Learning to Find English to Chinese Transliterations on the Web

Proceedings of the 2007 Joint Conference on Empirical Methods in Natural Language Processing and Computational Natural Language Learning, pp. 996–1004, Prague, June 2007. c⃝2007 Association for Computational Linguistics Learning to Find English to Chinese Transliterations on the Web Jian-Cheng Wu Department of Computer Science National Tsing Hua University 101, Kuangfu Road, Hsinchu, Taiwan d928322@oz.nthu.edu.tw Jason S. Chang Department of Computer Science National Tsing Hua University 101, Kuangfu Road, Hsinchu, Taiwan jschang@cs.nthu.edu.tw Abstract We present a method for learning to find English to Chinese transliterations on the Web. In our approach, proper nouns are expanded into new queries aimed at maxi- mizing the probability of retrieving trans- literations from existing search engines. The method involves learning the sublexi- cal relationships between names and their transliterations. At run-time, a given name is automatically extended into queries with relevant morphemes, and transliterations in the returned search snippets are extracted and ranked. We present a new system, TermMine, that applies the method to find transliterations of a given name. Evaluation on a list of 500 proper names shows that the method achieves high precision and re- call, and outperforms commercial machine translation systems. 1 Introduction Increasingly, short passages or web pages are be- ing translated by desktop machine translation soft- ware or are submitted to machine translation ser- vices on the Web every day. These texts usually contain some proportion of proper names (e.g., place and people names in “The cities of Mesopo- tamia prospered under Parthian and Sassanian rule.”), which may not be handled properly by a machine translation system. Online machine trans- lation services such as Google Translate1 or Yahoo! Babelfish2 typically use a bilingual dictionary that is either manually compiled or learned from a par- 1 Google Translate: translate.google.com/translate_t 2 Yahoo! Babelfish: babelfish.yahoo.com allel corpus. However, such dictionaries often have insufficient coverage of proper names and techni- cal terms, leading to poor translation performance due to out of vocabulary (OOV) problem. Handling name transliteration is also important for cross language information retrieval (CLIR) and terminology translation (Quah 2006). There are also services on the Web specifically targeting transliteration aimed at improving CLIR, including CHINET (Kwok et al. 2005) and LiveTrans (Lu, Chien, and Lee 2004). The OOV problems of machine translation (MT) or CLIR can be handled more effectively by learn- ing to find transliteration on the Web. Consider the sentence in Example (1), containing three proper names. Google Translate produces the sentence in Example (2) and leaves “Parthian” and “Sas- sanian” not translated. A good response might be a translation like Example (3) with appropriate transliterations (underlined). (1) The cities of Mesopotamia prospered under Parthian and Sassanian rule. (2) 城市繁榮下 parthian 達米亞、sassanian 統 治。 (3) 美索不達米亞3城市在巴底亞4和薩珊5統治下 繁榮起來。 These transliterations can be more effectively retrieved from mixed-code Web pages by extend- ing each of the proper names into a query. Intui- tively, by requiring one of likely transliteration morphemes (e.g., “巴”(Ba) or “帕”(Pa) for names beginning with the prefix “par-”), we can bias the search engine towards retrieving the correct trans- …

Content truncated to 500 words