Discovery Challenge Workshop and Challenge Organization Workshop Chairs Challenge Organizing Committee Programm Committee

This year’s Discovery Challenge was dedicated to solving of the video lecture recommendation problems, based on the data collected at VideoLectures.Net site. Challenge had two tasks: task 1 in which new-user/newitem recommendation problem was simulated, and the task 2 which was a simulation of the clickstream-based recommendation. In this overview we present challenge datasets, tasks, evaluation measure and we analyze solutions and results. 1 General description of the challenge VideoLectures.Net (VL.Net) is a free and open access multimedia repository of video lectures, mainly of research and educational character. The lectures are given by distinguished scholars and scientists at the most important and prominent events like conferences, summer schools, workshops and science promotional events from many fields of science. The website is aimed at promoting science, exchanging ideas and fostering knowledge sharing by providing high quality didactic contents not only to the scientific community but also to the general public. All lectures, accompanying documents, information and links are systematically selected and classified through the editorial process taking into account also users’ comments. This challenge was organized through the support of the EU project e-LICO. The aims of the challenge are multifold: from research in recommender systems, improvement of the current recommender system of the VL.Net site, to provision of the problem and datasets to the research community. The challenge consisted of two main tasks. Due to the nature of the problem, each of the tasks has its own merit: task 1 simulates new-user and new-item recommendation cold-start mode, while task 2 simulates clickstream (implicit preference) based recommendation. Due to the privacy preserving constraints, data from VL.Net website includes neither explicit nor implicit user profiles. Instead, implicit profiles embodied in viewing sequences (clickstreams) have been transformed, so that no individual viewing sequence information can be revealed or reconstructed. This transformed, viewing related data includes: i) lecture co-viewing frequencies ii) pooled viewing sequences (whose construction will be described later) and is accompanied with rich lecture description related information available: lecture category taxonomy, lecture names, descriptions and slide titles (where available), authors, institutions, lecture events and timestamps. Unlike most of the other publicly available recommendation problems datasets, this dataset contains original content, names and taxonomy. The dataset of the challenge (including 4 http://videolectures.net 5 http://www.e-lico.eu 6 http://lis.irb.hr/challenge/