Machine Translation and the Arabic Translation Model: The Current Scene

Abstract

Smart technology and artificial intelligence in translation has given impetus to neural machine translation. As one of the languages processed automatically, performance improved with the abundance of Arabic-language data available on the Internet.  This paper investigates the extent to which the Arabic language used on the Internet is linguistically sound, given that smart technology is fueled by the raw data that is pumped daily into forums, blogs, social networks, and websites, with all the errors it contains, which are repeated upon processing and, if not corrected, remain the default option. As such, it is essential to focus on the presence of the Arabic language on the Internet as a top priority in the short term, to preserve the linguistic integrity of published data, and promote regular scientific studies that monitor the use of the Arabic language, assess linguistic integrity, and determine the contexts and areas of machine translation use and its impact on the Arabic language on the Internet, thereby ensuring the preservation of the Arabic language and keeping pace with current technological advances.