Linkurious Days

Linkurious Days London - 03 December 2024

9:00

Welcome Breakfast

9:35

Introduction & Vision

Sébastien Heymann

CEO | Linkurious

9:35

9:50

Modernizing the AML/CTF practice with state of the art graph technology

Alejandro Rivas Vega

Consumer FIU Analyst | Ria Money Transfer

9:50

10:20

Our graph technology journey: Getting graph insights to end users

Project head & Data scientist

UK Government Department

10:20

11:10

Threat modeling: Mitigating cybersecurity risks with graphs

Alexander Wallace

Threat lab Technology Lead | Lloyds Bank

11:10

11:40

Decision Intelligence: from Theory to Practice

Alan Brown

Chief Technology Officer | Aptitude Global

11:40

12:10

Q&A Panel Discussion

12:30

Lunch

13:45

Creating knowledge graphs from structured and unstructured data by leveraging entity resolution

Paco Nathan

Principal DevRel Engineer | Senzing

13:45

Knowledge graphs have become popularized within the context of Generative AI, for example in GraphRAG use cases to "ground" LLMs with domain-specific facts and mitigate "hallucinations" within the results of chatbots, agents, etc. However, there are much broader use cases for knowledge graphs in industries beyond GraphRAG. For example, investigative graph practices provide foundations for use cases such as anti-money laundering (AML), ultimate beneficial owner (UBO), and other areas of addressing risks, financial crime, sanctions compliance, cyber attacks, counter-terrorism, and so on. Moreover, graph algorithms can be used to "focus a lens" on data at scale, augmenting visualizations to reveal hidden insights in ways that graph queries alone cannot produce -- for a wide range of applications. There are also machine learning techniques emerging for measuring the quality of a graph as it evolves through data updates, identifying potential problems areas to troubleshoot.

While much of the contemporary AI literature emphasizes how to construct knowledge graphs automatically using LLMs, in this talk we will step back and examine the workflows needed for generating knowledge graphs from structured and unstructured data sources within regulated environments. Employing state-of-the-art open models and open source libraries at each step, we will explore how to combine the use of entity resolution and entity linking to produce graphs that emphasize data quality, curation, and feedback from domain experts, while making affordances for audits, evidence-based decision making, and practices required for mission-critical enterprise applications in highly-regulated environments.

14:15

From raw data to rich insights: Unlock the power of unstructured data