Finding Fraud Part Two Revised

A few months ago I wrote up how to use the graph algorithms library to find fraud rings in bank data. The graph algorithms plugin has been a huge hit and received a promotion to a fully supported library with a team of developers, data scientists and product managers behind it. It was partially rewritten and given a fresh name. It is now called the Graph Data Science Library.

We’re going to give that fraud blog post a fresh look as well, change it to use the new library as well as throw more data at it. Please be sure you go back and read the original post right now so it’s fresh in your mind what we are going to do. Make sure you have the latest version of Neo4j Desktop running ( at least version 1.2.5 ), create a new graph with version 3.5.15 and install the plugin:

The Real Property Graph

Is not that thing above. That’s a Chart, not a Graph. But anyway…Neo4j is designed to support the property graph model natively. There are a host of other technologies that can bolt-on a “graph layer” of some kind. However it doesn’t make them a graph database. It’s like adding a rear spoiler to a van, sure it may look cool… or ridiculous, but it won’t make it a race car. Don’t fall for it. If you need fast graph queries, use a real graph database. But today we won’t talk about that. Instead we’re going to talk about the real property graph…

Finding Motifs in Cypher for Fun and Profit

If you are friends with Jessie, and Jessie is friends with Amy, there is a good chance you’ll eventually become friends with Amy too. In terms of a graph, this would be like a graph with three nodes and two relationships eventually building a third relationship to form a clique. This simple concept is one of the basis for recommendation engines. There are fancy terms for it, like “triadic closure” but basically it just means we are making triangles. But what about Amy’s friend Delilah? Is there a good chance now that you are friends with Amy that you’ll become friends with her? What about Jessie and Delilah? Can we extend the pattern to four nodes or five nodes and go beyond our simple triangle? Continue reading

Graph Analytics Book Jupyter Notebook for Chapter 8

If you’ve been going through the free Graph Algorithms book from Mark Needham and Amy E. Hodler you’ll eventually get to “Chapter 8: Using Graph Algorithms to Enhance Machine Learning”. This is a long chapter which walks us through how to use Graph Features to build and improve machine learning models. If you need a little help with it, take advantage of this public Jupyter notebook on Anaconda. Give it a shot, and let me know if you run into any issues.