Extracting English-Korean Transliteration Equivalence from Domain-Specific Dictionaries

Automatic translation knowledge acquisition or automatic bilingual dictionary construction has become an important first step for natural language applications such as machine translation and cross-language information retrieval. Transliterations are used to translate proper names and technical terms especially from languages in Roman alphabets to languages in non-Roman alphabets such as from English to Korean, Japanese, and Chinese. Transliteration equivalence refers to a set composed of one foreign word and its possible transliterations. Transliterations are one of the main sources of the out-ofvocabulary (OOV) problem, because transliteration is a productive process. Many Korean domain-specific terms are composed of transliterations. Therefore, translation knowledge on transliteration equivalence is important for natural language applications to process domain-specific texts. In this paper, we propose an algorithm recognizing transliteration equivalence or transliteration pairs in domain-specific dictionaries using machine transliteration. Our method shows about 99% precision and 73% recall rate.