Recognition and translation Arabic-French of Named Entities: case of the Sport places

The recognition of Arabic Named Entities (NE) is a problem in different domains of Natural Language Processing (NLP) like automatic translation. Indeed, NE translation allows the access to multilingual information. This translation doesn’t always lead to expected result especially when NE contains a person name. For this reason and in order to ameliorate translation, we can transliterate some part of NE. In this context, we propose a method that integrates translation and transliteration together. We used the linguistic NooJ platform that is based on local grammars and transducers. In this paper, we focus on sport domain. We will firstly suggest a refinement of the typological model presented at the MUC Conferences we will describe the integration of an Arabic transliteration module into translation system. Finally, we will detail our method and give the results of the evaluation.