By Deepak Rangarao of IBM Analytics/Ecosystem, published October 22, 2018

This is our first guest blog on


Identity fraud isn’t a new topic. It costs government agencies/businesses a lot of money. As shown above, it is important to look proactively for scenarios where one person is pretending to be different people to benefit from different social programs. This fraud costs the taxpayer.

Respondeat Superior (Latin: “let the master answer”), a little known doctrine applicable under certain circumstances, holds employers responsible for their employee’s actions. A case for continuous vetting of employees. Something that even warranted a Presidential Executive Order.


We are seeing an increasing adoption of Machine Learning (ML) and Artificial Intelligence (AI) in everything we do but the outcomes from ML/AI are only as good as the data that is used. How do we get quality datasets to derive quality outcomes?

We could leverage Deduplication libraries that help find duplicate records and help us cleanse the dataset. But that will only take us so far. We can do better than that.

As Jeff Jonas would say it …. “Here’s the big idea” Context Gathering a.k.a Entity Resolution!

“Better understanding something by taking into account the things around it.”

What we really need is the ability to build context in addition to cleansing the data and removing duplicates. We need the ability to look at large datasets that span across multiple data sources and identify relationships and use it in conjunction with disclosed relationships to derive entity intelligence.

We would then know that Mary Smith was also pretending to be Marisol Smith. We would also know that Martin Johnson had unsavory friends. As a result, we should really watch out for potential consequences as a result of his emergency contact Emmet Johnson being on a sanctions list.

Entity Resolution is not just to get the bad guys. It can also help serve our customer’s better. By having a holistic view of our customers, we better understand their needs and are able to offer them better services.

With IBM Cloud Private for Data and our ecosystem partner, Senzing, we can now realize this dream on large datasets from diverse data sources using a cloud-native, private cloud platform for data and analytics. Watch the video below for a quick demonstration of some use cases. If you want to try it and see it in action you now have options :