A Muitilingual News Summarizer

A Multilingual News Summarizer Hsin-Hsi Chen Department of Computer Science and Information Engineering National Taiwan University Taipei, TAIWAN, R.O.C. hh_chen@csie.ntu.edu.tw Chuan-Jie Lin Department of Computer Science and Information Engineering National Taiwan University Taipei, TAIWAN, R.O.C. cjlin@nlg2.csie.ntu.edu.tw Abstract Huge multilingual news articles are reported and disseminated on the Internet. How to extract the key information and save the reading time is a crucial issue. This paper proposes architecture of multilingual news summarizer, including monolingual and multilingual clustering, similarity measure among meaningful units, and presentation of summarization results. Translation among news stories, idiosyncrasy among languages, implicit information, and user preference are addressed. Introduction Today many web sites on the Internet provide online news services. Multilingual news articles are reported periodically, and across geographic barrier to disseminate to readers. Readers can access the news stories conveniently, but it takes much time for people to read all the news. This paper will present a personal news secretariat that helps on-line readers absorb news information from multiple sources in different languages. Such a news secretariat eliminates the redundant information in the news articles, reorganizes the news for readers, and helps them resolve the language barriers. Reorganization of news is some sort of document summarization, which creates a short version of original document. Recently, many papers touch on single document summarization (Hovy and Marcu, 1998a). Only a few touch on multiple document summarization (Chen and Huang, 1999; Mani and Bloedorn, 1997; Radev and McKeown, 1998) and multilingual document summarization (Hovy and Marcu, 1998b). For multilingual multiple news summarization, several issues have to be addressed: (1) Translation among news stories in different languages The basic idea in multiple document summarizations is to identify which parts of news articles present similar reports. Because the news stories are in different languages, some kind of translation is required, e.g., term translation. Besides the problem of translation ambiguity, different news sites often use different names to refer the same entity. The translation of named entities, which are usually unknown words, is another problem. (2) Idiosyncrasy among languages Different languages have their own specific features. For example, a Chinese sentence is composed of characters without word boundary. Word segmentation is indispensable for Chinese. Besides, Chinese writers often assign punctuation marks at random, how to determine a meaningful unit for similarity checking is a crucial issue. Thus some tasks may be done for specific languages during summarization. (3) Implicit information in news reports Some information is implicit in news stories. For example, the name of a country is usually not mentioned in a news article reporting an event that happened in that country. On the contrary, the country name is important in foreign news. Besides, time zone is used to specify date/time implicitly in the news. (4) User preference When users want to read documents in their familiar languages, news fragments in some Figure 1. Architecture of Our Multilingual Summarization System languages are preferred to those in other languages. Even machine translation should be introduced to translate news fragments. Besides, if a user …

Content truncated to 500 words