Automated Institutional Query and Feedback Processing using DistilBERT in a Unified Management System

Educational institutions today face numerous challenges in managing administrative tasks, such as manual attendance tracking, fragmented communication, and lack of real-time academic performance monitoring. These challenges lead to inefficiencies, errors, and delays in decision-making. Furthermore, traditional methods of managing student records, faculty performance, and class schedules are time-consuming and prone to errors. Existing systems often fail to provide an integrated approach, leading to disconnected platforms for different stakeholders—students, faculty, and administrators. This paper presents a Unified Institution Management System that leverages advanced technologies to address these issues. The proposed system integrates a role-based web and mobile application that automates attendance tracking, real-time academic performance monitoring, and centralized communication. By utilizing DistilBERT, a transformer-based deep learning model, the system efficiently handles Natural Language Processing (NLP) tasks such as sentiment analysis and feedback categorization, which enhances communication and academic transparency. The system also ensures secure, structured access for all user roles—administrators, HODs, faculty, and students—thus streamlining administrative operations and reducing overhead. Key features of the system include automated attendance management, real-time class scheduling, dynamic faculty substitution, and a comprehensive dashboard for academic monitoring. This approach provides a unified, scalable, and user-friendly platform designed to enhance institutional productivity and decision-making. The proposed system aims to modernize educational administration by improving efficiency, reducing errors, and providing stakeholders with timely, accurate information.