Design and Development of Attendance Management and Analysis System using LLM

Abstract

This research addresses the present issues in attendance management system, including poor faculty user interface when handling large volume of student data and the time-consuming process of analysing and visualizing attendance and other student related information from the database using SQL queries. To resolve these challenges, a user friendly, web-based attendance management system is developed. This system can be deployed on both desktop and standalone devices. This system incorporates an SQL chatbot powered by a Large Language Model (LLM) to analyse and retrieve information from the database. The server side of this application is built with PHP scripting language, with MySQL as the database for managing students, faculties, subjects, timetable, and attendance data. The frontend of this software is developed using HTML 5.0, Sassy CSS (SCSS), and JavaScript to ensure a responsive and user-friendly user interface. A key feature of this system is the integrated MySQL chatbot, built with Python using the streamlit (U/I) framework, LangChain NLP library and OpenAI API. This chatbot interface directly connects with attendance database, allowing administrators to analyse and visualize data through natural language commands, eliminating the need for simple/complex SQL queries. Through natural language interaction, the chatbot can perform various database operations such as updating, modifying, and deleting records, thus significantly enhancing data management efficiency and usability.