While the amount of user-generated content has been skyrocketing since the advent of social media and social networks, intelligent approaches to process and make sense of these huge masses of data produced by over a billion users are rather rare so far. Hence, we solicit innovative technical papers with a focus on user-generated content and addressing problems in the fields of machine learning, multimedia, or information retrieval. Also contributions that combine two or more of these fields are highly welcome. We invite authors to submit regular technical papers of up to 8 pages as well as short position or demo papers of 2-4 pages. Regardless of their category, submissions must follow the ACM author guidelines. Paper submissions must be original and not submitted to or accepted by any other conference or journal. All submissions to this workshop will be peer-reviewed by at least three Program Committee members. The review process will be double-blind. Proceedings will be released under a Creative Commons license.

Submissions tackling, for example, one of the following challenges are highly welcome:

• Explore the usage of several types of social data, including ratings, reviews, tags, comments, hyperlinks, geo‐located data, linked data, multimedia items, and playlists.
  

• Extract from these raw data several types of knowledge (users and data): relations, user opinions, user preferences, semantic relationships/description of multimedia objects, sentiment analysis, community detection
  

• Exploit novel data mining and information retrieval techniques: expert finding, recommendation computation, similarity evaluation, network analysis, information visualization, multimedia retrieval, semantic indexing, evaluation of systems with implicit data from social media, social and human computation

• Define new information search problems: context‐dependent recommendation, definition of key users, identification of relevant locations, cross‐domain multimedia recommendation, games and multimedia, cross‐modal social content analysis
  

• Evaluate the benefits of such techniques: live users experiments, new off‐line evaluation methods