Abstract
The AI Integrated Student Management System (AI-SMS) was developed to address inefficiencies in managing student data and administrative workflows in the Department of Information Technology at the University of Sri Jayewardenepura. The current manual processes and isolated digital tools create delays, inaccuracies, and an increased workload for administrative staff and faculty.
The proposed system is a comprehensive platform that integrates AI to automate letter generation, manage student data, and facilitate real-time communication. It features a dynamic dashboard tailored for students, lecturers, and administrators, enabling streamlined access to records and updates. Key functionalities include secure OAuth-based authentication, centralized data management, and real-time status tracking for letter requests. By incorporating Gemini 1.5 Flash, the system generates AI-powered academic letters, such as recommendation and internship introduction letters, significantly reducing the reliance on manual effort.
The system was developed using ReactJS for the frontend, Node.js and Express.js for backend operations, and MySQL as the database. Modules for bulk user registration, profile management, news dissemination, and letter review ensure end-to-end functionality. Rigorous testing, including unit, integration, and user acceptance testing, confirmed the system’s reliability and its alignment with the department’s objectives.
The AI-SMS offers notable improvements in efficiency and accuracy, empowering staff to focus on academic and strategic tasks while enhancing the student experience through timely responses and clear communication.
Background
The Department of Information Technology offers a B.Sc. in Business Information Systems (Special) and emphasizes blending IT and management principles. However, the absence of a comprehensive student management system has hampered administrative efficiency. Current practices rely on manual processes and Excel-based data management, leading to delays, inconsistencies, and challenges in meeting student needs effectively.
System Analysis
Problem Analysis
Manual operations resulted in inefficiencies such as delayed responses, a heavy administrative workload, and challenges in real-time communication. Processes like letter generation depended entirely on individual faculty members, creating bottlenecks and reducing overall productivity. Data management was confined to basic student details, failing to capture holistic student profiles.
Requirement Analysis
Functional Requirements:
- Student registration, profile updates, letter requests, and dashboard notifications.
- HOD and lecturer functionalities for letter approval and student record management.
Non-Functional Requirements:
- Secure user authentication using OAuth.
- Scalable architecture with user-friendly interfaces and data reliability.
System Design
The system was conceptualized with a centralized architecture featuring distinct modules for user management, letter generation, and communication. Entity-Relationship diagrams and activity workflows were created to ensure optimal database structure and system efficiency. A dynamic dashboard provides personalized views for students, lecturers, and the HOD, ensuring streamlined interaction and information dissemination.
System Development
Development Methodology
The project employed an agile methodology to ensure iterative development and stakeholder feedback at every stage. This approach allowed for rapid prototyping and adjustments based on testing outcomes.
Technologies Used
- Frontend: ReactJS, Bootstrap
- Backend: Node.js, Express.js
- Database: MySQL
- AI Tools: Gemini 1.5 Flash for letter generation
- Security: OAuth for authentication and bcrypt.js for password encryption
Developed System
Key features include an AI-powered letter generation tool, real-time status tracking for requests, and automated news dissemination. Users can securely log in and access personalized dashboards. The system also supports bulk data uploads for streamlined user registration.
Testing and Evaluation
Extensive testing included unit, integration, and user acceptance testing.
- Unit Testing: Verified individual functionalities such as login and data updates.
- Integration Testing: Assessed interactions between modules, such as letter approval workflows.
- User Acceptance Testing: Validated usability through end-user trials.
Performance evaluation highlighted fast response times (<3 seconds) and reliable operations under typical workloads.
Conclusion
The AI Integrated Student Management System adds substantial value to the department by automating repetitive tasks, improving communication, and centralizing data management. While the system addresses key challenges, future integration with existing university systems like LMS and the exams unit could further enhance its capabilities. Additionally, implementing multi-factor authentication and mobile access will strengthen security and usability. The project demonstrates the potential of AI-driven solutions in transforming administrative operations, paving the way for improved academic and administrative experiences.