Improve Graph Analytics With Entity Resolution

Many organizations today are unlocking the power of their data by using graph databases to feed downstream analytics, create graph visualizations and more. Yet, when different graph nodes represent the same entity, graphs get messy. Watch this video with Senzing CEO Jeff Jonas on how adding entity resolution to a graph database condenses network graphs to improve analytics and save your analysts time.

When you use entity resolution to resolve graph nodes, your downstream analytics become much more effective. Critical graph visualizations with overly complex nodes and connections transform into network graphs that are much easier and faster for analysts to understand.

The quick-to-deploy Senzing^® entity resolution API enables graph database users to gain insights from their data they couldn’t see before. For example, instead of a graph with six related nodes, you get a resolved graph that condenses six nodes into one person. Adding entity resolution drastically reduces the time analysts spend extracting relevant information and insights from graph databases. You can try Senzing entity resolution for free in less than 15 minutes.

Timestamps
0:00 Graph Databases Connect Data for More Context
0:09 The Power of Accurate Graph Databases for Downstream Analytics
0:27 Entity Resolution Reduces Excess Graph Database Nodes
0:43 Entity Resolution and Graph Databases
1:09 Eliminate Time-Consuming Manual Processes
1:32 Senzing Entity Resolution Improves Entity-Resolved Graphs Out-of-the-Box

If you really want to unlock the power of data, you’re going to connect the data with other data, which gives it more context.

One of the great ways organizations are doing this today is graph databases. In graph databases, you’re getting these collections of nodes and how these nodes are connected. You’re then taking these node structures – nodes and edges – and you’re passing them to downstream analytics, and you’re putting them on charts for people to visualize, and more.

Well, the funny thing about this is that when these nodes are synonyms – say Bill, William, Willie and Billy – they are all (potentially) the same. What you’re missing here is entity resolution that can recognize that a bunch of nodes aren’t really, say, four degrees of separation, but that it’s really one person.

It becomes really beautiful when you combine entity resolution and graph databases. You get much more effective downstream analytics. You get much better machine learning from that entity-resolved graph data. And your network charts – that humans look at for investigative processes – aren’t just a cluster of pictures. They’re compressed pictures that make it much faster to figure out what’s going on.

What a lot of organizations are doing today is they’re looking at graph data and then manually figuring out if the records need to be collapsed, and then combining them. It takes a lot of effort, a lot of things can be missed, and – especially with the really clever bad actors – well, humans won’t even find. This is one way to better apply human resources, because why should you have your people do that when a machine can do it?

Senzing entity resolution allows people with graph databases to produce an entity-resolved graph out-of-the-box. So, it doesn’t look like six nodes that are all related when it’s really just one person. Instead of having a team of ten to thirty people looking at graphs and figuring out how to improve them, entity resolution is going to speed that up so those people can work on more interesting things.

Graph databases and entity resolution…sort of like peanut butter and jelly. No, strike that, like peanut butter and chocolate.

Related Videos