Correcting translation assignments is time-consuming for professors, especially when dealing with large classes. Manual correction of such assignments can lead to fatigue and burnout among professors and delay in delivering feedback to students, hindering learning efficiency. Developing an artificial intelligence model can facilitate the correction process and help professors provide timely feedback to the students. Therefore, this paper presents an intelligent model applying Natural Language Processing techniques and Machine Learning algorithms to automatically distinguish between the well-translated texts and the ones with poorer quality provided by students. In this regard, the paper also introduces a real dataset to be used to train the proposed intelligent model. The dataset is composed of original texts provided in assignments, translated texts submitted by students, and associated grade for translation quality determined by human experts in the translation domain following the evaluation rubrics used to assess the quality of the translated texts by professors in the field of translation. The proposed model achieves 0.087 as Root Mean Squared Error (RMSE) value.