With the changes in e-commerce and the new consuming paradigm, you don’t get to a place where there are just non-stop, fraudulent transactions. It’s quite tricky to use AI and NLP in Fraud Detection, but let’s dive in and see what’s possible.

So in the world of fraud detection, there are a few categories in e-commerce, and some of the big ones are bot detection, so you want to really start looking at whether or not, the transaction that is occurring on your e-commerce platform, is a true human, or whether it’s a bot.

You want to have a firewall or a point-of-entry system that allows you to detect these rather quickly and, really understand the kind of normal usage patterns that your site would get and probably track them through time.

One thing on people’s minds is the safety of their personal information. In terms of data breaches, that’s very often up to the company’s infrastructure and security people.

If you have a lot of companies get hacked by not having their databases available online but possibly by having an API, an actual interface to your database available, one of the keys so one of the tokens or one of the authentication keys, or secret access to use, whatever you call it there that gets found breached or copied somewhere and is public.

In terms of detection and forensics, there’s a lot of AI and there’s a lot of modeling and analysis that can happen, but this is something that needs to happen way before you get to a place where you’ve been breached.

For example, Insurance companies have systems or people in places that will do a bit of applicant research, so while you’re applying to their back office look through many signals, including any history they may have on you.

The other way that they use AI and NLP is on the claim side, so doing a similar thing and looking for signals that may state, why you have such a claim, it’s insurance so it can get messy, imagine somebody going through a divorce and they must pay out to the other party.

The first big neural networks deep learning moles were in a bank about risk tolerance and risk app type. This lends itself well to the use of NLP and AI in Retail Banking going forward. It was really based on strictly AI, not really looking at an NLP and those kinds of signals, largely looking at who’s going to default on certain bones.

In the commercial banking side and the investment banking side, they have a lot more money and a lot more budget, so they’re even further ahead in terms of the AI used and the techniques they use, however, in retail banking it largely is about risk tolerance and that’s a very nice way to look at fraud.