SailPoint’s Data Scientist Jostine Ho gives us a deep dive into the AI solutions used to fine-tune SailPoint Predictive Identity featuring network graphs and their importance to the platform.
Natalie Reina: Hello I’m Natalie Reina and today I’m with SailPoint data scientist Jostine Ho. At SailPoint the data science team works on key AI components that can fit into SailPoint Predictive Identity. So today we’re gonna ask Jostine a couple of questions, welcome Jostine.
Jostine Ho: Hi.
Natalie Reina: Hello, so my first question is can you kind of walk me through the approach that you take when you’re focusing on AI components for SailPoint Predictive Identity?
Jostine Ho: Yeah sure um so one way SailPoint is different from others in the IGA space is that we utilize something called network graphs, it is very basic and fundamental to the work we do. So network graphs give us a faithful representation of the social aspect and unique data structure in identity management. So here we’re not only interested in the one-dimensional things such as individual profiles and their attributes but we’re also interested in the multi-dimensional interactions and the large-scale insights that can only be brought to attention using machine learning and AI techniques.
Natalie Reina: So what is a network graph and why is it important?
Jostine Ho: Yeah so um to give you an idea the best examples I would give are Facebook and Twitter. So they describe their users and user interactions and their social activities using this spiderweb like graph, hence we call it a social network and it is the same idea when we look at IGA, where we have this highly social data structure that we can use to create an identity graph. And this identity graph is an accurate data model of the type of human activities that we’re interested in and these activities are best described by linking multiple identities and accounts in the form of a big connected information structure that can only be accessed by network graph.
Natalie Reina: So how will these graphs help organizations today stay secure and just stay up-to-date on what the latest innovations that identity has to offer?
Jostine Ho: Yeah that’s a good question, so um the way things are done before, using tabular spreadsheets, is that they rarely take into account the inherent connections and interactions in our data. So as a result, we’re losing a lot of valuable information as soon as we start recording it in such format but with network graphs we’re able to utilize machine learning and AI to help us identify and take actions based on the insights hidden in this web of information. And so here we’re optimizing large-scale systems doing things such as access modeling and smart recommendations and these capabilities help organizations have a more comprehensive and clear overview of their IGA landscape. This is all being done without sacrificing personalized recommendation at a granular level and I think that this is very powerful.
Natalie Reina: I think so too. Jostine thank you so much for your time this was very informative. This is Identity Talks be sure to check out more videos from our data science team on our digital hub.
Jostine Ho: Thank you.
You might also be interested in:
Find out how SailPoint can help your organization.