Analyzing and Predicting Community Preference of Socially Generated Metadata: A Case Study on Comments in the Digg Community

Abstract

Large-scale socially-generated metadata is one of the key features driving the growth and success of the emerging Social Web. Recently there have been many research efforts to study the quality of this metadata that relies on quality assessments made by human experts external to a Social Web community. We are interested in studying how an online community itself perceives the relative quality of its own user-contributed content, which has important implications for the successful self-regulation and growth of the Social Web. To this end, we study the community preference for user-contributed comments on the social news aggregator Digg. In our analysis, we study several factors impacting community preference. We propose a learning-based approach for predicting the community's preference rating of unseen comments, which can be used to promote high-quality comments and filter out low-quality comments based on the community's expressed preferences.