Many small and mid-sized e-commerce businesses use services from recommender system (RS) providers to outsource the implementation and operation of their RS. RS providers are recording users’ activities, computing statistical models, and delivering recommendations on demand.
Due to the nature of the collected data, which is usually implicit, the task of the RS provider is to deliver top-N recommendations and not really to predict what will be the rating that a user will give an item. YOOCHOOSE GmbH is such a RS provider that is specialized in calculating best matching top-N recommendations on a user base for different use cases like generating cross- or up-sell, exploit long tail and last but not least to keep the user entertained.
In this challenge, YOOCHOOSE is providing a collection of sequences of click events; click sessions. For some of the sessions, there are also buying events. The goal is hence to predict whether the user (a session) is going to buy something or not, and if he is buying, what would be the items he is going to buy. Such an information is of high value to an e-business as it can indicate not only what items to suggest to the user but also how it can encourage the user to become a buyer. For instance to provide the user some dedicated promotions, discounts etc’. The data represents six months of activities of a big e-commerce businesses in Europe selling all kinds of stuff such as garden tools, toys, clothes, electronics and much more.
A more detailed description of the challenge can be found in the Challenge section.
- Start of Challenge: November 5 2014
|Papers||July 7||August 1||August 10|
- Winners announced: RecSys 2015 (September 16)
The RecSys challenge workshop will be held during the ACM RecSys 2015 at Vienna, Austria, 16-20 September 2015. Participants are expected to submit a workshop paper (4 pages – template available here) detailing their results and findings. The winners of the challenge will present the solution in the main track of the conference. See the section about paper submission for more details.