Rozi, Fahrur (2026) Pengembangan Chatbot Berbasis LLM (Large Language Model) Untuk Mempermudah Konsultasi Dan Pencarian Informasi Kehalalan Produk. Masters thesis, Institut Teknologi Sepuluh November.
![[thumbnail of 6047241044-Master_Thesis.pdf]](http://repository.its.ac.id/style/images/fileicons/text.png) |
Text
6047241044-Master_Thesis.pdf Restricted to Repository staff only Download (7MB) | Request a copy |
Abstract
Kebutuhan akan informasi kehalalan produk semakin meningkat seiring tumbuhnya kesadaran masyarakat Muslim terhadap konsumsi yang sesuai dengan syariat Islam. Namun, keterbatasan interaktivitas dan akses cepat terhadap data sertifikasi halal seperti yang tersedia di platform SIHALAL masih menjadi kendala. Penelitian ini bertujuan untuk mengembangkan chatbot berbasis Large Language Model (LLM) yang dapat memberikan layanan konsultasi dan pencarian informasi halal secara interaktif, kontekstual, dan berbasis sumber resmi. Sistem ini mengimplementasikan pendekatan Retrieval-Augmented Generation (RAG) yang memungkinkan integrasi antara model generatif dan dokumen kehalalan dari SIHALAL BPJPH, fatwa MUI, serta regulasi halal lainnya. Sistem dikembangkan menggunakan bahasa pemrograman Python dengan framework Flask untuk backend dan frontend. Arsitektur sistem mengadopsi pipeline modular dari LangChain, yang terdiri atas komponen PDFLoader untuk membaca dokumen, OpenAIEmbeddings untuk representasi vektor, HNSWLib sebagai vector store, PromptTemplate, serta integrasi dengan model GPT melalui ChatOpenAI. Frontend berbasis Flask dirancang untuk memberikan antarmuka web sederhana dan responsif bagi pengguna dalam mengajukan pertanyaan terkait kehalalan produk. Pengujian dilakukan secara internal dengan metode penilaian akurasi dan relevansi jawaban terhadap pertanyaan berbasis dokumen. Hasil evaluasi menunjukkan bahwa chatbot yang dikembangkan mampu memberikan respons yang faktual, relevan, dan konsisten dengan sumber regulatif. Penerapan teknologi RAG dan LangChain dalam sistem chatbot ini memberikan kontribusi nyata dalam mendukung digitalisasi layanan halal serta edukasi publik berbasis AI
================================================================================================================================
The demand for halal product information is increasing in line with the growing awareness among Muslims about consumption that complies with Islamic law. However, limited interactivity and fast access to halal certification data, such as those provided by the SIHALAL platform remain a challenge. This research aims to develop a chatbot based on a Large Language Model (LLM) that can provide interactive, contextual, and officially-sourced halal consultation and information search services. The system implements a Retrieval-Augmented Generation (RAG) approach, enabling the integration of generative models with halal documents from SIHALAL BPJPH, Fatwa MUI, and other halal regulations. The system is developed using the Python programming language with the Flask framework for both backend and frontend. Its architecture adopts a modular pipeline from LangChain, consisting of components such as PDFLoader for reading documents, OpenAIEmbeddings for vector representation, HNSWLib as the vector store, PromptTemplate, and integration with the GPT model via ChatOpenAI. The Flask-based frontend is designed to provide a simple and responsive web interface for users to submit questions related to product halal status. Internal testing was conducted using evaluation methods based on the accuracy and relevance of responses to document-based questions. Evaluation results indicate that the developed chatbot is capable of delivering factual, relevant, and regulation-consistent responses. The application of RAG and LangChain technologies in this chatbot system makes a concrete contribution to the digitalization of halal services and AI-based public education
| Item Type: | Thesis (Masters) |
|---|---|
| Uncontrolled Keywords: | Chatbot, LLM, RAG, LangChain, SIHALAL, Flask, Python, Halal, GPT, AI |
| Subjects: | T Technology > T Technology (General) T Technology > T Technology (General) > T58.8 Productivity. Efficiency |
| Divisions: | Interdisciplinary School of Management and Technology (SIMT) > 78201-System And Technology Innovation |
| Depositing User: | Fahrur Rozi |
| Date Deposited: | 29 Jan 2026 03:56 |
| Last Modified: | 29 Jan 2026 03:56 |
| URI: | http://repository.its.ac.id/id/eprint/130936 |
Actions (login required)
 |
View Item |