What is Entity Resolution?
By Senzing, published September 21, 2022
If you’re curious about what entity resolution is, you’re in the right place. Watch the video below as Senzing Founder and CEO, Jeff Jonas, explains what entity resolution is and how we define it at Senzing.
Entity resolution is the process of determining when real world entities are the same, despite differences in how they are described. Entity resolution answers tough questions like: are these two records for the same person or is this business the same as that one?
Three Pillars of Entity Resolution:
Senzing® entity resolution software uses entity-centric learning to examine the similarities and differences between entities. And, to identify relationships between entities for even greater accuracy and better insights. We focus on three areas when resolving entities:
- Recognizing when real-world entities are the same, despite differences in how they are described. Why is this important? Data about people and organizations is often messy or structurally inconsistent and requires fuzzy matching to connect the dots.
- Identifying when real-world entities are different, despite similarities in how they are described. Why is this important? You can have two records that are virtually the same except they have a one letter difference, such as a Jr. and a Sr., and you need to know these are actually two identities to reduce false positives.
- Understanding relationships between entities in data including possible matches, disclosed relationships and undisclosed or nonobvious relationships.
Entity resolution for enhanced data matching and faster, better decisions.
By applying all three pillars, Senzing entity resolution improves accuracy, provides more context and enables users to make better decisions faster.
Do you have more questions about what entity resolution is or how you should implement it? Contact us today to schedule a call with an entity resolution expert.
Video Transcript
Timestamps
0:00 Intro
0:16 What is Entity Resolution?
I thought I would just take a moment to explain what entity resolution is in the way that we think about it at Senzing. There’s a lot of different definitions for it, but ours is a little bit particular.
0:16 What is Entity Resolution?
So part one is recognizing when two records relate to the same identity despite having been described differently. You might see a whole bunch of differences between names, addresses, and fuzzy matching is the kinds of things that are necessary to be able to draw those two records together, okay.
The second part is recognizing when two records don’t relate to the same identity despite having been described similarly. You can have two records that virtually everything is the same except one letter off. One can say jr and one can say sr for junior and a senior, and you know, those are actually different identities.
And finally, if you want to do a really good job with entity resolution, you need to be able to track and manage the relationships between records, which might include a possible match that’s a form of a relationship or a disclosed relationship where you’re informed that these two people are say twins, and being able to record those relationships helps complete what’s necessary to do high-quality entity resolution.
There you go, entity resolution, defined Senzing style.