Summarizing Encyclopedic Term Descriptions on the Web

Summarizing Encyclopedic Term Descriptions on the Web Atsushi Fujii and Tetsuya Ishikawa Graduate School of Library, Information and Media Studies University of Tsukuba 1-2 Kasuga, Tsukuba, 305-8550, Japan {fujii,ishikawa}@slis.tsukuba.ac.jp Abstract We are developing an automatic method to compile an encyclopedic corpus from the Web. In our previous work, paragraph-style descriptions for a term are extracted from Web pages and organized based on domains. However, these descriptions are independent and do not comprise a condensed text as in hand-crafted encyclopedias. To resolve this problem, we propose a summarization method, which produces a single text from multiple descriptions. The resultant sum- mary concisely describes a term from dif- ferent viewpoints. We also show the effec- tiveness of our method by means of experi- ments. 1 Introduction Term descriptions, which have been carefully orga- nized in hand-crafted encyclopedias, are valuable linguistic knowledge for human usage and compu- tational linguistics research. However, due to the limitation of manual compilation, existing encyclo- pedias often lack new terms and new definitions for existing terms. The World Wide Web (the Web), which contains an enormous volume of up-to-date information, is a promising source to obtain new term descriptions. It has become fairly common to consult the Web for descriptions of a specific term. However, the use of existing search engines is associated with the follow- ing problems: (a) search engines often retrieve extraneous pages not describing a submitted term, (b) even if desired pages are retrieved, a user has to identify page fragments describing the term, (c) word senses are not distinguished for polyse- mous terms, such as “hub (device and center)”, (d) descriptions in multiple pages are independent and do not comprise a condensed and coherent text as in existing encyclopedias. The authors of this paper have been resolving these problems progressively. For problems (a) and (b), Fujii and Ishikawa (2000) proposed an auto- matic method to extract term descriptions from the Web. For problem (c), Fujii and Ishikawa (2001) improved the previous method, so that the multiple descriptions extracted for a single term are catego- rized into domains and consequently word senses are distinguished. Using these methods, we have compiled an ency- clopedic corpus for approximately 600,000 Japanese terms. We have also built a Web site called “Cy- clone”1 to utilize this corpus, in which one or more paragraph-style descriptions extracted from differ- ent pages can be retrieved in response to a user input. In Figure 1, three paragraphs describing “XML” are presented with the titles of their source pages. However, the above-mentioned problem (d) re- mains unresolved and this is exactly what we intend to address in this paper. In hand-crafted encyclopedias, a single term is de- scribed concisely from different “viewpoints”, such as the definition, exemplification, and purpose. In contrast, if the first paragraph in Figure 1 is not described from a sufficient number of viewpoints for XML, a user has to read remaining paragraphs. However, this is inefficient, because the descriptions are extracted from independent pages and usually include redundant contents. To …

Content truncated to 500 words