At the Annual LIACS Business Event on 18 January in the Leiden University Academy Building, the central message was how to find irregularities in Big Data, and how it can help you to track faults – and opportunities. Over 80 visitors heard how anomaly detection in data science can play a role in society.

Big Data and anomalies – Chasing the unknown

Attendees from industry and government were excited about the event, the insights they gathered from the presentations, the new contacts they were able to establish, and the expertise in data science and anomaly detection that the scientists at LIACS offered – not only from an academic perspective, but also with the application-oriented perspective and experience.

Insights

The most important insights that could be extracted from the event:

• Most business focus on the average numbers, but in stead, it is important to understand the origin of deviating numbers. Those are the anomalies in your big data.
• When you know the origin of an anomaly, you will understand whether you should prevent it from happening and consequently, act on it.

Presentations

Take another look at the presentations of our speakers.

Prof. dr. Thomas Bäck (LIACS)	Anomalies are not normal - Are they?
Dr. Sebastian Kreissl (BMW Germany)	Outlier detection in car manufacturing
Bert van Nistelrooij and Hans Paalvast (Zorginstituut Nederland)	Optimizing health services - Defining "Proper Care"
Dr. Wojtek Kowalczyk (LIACS and National Fraud Detection Center)	Fighting Fraud in financial services
Mark van den Akker (CEO Supergraph)	From Standard Deviation to Decision
Dr. Cor Veenman (LIACS and National Forensic Institute)	How to identify criminal behaviour - "Whodunit with data science"