Web-Based Terminology Translation Mining

R. Dale et al. (Eds.): IJCNLP 2005, LNAI 3651, pp. 1004 – 1016, 2005. © Springer-Verlag Berlin Heidelberg 2005 Web-Based Terminology Translation Mining Gaolin Fang, Hao Yu, and Fumihito Nishino Fujitsu Research and Development Center, Co., LTD. Beijing 100016, China {glfang, yu, nishino}@frdc.fujitsu.com Abstract. Mining terminology translation from a large amount of Web data can be applied in many fields such as reading/writing assistant, machine translation and cross-language information retrieval. How to find more comprehensive re- sults from the Web and obtain the boundary of candidate translations, and how to remove irrelevant noises and rank the remained candidates are the challeng- ing issues. In this paper, after reviewing and analyzing all possible methods of acquiring translations, a feasible statistics-based method is proposed to mine terminology translation from the Web. In the proposed method, on the basis of an analysis of different forms of term translation distributions, character-based string frequency estimation is presented to construct term translation candidates for exploring more translations and their boundaries, and then sort-based subset deletion and mutual information methods are respectively proposed to deal with subset redundancy information and prefix/suffix redundancy information formed in the process of estimation. Extensive experiments on two test sets of 401 and 3511 English terms validate that our system has better performance. 1 Introduction The goal of Web-based terminology translation mining is to mine the translations of terminologies or proper nouns which cannot be looked up in the dictionary from the Web using a statistical method, and then construct an application system for read- ing/writing assistant (e.g. Mont Blanc�万宝龙, 白朗峰). Translators and technical researchers cannot yet obtain an accurate translation after many lookup efforts when they encounter terminology or proper noun during translating or writing foreign lan- guage. According to Web statistics by Google, 76.59% of Web pages are English. In China, statistical results by China Internet Network Information Center in July 2004 show that the number of Internet users has reached 94 million, and nearly 87.4% of users have educational backgrounds beyond high school. These users can smoothly read general English pages, but some terminologies in the Web hamper them to exactly un- derstand the whole content. Some skilled users perhaps resort to a Web search engine, but they cannot obtain effective information from a large amount of retrieved irrelevant pages and redundancy information. Thus, it is necessary to provide a system to auto- matically mine translation knowledge of terms or proper nouns using abundant Web information so as to help users accurately read or write foreign language. The system of Web-based terminology translation mining has many applications. 1) Reading/writing assistant, as one part of computer-assisted language learning (CALL) used in the E-learning. During reading or writing, users often meet terms Web-Based Terminology Translation Mining 1005 whose translations cannot be found in the dictionary, but this system can help them mine native and accurate translations from the Web. 2) The tool for constructing bi- lingual dictionary. The system can not only provide translation candidates for compil- ing bilingual lexicon, but …

Content truncated to 500 words