论文信息 - Cross-dataset Clustering: Revealing Corresponding Themes across Multiple Corpora

Cross-dataset Clustering: Revealing Corresponding Themes across Multiple Corpora

We present a method for identifying corresponding themes across several corpora that are focused on related, but distinct, domains. This task is approached through simultaneous clustering of keyword sets extracted from the analyzed corpora. Our algorithm extends the information-bottleneck soft clustering method for a suitable setting consisting of several datasets. Experimentation with topical corpora reveals similar aspects of three distinct religions. The evaluation is by way of comparison to clusters constructed manually by an expert.

Ido Dagan | Eli Shamir | Zvika Marx

[1] Douglas R. Hofstadter,et al. Fluid Concepts and Creative Analogies , 1995 .

[2] H. Sebastian Seung,et al. Learning the parts of objects by non-negative matrix factorization , 1999, Nature.

[3] Margaret A. Boden,et al. Douglas Hofstadter and the Fluid Analogies Research Group, Fluid Concepts and Creative Analogies: Computer Models of the Fundamental Mechanisms of Thought , 2004, Minds and Machines.

[4] Dedre Gentner,et al. Structure-Mapping: A Theoretical Framework for Analogy , 1983, Cogn. Sci..

[5] Thomas M. Cover,et al. Elements of Information Theory , 2005 .

[6] Ido Dagan,et al. Cross-component Clustering for Template Induction , 2002 .

[7] Joachim M. Buhmann,et al. Coupled Clustering: A Method for Detecting Structural Correspondence , 2001, J. Mach. Learn. Res..

[8] Naftali Tishby,et al. The information bottleneck method , 2000, ArXiv.

[9] Naftali Tishby,et al. Distributional Clustering of English Words , 1993, ACL.

[10] Inderjit S. Dhillon,et al. Concept Decompositions for Large Sparse Text Data Using Clustering , 2004, Machine Learning.