The Community for Technology Leaders
Green Image
Issue No. 06 - June (2013 vol. 25)
ISSN: 1041-4347
pp: 1227-1239
Yangqiu Song , Microsoft Research Asia, Beijing
Shimei Pan , IBM Research - T. J. Watson Center, Hawthorne
Shixia Liu , Microsoft Research Asia, Beijing
Furu Wei , Microsoft Research Asia, Beijing
Michelle X. Zhou , IBM Research - Almaden Center, San Jose
Weihong Qian , IBM Research-China, Beijing
In this paper, we propose a novel constrained coclustering method to achieve two goals. First, we combine information-theoretic coclustering and constrained clustering to improve clustering performance. Second, we adopt both supervised and unsupervised constraints to demonstrate the effectiveness of our algorithm. The unsupervised constraints are automatically derived from existing knowledge sources, thus saving the effort and cost of using manually labeled constraints. To achieve our first goal, we develop a two-sided hidden Markov random field (HMRF) model to represent both document and word constraints. We then use an alternating expectation maximization (EM) algorithm to optimize the model. We also propose two novel methods to automatically construct and incorporate document and word constraints to support unsupervised constrained clustering: 1) automatically construct document constraints based on overlapping named entities (NE) extracted by an NE extractor; 2) automatically construct word constraints based on their semantic distance inferred from WordNet. The results of our evaluation over two benchmark data sets demonstrate the superiority of our approaches against a number of existing approaches.
Clustering algorithms, Semantics, Hidden Markov models, Sparse matrices, Clustering methods, Humans, Computational modeling, text clustering, Constrained clustering, coclustering, unsupervised constraints

W. Qian, F. Wei, M. X. Zhou, S. Liu, S. Pan and Y. Song, "Constrained Text Coclustering with Supervised and Unsupervised Constraints," in IEEE Transactions on Knowledge & Data Engineering, vol. 25, no. , pp. 1227-1239, 2013.
207 ms
(Ver 3.3 (11022016))