Graph Databases: Law Enforcement, Fraud Detection and Social Networks

Today I am introducing a talk I did as part of employee knowledge sharing during my work with Teltech as a Golang developer. Previously, I had gained some exposure to Neo4J Graph Databases during a previous contract with GraphAware. I wanted to talk about the topic while the knowledge was still fresh in my mind. I held the talk over Zoom in November of 2021.

Enterprise application developers rarely use Graph Databases. My colleagues lacked experience with the topic. Therefore, it seemed like a good idea to do a talk. Thankfully, the folks at Neo4J publish some very nice whitepapers in which they outline use cases. Using their material and some hands-on experience, I did a short talk in front of my coworkers.

My audience was a little constrained, probably owing to the serious topics that we discussed. I am grateful that I got some questions, though. Every presenter wants their listeners to engage.

Why Graphs?

I give a brief theoretical and historical introduction into Graphs as a concept from discrete mathematics. It is a concept that has applications in some important technologies we use daily. In fact, both Google and Apple maps use this type of math to route you to your destination. It’s first use in analyzing paths goes back to 18th century Prussia.

The second and third part of the talk focus on two use cases – law enforcement and fraud detection. An underlying methodology connects both of these use cases. Experts study social networks to figure out what happened. They solve real-world problems looking at how the people involved are connected. Finally, graph databases are useful as they provide an easy way to traverse and visualize the realations involved.

I want to be clear that I am in no way involved nor was at any point with any organizations mentioned in the talk except Teltech and GraphAware.

Ivan Šarić

Engineering Blog