Introduction

Graph databases have been revolutionizing the world as we know it. In the past decade, they have proven their usefulness in tons of cases, from social media networks to fraud detection to recommendation engines. The true strength of graph databases is the ability to efficiently represent the relationshipsbetween pieces of information. This especially applies to data that is highly connected, like hospital records.

In this blog, I’ll be walking through an example of how we take healthcare data stored in a graph database and visually show a sample of that information in 3D. Because of the large number of connections, 2D network graphs can quickly become cluttered and unreadable; however, these issues can be handled by changing our perspective. In cases where the 2D option fails, 3D graphs provide an expansive view of the data that is much easier on the eyes.

Since the data is stored in a graph database, it only makes sense to visualize it with a graph. That way, all we have to do is translate the connections that already exist in the database into something visible.

With that background, let’s start making our 3D graph masterpiece!

The full code for this blog can be found on my Github.

Designing a Graph Database Schema

The graph database I used for this project was built on TigerGraph using synthetic patient data generated by a neat, open-source program called Synthea. If you are unfamiliar with Synthea, check out this short blog I wrote or have a look at their website.

If you’re interested in seeing the process I used to design this healthcare network, check out my last blog post.

Otherwise, feel free to design your own graph or use a premade one. TigerGraph has tons of tutorials and examples that can help you design your own graph. Optionally, you can reference an article I wrote that takes you through the general graph design process.

Everything I mention from this point on can be done with any graph.

#data-science #analytics #3d #graph #visualization #data analytic