27-29 November, Vilnius

Conference about Big Data, High Load, Data Science, Machine Learning & AI

Early Bird Ends In:

Day(s)

:

Hour(s)

:

Minute(s)

:

Second(s)

Santiago Cabrera-Naranjo

Teradata, Germany

Biography

Santiago is Consulting Director at Teradata Think Big Analytics. He is thereby part of the Consulting Leadership Team in Germany. Furthermore, he is responsible of Consulting Services, advising Enterprises within their big data and advanced analytics strategy; leading them to innovative best-practice implementations with the goal to speed up time-to-market regardless their technology stack and initial organization.

Furthermore, Santiago has been responsible to launch and grow Teradata’s Hub in Berlin and is active as keynote speaker. His computer science studies helped him to collect experience from different point of views across several sectors. From building up data infrastructures for Rocket-Internet Ventures to founding himself stampfy after winning the German StartupBus in 2011 at the Age of 24. Before joining Teradata, Santiago was responsible of building up the Analytical Landscape for BILD and for the adoption of new cutting-edge technologies at Axel Springer SE.

Talk

It is about Augmented Intelligence, not Artificial Intelligence

Computers alone using artificial intelligence currently could not defeat an adaptive and fast changing adversary. We should re-think AI as the effective use of information technology in augmenting human intelligence through a Human-Computer Symbiosis.

AI has encountered many fundamental obstacles, practical as well as theoretical for which augmented intelligence seems to be a more effective & logical solution. A good example of this is Fraud – Fighting fraud is ambiguous. For instance, a financial institution in Northern Europe used AI to amplify human intelligence. Some of the challenges are for example the class imbalance of 100,000:1 non-fraud vs. fraud and the fast evolution of fraud sophistication with similar intelligent software.

We will discuss how machine learning was able to reduce false positives by 35% and improve detection of true positives — actual fraud, at roughly the same percent. When the Bank added deep learning the numbers almost doubled to a 60% reduction in false positives and a 50-ish improvement in detecting actual fraud. What is the Human-Computer Symbiosis in this example and how does it work in production?