GraphIQ: Intelligent Social Influencer Ranking with Neo4j and Diverse Topical Contexts

Session Track: App Dev

Session Time: 05:15 am - 05:30 am November 6

Session description

Online Social Networks (OSNs) like Twitter are rich, dynamic platforms where influence spreads through complex and evolving interactions. Accurately identifying key influencers requires moving beyond surface metrics to a deeper, graph-based understanding of user relationships and topical relevance. In this session, attendees will learn how to model OSN data as a topical graph using Neo4j and apply the Aggregation Consensus Rank Algorithm (ACRA)—a novel method that aggregates multiple Twitter metrics to identify and rank the top-k influential users across time intervals. The session will demonstrate how Neo4j’s native graph capabilities support advanced influence analysis with both structural and semantic insights. We will also explore how to integrate machine learning models directly into Neo4j workflows, enhancing the precision and contextual relevance of influencer detection. Additionally, the session will introduce GraphRAG, a graph-based Retrieval-Augmented Generation (RAG) approach for enabling explainable AI on top of connected data. By the end of the session, attendees will understand how to build end-to-end solutions that combine graph technology, machine learning, and generative AI to uncover, rank, and explain influence in OSNs—offering new perspectives and tools for working with real-world social network data.