“How to Find a Killer Using Graphs: What Sherlock Was Really Doing with That Red String”
Every investigator faces the same challenge: connecting scattered facts and finding hidden patterns in complex data. Traditional analysis tools often fall short when dealing with the relational nature of investigative work, where context and connections are more important than isolated data points.
In this session, Paweł Gołąb will demonstrate Networks Notebook, an innovative system that transforms multimodal investigative data into knowledge graphs for automated analysis. You’ll discover how this murder board-inspired solution processes documents and images into interconnected graph structures, then applies sophisticated analytical methods to uncover hidden insights.
The presentation will showcase a two-stage analysis approach that combines precise graph queries with the interpretation of generative AI. Unlike traditional vector-based RAG architectures that flatten relationships and lose informational context, this graph-powered system preserves the relational nature of reality. You’ll learn how “”cold”” graph analysis using Cypher queries identifies key structures and anomalies, which language models then interpret to generate actionable hypotheses that neither technology could produce alone.
Attendees will explore practical applications through a case study analyzing a crime novel, demonstrating how the system detects probable event sequences, information flow paths, knowledge gaps, and conflicting witness statements. While the system is currently being deployed by investigative journalists at a major international publisher, Paweł will focus on the technical architecture and methodology that makes such real-world applications possible.
You’ll walk away understanding how to build automated knowledge graph extraction from text and images, implement graph-based RAG systems, and apply drill-down analysis techniques that follow each thread until the whole story unravels. The session covers practical implementation strategies for security, audit, legal, and medical applications, providing sufficient technical depth for immediate application in your projects.
Bio
Founder of Million Monkeys Software and co-founder of Bielik Consulting. Has collaborated on a variety of projects, ranging from mobile and web applications to criminal analysis systems. Passionate about projects that merge science with business. Currently focused on graph analytics, its connections to artificial intelligence, and the implementation and promotion of Bielik – the Polish language model. In his spare time, he lectures on computer science at the University of Warsaw.