EPFL Master students win Data Mining Cup competition
The Data Mining Cup (DMC) is an international competition organized each year by prudsys AG, a leading data-mining company. The aim of the DMC is “to enthuse domestic and international students for intelligent data analysis (data mining) and challenge them to find the best solution to a data mining problem in competition with others.” The top ten teams are invited to the awards ceremony in Berlin, with travel and accommodations expenses paid, and the top three teams win cash awards of €2,000, €1,000, and €500.
This year, the DMC’s challenge was “retail meets machine learning”; the task was to create a sales forecast for an online shop that adapts its product prices dynamically and automatically. The competition participants had to develop a mathematical model that could predict sales by using anonymised historical transaction data from a real mail-order pharmacy.
202 student teams from 150 universities in 48 countries participated in the competition, which ended on 17 May 2017. EPFL was represented by two teams, both from Mathematics.
One of these teams, consisting of five Master students, Loris Michel, Melvin Kianmanesh Rad, Lavinia Ghita, Xavier Bays, and Brice Repond, won first place in the competition after successfully developed a data-mining model that could be implemented to meet the online store’s needs. These students had already taken part in the 2015 competition, when two EPFL Mathematics teams came in the 4th and 8th places.
The second place in 2017 went to a team from ETHZ, led by Alain-Sam Cohen, a former EPFL student in Mathematics, and a team from Southern University (Nanjing, China) won third place.
The winning teams travelled to Berlin for the award ceremony, which took place during a conference on 27 and 28 June.
Another team of EPFL Math students came in 13th place. The team was a mix of third-year and Master students: Yassine Benyahia, Aldo Podesta, Wallyson Lemes de Oliveira, Hamza Sayah, Antoine Launay, and Francesco Palma.
DMC announcement: http://www.data-mining-cup.de/en/wettbewerb/preistraeger.htm