Dudley North visits North London: Learning When to Transliterate to Arabic

We report the results of our work on automating the transliteration decision of named entities for English to Arabic machine translation. We construct a classification-based framework to automate this decision, evaluate our classifier both in the limited news and the diverse Wikipedia domains, and achieve promising accuracy. Moreover, we demonstrate a reduction of translation error and an improvement in the performance of an English-to-Arabic machine translation system.