4o. Human evaluation by 25 users rated the answer quality in terms of relevance (4.89), accuracy (4.80), and completeness (4.68). A satisfaction survey of 27 users also indicated high scores in answer quality (4.73), reuse intention (4.41), and UI/UX appearance (4.34).
Item Type: | Thesis (Bachelor) |
---|---|
Creators: | Creators NIM Email ORCID Louis, Ziven NIM03082210017 zivenlouisuph@gmail.com UNSPECIFIED |
Contributors: | Contribution Contributors NIDN/NIDK Email Thesis advisor Maulana, Ade NIDN0317049201 Ade.maulana@lecturer.uph.edu |
Uncontrolled Keywords: | Chatbot; Retrieval-Augmented Generation (RAG); Peternakan Ayam; Llama 3.1; Ketahanan Pangan; Evaluasi Model; Kecerdasan Buatan |
Subjects: | Q Science > QA Mathematics > QA75 Electronic computers. Computer science |
Divisions: | University Subject > Current > Faculty/School - UPH Medan > School of Information Science and Technology > Informatics Current > Faculty/School - UPH Medan > School of Information Science and Technology > Informatics |
Depositing User: | Ziven Louis |
Date Deposited: | 21 Jul 2025 08:33 |
Last Modified: | 21 Jul 2025 08:33 |
URI: | http://repository.uph.edu/id/eprint/69819 |