Modern Fraud Prevention using Deep Learning

A session at GOTO Copenhagen 2015

  • Phil Winder

Tuesday 6th October, 2015

2:30pm to 3:20pm (CET)

In any language and all cultures, financial gain is the aim of most people. Unfortunately, many people (and companies!) resort to techniques that lie somewhere between immoral and criminal. Currently, many companies rely on a combination of external scoring services and internal manual fraud prevention procedures. This presentation will introduce ideas that are able to automate the detection of fraud and automate the analysis of new unknown fraud patterns.

Fraud detection is very specific to a particular domain. For example, credit card fraud is completely different to benefit fraud. This means that solutions must be bespoke and must fulfill the company's specific requirements. Mortgage fraud will be used as an example throughout this talk, although the techniques are applicable to any domain. It will discuss using software, databases and cluster computing frameworks to automate detection. It will also introduce the field of Machine Learning, with a particular focus on Deep Learning. Modern deep learning algorithms allow machines to analyze huge amounts of data to automate the discovery of new fraud vectors.

If you are thinking about implementing a fraud detection system for your business, and are interested in the cutting-edge of fraud detection technology, then this is the talk for you. Alternatively, feel free to visit me at the Trifork stand to talk about your specific domain. If you don't know what Machine Learning is and would like a buzzword-free introduction and example, this is also the talk for you.

About the speaker

This person is speaking at this event.
Phil Winder

Engineer at Trifork Leeds

Sign in to add slides, notes or videos to this session

Tell your friends!


Time 2:30pm3:20pm CET

Date Tue 6th October 2015

Short URL


View the schedule


See something wrong?

Report an issue with this session