Smart University Assistant: Leveraging Graph Databases and Vector Embeddings for Efficient Information Retrieval

This research presents the Smart University Assistant (SUA), an AI-driven system designed to improve student access to campus resources at Utah Valley University (UVU) through a combination of graph databases and vector embeddings. Unlike traditional keyword-based search systems, SUA extracts and stores information at the paragraph level in a Neo4j graph database, enabling precise and context-aware retrieval. Each paragraph is embedded into a high-dimensional vector space using models like SentenceTransformers, facilitating semantic search through similarity matching. By integrating structured relationships with deep semantic understanding, SUA provides more accurate, contextually relevant responses to student queries. This approach enhances the efficiency and scalability of university resource discovery, setting a new standard for AI-driven academic support systems.