Accuracy Analysis of Artificial Intelligence in Arabic Language Translation and Grammatical Rule Mapping

Abstract

The development of Artificial Intelligence (AI) has been used in various applications that are able to translate Arabic and map its grammatical rules automatically. In the field of linguistics, especially Arabic, AI has the potential to help and speed up the translation process as well as complex grammar analysis. However, some linguistic aspects, such as sentence context, nuances of meaning, and i’rab interpretation, are still a challenge for AI systems. This article presents the analysis of AI's ability to punctuate, translate and map Arabic rules, and assess the extent to which it can replace humans. The research method used was a literature study with a content analysis approach. The objects studied were Arabic mahfudzhot selected based on the objectives with the book to analyze the AI with Kitab Nahwu Wadhih. This study found that the three AIs analyzed have their own advantages and disadvantages. ChatGPT has the largest percentage score in punctuation (harakat) and Arabic rule mapping (i’rab). As for the translation, Gemini has a superior percentage score compared to ChatGPT and Perplexity. Although AI is able to perform literal translation and structural analysis, the shortcomings were frequently showed as its results, so a human is needed to ensure accuracy and contextualized interpretation. These results showed that AI can serve as an effective tool, but cannot fully replace human expertise in Arabic linguistics.

Keywords

• arabic language rules
• artificial intelligence
• linguistics
• translation