Holy Moses! Leveraging Existing Tools and Resources for Entity Translation

AbstractRecently, there has been an emphasis on creating shared resources for natural language processing applications. This has resulted in the development of high-quality tools and data, which can then be leveraged by the research community as components for novel systems. In this paper, we reuse an open source machine translation framework to create an Arabic-to-English entity translation system. The system first translates known entity mentions using a standard phrase-based statistical machine translation framework, which is then reused to perform name transliteration on unknown mentions. In order to transliterate names more accurately, we introduce an algorithm to augment a names database with name origin and frequency information from existing data resources. Origin information is used to learn name origin classifiers and origin-specific transliteration models, while frequency information is used to select amongst n-best transliteration candidates. This work demonstrates the feasibility and benefit of adapting such data resources and shows how off-the-shelf tools and data resources can be repurposed to rapidly create a system outside their original domain.