• Login
    View Item 
    •   SUSpace Home
    • Faculty of Science and Engineering
    • Department of Computer Science and Engineering
    • 2026-2030
    • View Item
    •   SUSpace Home
    • Faculty of Science and Engineering
    • Department of Computer Science and Engineering
    • 2026-2030
    • View Item
    JavaScript is disabled for your browser. Some features of this site may not work without it.

    An AI-Driven School Management System with Intelligent Student Assistance

    Thumbnail
    View/Open
    CSE-260286.pdf (377.2Kb)
    Date
    2026-05-02
    Author
    Sarker, Md. Al-Amin
    Metadata
    Show full item record
    Abstract
    School management software has become essential infrastructure for modern educational institutions, yet most existing platforms remain transactional in nature, offering little in the way of personalized academic support for students. At the same time, advances in Large Language Models have made conversational, context-aware tutoring practical, but such tools usually exist as standalone services disconnected from the institutional data that would make them most useful. This work bridges that gap. We present ClassMatrix, an AI-driven school management system that integrates a complete administrative platform with intelligent student assistance powered by OpenAI’s GPT models. The system is built using Laravel 13, Inertia.js v3, Vue 3, and Tailwind CSS v4. It supports five user roles Super Admin, Admin, Teacher, Student, and Parent through a role-based access control layer implemented with the Spatie permission package. Functionality is organized into eight delivery phases covering academic setup, daily operations, examinations, finance and human resources, library, transport, hostel, online learning, and an AI-enabled student panel. The student panel includes a personal-teacher chatbot that draws on the student’s own academic record (enrolled subjects, recent exam scores, attendance) to generate contextually grounded responses, together with an automated study-routine generator. Voice input and output are implemented through browser-native Web Speech APIs, eliminating any per-request cost for speech processing. The completed implementation comprises 190 routes, 64 Eloquent models, 61 Vue pages, 31 database migrations, and 16 backed enums. GPT-4o is used for routine generation due to its stronger structured-output capability, while the lighter GPT-4o-mini handles conversational tutoring at lower cost. Functional validation across all eight phases confirms that integrating institutional data with LLM prompts produces noticeably more relevant and personalized assistance than generic chat tools. The work demonstrates that a tightly integrated school management platform with AI capabilities is both technically feasible and pedagogically valuable, particularly for institutions seeking modern, low-cost alternatives to commercial enterprise solutions.
    URI
    http://suspace.su.edu.bd/handle/123456789/3038
    Collections
    • 2026-2030 [18]

    Copyright © 2022-2025 Library Home | Sonargaon University
    Contact Us | Send Feedback
     

     

    Browse

    All of SUSpaceCommunities & CollectionsBy Issue DateAuthorsTitlesSubjectsThis CollectionBy Issue DateAuthorsTitlesSubjects

    My Account

    LoginRegister

    Copyright © 2022-2025 Library Home | Sonargaon University
    Contact Us | Send Feedback