Data Science Game 2016
A world-class competition
For the second year in a row the Data Science Game, a data science and machine learning international student contest, was held in Paris, France in September, 2016. 143 teams, three times as much as in 2015, representing more than 50 universities from 28 different countries (view the map), faced real-life business challenges. To stand out in this competition, students were asked to conceive and implement predictive models in order to solve Big Data-related issues.
A thrilling qualifier on computer vision and renewable energy
The 143 teams were first cut down to 20 through an online qualification phase which lasted for about a month. During this stage, candidates were given a challenge based on solar energy production optimization.
In order to map such production potential in France, the Data Science Game partnered with Etalab - the French public agency in charge of Open Data and Data use in the French administration. The State Agency created the OpenSolarMap project which provides satellite images of about 80,000 building roofs. Automated classification of roof orientation is a true challenge for Etalab.
For the 2016 Data Science Game participants, the challenge was to develop an algorithm which would recognise the orientation of a roof based on a satellite photograph by building on more than 10,000 roof photographs categorized using crowdsourcing. The majority of the top 40 teams used Deep Learning methods, which have proved to be particularly efficient on Computer Vision issues and in the context of Big Data.
A unique setting for the Data Science world finals
The final phase was held in the Château Les Fontaines in Paris area, France. The gathering saw 80 students competing in a fierce hackathon to build the best predictive algorithm, using Machine Learning models. This time, the challenge dataset contained requests for automobile insurance quotes received by AXA, the 1st international insurance brand, from different brokers and comparison websites.
The participants were asked to predict whether the person who requested a given quote bought the associated insurance policy. The mammoth 30-hour hackathon fostered the creation of some of the most innovative and refined solutions to this insurance-related problem.
The Moscow Institute of Physics and Technology (Russia), Cambridge University (United Kingdom) and Skoltech University (Russia) won the three first prizes, while Université Pierre et Marie Curie (France) and University of Padova (Italy) ranked 4th and 5th, respectively.
|Prize Capgemini, for 1st place||Russian Data Mafia||Moscow Institute of Physics and Technology||Russia|
|Prize AXA, for 2nd place||Cantab||Cambridge University||United Kingdom|
|Prize Microsoft, for 3rd place||We just want our name to be the longest one||Skoltech University||Russia|
|Prize Milliman, for 4th place||Jonquille||Université Pierre et Marie Curie||France|
|Prize Numberly, for 5th place||Brosio2BeWild||University of Padova||Italy|
|Prize QuantCube Technology, for Jury’s innovation prize||ml_noobs||Moscow State University||Russia|
Shortlisted countries included: France, the Netherlands, Russia, Germany, the UK, Singapore, USA, Japan, India, and Italy.
Last year’s edition saw an amazingly high level of competition. We are looking forward to seeing even more highly-skilled students in Data Science competing this year. Stay tuned for more information on the 2017 qualifier phase!