AI-Enabled Intelligent E-Library Ecosystem: A PWA-Based Framework with Predictive Risk Analytics and Dual-Portal Architecture
Tejaswi Kale, Swarupa Khaire, Sanika Sonawane, Shraddha Dhavale, Prof. Soniya Dhotre, Prof. Tejal Rane, Dr. Sandeep Kadam
The evolution of digital learning environments requires library systems that move beyond basic catalog automation toward intelligent, user-centric ecosystems. This research introduces an Intelligent E-Library Ecosystem developed as a Progressive Web Application (PWA) to enhance accessibility, automation, and operational transparency within academic institutions. The proposed system incorporates several advanced features that are rarely unified within conventional library platforms, including a formalized lost-book reporting workflow, real-time seat reservation management, automated due-date reminders, an integrated online PDF reader, a centralized digital notice board, predictive borrower risk analysis, and scheduled email notification services. A secure dual-portal architecture separates administrative control from student interaction, ensuring structured access management and data integrity. A Python–Flask-based machine learning microservice performs behavioral risk assessment to support proactive decisionmaking. Offline functionality is enabled through Service Worker implementation, ensuring continuity of access in low-connectivity environments. Experimental validation demonstrates improvements in process efficiency, user engagement, monitoring accuracy, and administrative responsiveness when compared to traditional manual and semi-digital library systems. Index Terms—Library Automation, Progressive Web App (PWA), Machine Learning, E-book Management, Late Fee Calculation, Lost Book Claim, Role-Based Access Control, Notice Board.

