The central topic of the DATA MINING CUP 2016 competition was the prediction of the return rates for a fashion distributor. Returns have been a major cost driver for online shops for many years. This is particularly the case for assortments in the fashion industry. Many approaches to solve this problem are based on forecasting models. Predicting returns was the central topic in the DATA MINING CUP of 2014. 2016 we looked at the topic in more detail focusing in particular on the influence of discounted items and vouchers on return rates.
A fashion distributor sells articles of particular sizes and colors to its customers. In some cases items are returned to the distributor for various reasons. The order data and the related return data were recorded over a two-year period. The aim is to use this data and methods of data mining to build a model which enables a good prediction of return rates.
1st place: Team 2 of University of California (Davis/USA) (EUR 2,000.00)
2nd place: Team 2 of Iowa State University (USA) (EUR 1,000.00)
3rd place: Team 2 of Technical University Darmstadt (Germany) (EUR 500.00)