Feature selection is widely used in preparing highdimensional data for effective data mining. Increasingly popular social media data presents new challenges to feature selection. Social media data consists of (1) traditional high-dimensional, attribute-value data such as posts, tweets, comments, and images, and (2) linked data that describes the relationships between social media users as well as who post the posts, etc. The nature of social media also determines that its data is massive, noisy, and incomplete, which exacerbates the already challenging problem of feature selection. In this paper, we illustrate the differences between attributevalue data and social media data, investigate if linked data can be exploited in a new feature selection framework by taking advantage of social science theories, extensively evaluate the effects of user-user and user-post relationships manifested in linked data on feature selection, and discuss some research issues for future work. 1

Abstract. Community detection is an unsupervised learning task that discovers groups such that group members share more similarities or interact more frequently among themselves than with people outside groups. In social media, link information can reveal heterogeneous relationships of various strengths, but often can be noisy. Since different sources of data in social media can provide complementary information, e.g., bookmarking and tagging data indicates user interests, frequency of commenting suggests the strength of ties, etc., we propose to integrate social media data of multiple types for improving the performance of community detection. We present a joint optimization framework to integrate multiple data sources for community detection. Empirical evaluationonbothsyntheticdataandreal-worldsocialmediadatashowssignificant performance improvement of the proposed approach. This work elaborates the need for and challenges of multi-source integration of heterogeneous data types, and provides a principled way of multi-source community detection. Keywords: CommunityDetection,Multi-sourceIntegration,SocialMediaData 1