Implementasi Large Language Model (LLM) dengan LangChain untuk Perancangan Asisten Virtual Informasi BPJS.

Dary, Dany and Rahinda, Muhammad Fayyadh (2025) Implementasi Large Language Model (LLM) dengan LangChain untuk Perancangan Asisten Virtual Informasi BPJS. Project Report. [s.n.], [s.l.]. (Unpublished)

Text 5025211237_5025221224-Project_Report.pdf - Accepted Version Restricted to Repository staff only Download (4MB) | Request a copy

Abstract

informasi interaktif dan komprehensif terkait layanan BPJS. Proyek ini bertujuan untuk memfasilitasi interaksi pengguna dengan data BPJS, baik melalui kueri database maupun informasi yang bersumber dari dokumen. Pengguna utama aplikasi yang kami kembangkan mencakup masyarakat umum yang mencari informasi BPJS, serta pengguna dengan peran khusus seperti pegawai dan administrator. Aplikasi ini dibangun dengan antarmuka pengguna (frontend) menggunakan React dan Tailwind CSS, serta sistem backend yang dikembangkan dengan Flask (Python). Fitur-fitur utama meliputi interaksi chatbot dinamis, mode kueri database untuk pencarian data spesifik (misalnya, data pengguna atau keluhan), mode dokumen SOP untuk ekstraksi informasi dari PDF, serta manajemen pengguna dengan sistem registrasi, login, dan riwayat chat. Integrasi dengan Large Language Model (LLM) seperti ChatGroq dan OpenRouter API, bersama dengan database MySQL, memungkinkan pemrosesan bahasa alami dan pengelolaan data yang efisien.
====================================================================================================================================
The BPJS Chatbot is a virtual assistant developed to provide interactive and comprehensive information related to BPJS services. This project aims to facilitate user interaction with BPJS data, both through database queries and information sourced from documents. The primary users of the developed application include the general public seeking BPJS-related information, as well as users with specific roles such as employees and administrators. The application is built with a user interface (frontend) using React and Tailwind CSS, while the backend system is developed using Flask (Python). The main features include dynamic chatbot interaction, a database query mode for retrieving specific data (such as user data or complaints), a SOP document mode for extracting information from PDF files, and user management functionalities including registration, login, and chat history. Integration with Large Language Models (LLMs) such as ChatGroq and the OpenRouter API, along with a MySQL database, enables efficient natural language processing and data management.

Item Type:	Monograph (Project Report)
Uncontrolled Keywords:	Chatbot, BPJS, React, Flask, MySQL, LLM, Dokumen
Subjects:	Q Science > QA Mathematics > QA336 Artificial Intelligence Q Science > QA Mathematics > QA76.758 Software engineering
Divisions:	Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Informatics Engineering > 55201-(S1) Undergraduate Thesis
Depositing User:	Muhammad Fayyadh Rahinda
Date Deposited:	08 Jan 2026 04:13
Last Modified:	08 Jan 2026 04:13
URI:	http://repository.its.ac.id/id/eprint/129365

Actions (login required)

View Item