CLGVSM: Adapting Generalized Vector Space Model to Cross-lingual Document Clustering

Proceedings of the 5th International Joint Conference on Natural Language Processing, pages 580–588, Chiang Mai, Thailand, November 8 – 13, 2011. c⃝2011 AFNLP CLGVSM: Adapting Generalized Vector Space Model to Cross- lingual Document Clustering Guoyu Tang1,2, Yunqing Xia1, Min Zhang2, Haizhou Li2,Fang Zheng1 1Dept. of Comp. Sci. & Tech., Tsinghua University, Beijing 100084, China sweetyuer@gmail.com,{yqxia,fzheng}@tsinghua.edu.cn 2Institute for Infocomm Research, A-STAR, Singapore {mzhang, hli}@i2r.a-star.edu.sg Abstract Cross-lingual document clustering (CLDC) is the task to automatically or- ganize a large collection of cross-lingual documents into groups considering con- tent or topic. Different from the tradition- al hard matching strategy, this paper ex- tends traditional generalized vector space model (GVSM) to handle cross-lingual cases, referred to as CLGVSM, by incor- porating cross-lingual word similarity measures. With this model, we further compare different word similarity measures in cross-lingual document clus- tering. To select cross-lingual features ef- fectively, we also propose a soft- matching based feature selection method in CLGVSM. Experimental results on benchmarking data set show that (1) the proposed CLGVSM is very effective for cross-document clustering, outperform- ing the two strong baselines vector space model (VSM) and latent semantic analy- sis (LSA) significantly; and (2) the new feature selection method can further im- prove CLGVSM. 1 Introduction The globalization of business environment urges organizations to maintain documents in different language. Obviously, organizations and research communities nowadays encounter the challenge of cross-lingual document clustering (CLDC). Document clustering seeks to automatically or- ganize a large collection of documents into groups of similar documents. Various document clustering technologies have been proposed to deal with monolingual documents. The classical solution to monolingual docu- ment clustering is vector space model (VSM), which explores bag of words (BOW) to construct feature space. Each document is converted to a VSM vector. Serious problem occurs when words are matched to the features using the hard matching strategy. For example, when the word coast is selected as a feature, word seashore will not contribute to hard matching unless it is also selected as a feature. Different semantic document representation models have been proposed to address the short- coming of VSM. Some semantic document rep- resenting models such as Latent Semantic Analy- sis (LSA) (Landauer et al., 1998) and Latent Di- richlet Allocation (Blei et al., 2003) implicitly capture statistical semantics by mapping docu- ments to a lower dimension space. Other models such as Generalized Vector Space Model (GVSM) (Wang et al., 1985) extract statistical semantics in an explicit way by directly estimat- ing measures of correlations between words. The above models are designed for monolin- gual document sets and cannot be applied to cross-lingual scenario unless a bridge is created to connect cross-lingual features. Carbonell et al. (1997) use semantic models in parallel corpus. As features are selected from a common paral- lel/comparable corpus, which is usually different from the test cross-lingual document, over-fitting problem inevitably happens. Other researchers propose to translate features or documents with bilingual dictionary or ma- chine translation tools (Mathieu et al., 2004).However, ambiguity happens constantly and it is …

Content truncated to 500 words