Werner Geyer

Dr Werner Geyer gave a guest seminar on "Fuelling the social web with recommender systems", on 24th September in UCD

ABSTRACT
The Social Web has enjoyed huge popularity in recent years, attracting millions of visitors on sites such as Facebook, Delicious, or YouTube. At the same time, enterprises are deploying social software in the workplace to support internal communication and collaboration, community building, expertise location and sharing of work-related knowledge. The Social Web provides numerous opportunities for recommender technology and, in turn, recommender technologies can play a part in fuelling the success of the Social Web phenomenon. Traditionally, recommender systems have been used for information discovery. Our research at the IBM Center for Social Software over the past couple of years has focused on applying recommendation techniques to enterprise social media sites in order to increase adoption and participation. I will provide an overview of our research projects and progress thus far and present study results from three recommender systems in more detail. The first system targets new users of a social networking site during the sign-up process in order to make the site more sticky (i.e. increase adoption), the second system suggests topics in order to inspire bloggers to write, and the third system supports personalized discovery of interesting discussions using sentiment analysis.

BIOGRAPHY
Werner Geyer is a Research Staff Member at the IBM Center for Social Software in Cambridge, MA. His areas of research are in social software and computer-supported cooperative work. After his post-doc with IBM Research in New York in 2002, he moved to Boston and started new research on activity-centric collaboration. He spent one year with Lotus Development transferring research results into product, influencing the Lotus Connections Activities product. Since January 2007, Werner has been working on beehive, IBM's internal social networking site, focusing on social software in the enterprise with a particular interest in recommender systems in social software.