CPS Week 2018: Fraunhofer Portugal hosts Machine Learning Competition



Fraunhofer Portugal’s Machine Learning Competition will be held at CPS Week 2018, on April 10th, in Porto. The challenge? To catch a criminal by evaluating biometrical gait information retrieved from a smartphone’s inertial sensor data, while using machine learning techniques.


Porto has been stunned by a series of crimes!

For weeks the police has been puzzled. They have several suspects but no definite proof against them. However, the criminal has struck again and this time he might have made a mistake!

While robbing the Stock Exchange Palace, he left behind a smartphone. You know the smartphone's registered sensor data, that can be compared to the suspects', thus finding the real culprit. You volunteer to help, but the clock is ticking…


This is the challenge set by Fraunhofer Portugal’s Machine Learning Competition, which will be held at CPS Week 2018, on April 10th, at Porto. Fraunhofer Portugal will challenge all participants to evaluate biometrical authentication based on smartphones’ inertial sensor data, while using machine learning techniques.

To make the challenge more appealing, this problem will be presented as a crime scene investigation, where a smartphone belonging to a criminal was found as evidence and the registered sensor data needs to be compared to several suspects in custody.

The main purpose of the competition is to match test data to a single subject from the train dataset. The results will be based not only on the algorithms score given to the correct subject but also on the scores associated with the wrong people, to ensure that ambiguous results are not being generated.

While the results of this main problem will be presented in their own separate table, additional challenges will be unlocked as each team progresses. These challenges will ask the participants to detect additional traits or habits of each subject of the train dataset, such as identifying different activity intensity levels, specific occurrences like jumps, falls or moments when pairs of subjects met each other and have similar sequences of activities.

The teams that best showcase the efficacy of their biometric authentication algorithms will be named the winners.

In order to participate you must register for CPS Week 2018 before March 30th.