From the December 2014 issue
An Unsupervised Feature Selection Framework for Social Media Data
By Jiliang Tang and Huan Liu
The explosive usage of social media produces massive amount of unlabeled and high-dimensional data. Feature selection has been proven to be effective in dealing with high-dimensional data for efficient learning and data mining. Unsupervised feature selection remains a challenging task due to the absence of label information based on which feature relevance is often assessed. The unique characteristics of social media data further complicate the already challenging problem of unsupervised feature selection, e.g., social media data is inherently linked, which makes invalid the independent and identically distributed assumption, bringing about new challenges to unsupervised feature selection algorithms. In this paper, we investigate a novel problem of feature selection for social media data in an unsupervised scenario. In particular, we analyze the differences between social media data and traditional attribute-value data, investigate how the relations extracted from linked data can be exploited to help select relevant features, and propose a novel unsupervised feature selection framework, LUFS, for linked social media data. We systematically design and conduct systemic experiments to evaluate the proposed framework on data sets from real-world social media websites. The empirical study demonstrates the effectiveness and potential of our proposed framework.
Editorials and Announcements
- Get Your Journals as eBooks for Free
- TKDE celebrates its 25th Anniversary. Editor-in-Chief Jian Pei says, "We are celebrating the 25th Anniversary of TKDE. Since its first issue in March 1989, TKDE has published 2,981 articles, and another 220 articles in the early access portal. With 898 submissions and 79 accepted articles in 2012, TKDE is now the premier journal in the broad and general fields of data management, data mining, and knowledge engineering. We thank all the authors, reviewers, and readers for their continuing support to TKDE. As always, we are eager to hear your ideas and suggestions, and will do our best to meet your expectations. With all your passions, contributions, and supports, TKDE is embracing the new era of big data and big data analytics. Happy birthday to TKDE!"
- eBooks of issues of TKDE can now be downloaded from the Computer Society Digital Library
- Editorial: State of the Transactions (January 2014)
- Editorial (August 2013)
- New EIC Editorial (Feb 2013)
- Outgoing EIC Editorial (Feb 2013)
- State of the Journal (Feb 2012)
- EIC Editorial (January 2011)
- Special Section on the International Conference on Data Engineering (June 2014)
- Special Section on the 27th International Conference on Data Engineering (ICDE 2011)(Oct 2012)
- Special Section on Keyword Search on Structured Data (Dec 2011)
- Cloud Data Management (Sept 2011)
- Special Section on the 26th International Conference on Data Engineering (Aug 2011)
Access recently published TKDE articles
Subscribe to the RSS feed of latest TKDE content added to the digital library.
Sign up for the Transactions Connection newsletter.
IEEE Transactions on Knowledge and Data Engineering (TKDE) is an archival journal published monthly designed to inform researchers, developers, managers, strategic planners, users, and others interested in state-of-the-art and state-of-the-practice activities in the knowledge and data engineering area.
Read the full scope of TKDE