The goal of this workshop is to study the application of machine learning and AI approaches and algorithms to social media, big data and the web. We particularly welcome submissions that go beyond the “simple” computational approaches and try to discover the “hidden” knowledge/insights, evaluate how “good” they are, and how they can be validated for accuracy and reality. Hence, the workshop aims to attract and discuss various novel aspects of knowledge learning, recommendation, community discovery, social influence and prediction from social media, big data and the web. In short, the workshop invites topics that deal with user activities and social predictable behavior that is inferred from analysis and mining of the social media, big data, or web using suitable machine learning and AI methods. Thus, our goal is to bring together researchers and practitioners from around the world in the machine learning, AI, natural language processing, user analysis, big data and recommendation communities interested in 1) exploring different perspectives and approaches to mine hidden behavioral aspects of (complex) social media data, web data and big data, 2) inferring user and social influence, hidden activities and recommendation and 3) building models and frameworks for evaluating the designed approaches.

In our first workshop on Modeling Social Media (MSM 2010 at ACM HT in Toronto Canada), we explored various models of social media ranging from user modeling, hypertext models, software engineering models, sociological models and framework models. In our second workshop (MSM 2011 at IEEE SOCIALCOM in Boston, USA), we addressed the user interface aspects of modeling social media. In our third workshop (MSM 2012 at ACM HT in Milwaukee, USA), we looked at the collective intelligence in social media, i.e. making sense of the content and context from social media websites such as Facebook, Twitter, Google+ and Foursquare by analyzing tweets, tags, blog posts, likes, posts and check-ins, in order to create a new knowledge and semantic meaning. In our fourth workshop (MSM 2013 at ACM WWW in Paris, France), we discussed mining, modeling and recommending “things” in social media. In our fifth and sixth workshops (MSM 2014 at ACM WWW in Seoul, Korea and MSM 2015 at ACM WWW in Florence, Italy), we focused on mining big data on social media and the web. In the last workshop (MSM 2016 at ACM WWW in Montreal, Canada), we attracted worldwide researchers’ attention to the field of behavioral analytics using web and social media data. For this year’s workshop, we aim to attract researchers from all over the world working on Machine Learning and AI models for social media data analytics and predictive insights. Social networks such as Facebook, Twitter, and LinkedIn have paved the way for generating huge amount of diverse data in a short period of time. Such social media data require the application of big data analytics to produce meaningful information to both information consumers and data generators. Machine learning and AI techniques are particularly effective in situations where deep and predictive insights need to be uncovered from such social media data sets that are large, diverse and fast changing. Following the discussion at our workshop at WWW2016, we aim to focus on machine learning and AI driven data analytics and predictive modeling on social media and the web. Contrary to last year’s workshop, we would like to particularly invite researchers that are interested in going beyond standard analytics approaches and discovering the knowledge/insights hidden in the large and fast-changing social media data.