IRIS-Bot: Institutional Research Information System using Intelligent Automation and Deep Learning

Managing and verifying research publications within academic institutions is a major challenge due to fragmented storage, inconsistent metadata, duplicate submissions, and the lack of automated validation against trusted scholarly sources. Manual compilation of institutional research data—required for accreditation, rankings, audits, and faculty evaluation— is time-consuming, error-prone, and inefficient. To address this, we propose IRIS-Bot: Institutional Research Information System, an intelligent desktop application that automates end-to-end research publication management. The system integrates multiple advanced components, including automated PDF metadata extraction, DOI/CrossRef lookup, citation and indexing verification, journal authenticity checking, duplicate detection, semantic search, and domain classification. Using deep learning-based sentence embeddings and pgvector-powered similarity search, IRIS-Bot provides accurate retrieval and clustering of research documents. A modular backend architecture with PostgreSQL as the primary datastore ensures scalability, while serves as a lightweight for offline use. The Qt-based graphical interface provides an interactive and user-friendly environment for administrators and faculty. Experimental evaluation demonstrates that IRIS-Bot significantly improves accuracy in metadata extraction, reduces duplication errors, and enhances search efficiency using semantic embeddings compared to traditional keyword-based methods. The proposed system offers a unified solution that enables institutions to maintain clean, verified, and searchable research repositories with minimal manual effort. IRIS-Bot can be effectively deployed for academic audits, accreditation processes, and institutional research analytics, contributing to improved data quality and automation in higher education environments Keywords: Institutional Research System, Deep Learning, NLP, Metadata Verification, CrossRef, Automation, Academic Data Integration