Automatic Arabic Grading System for Short Answer Questions

The era of technology and digitalization has been advantageous to the educational sector. examination system is one most important pillars that have affected. As automatic exam grading a revolution in history development, therefore started replace traditional assessment system. allows examiners automatically assign grades for students” answers compared model answers. And, generate results based on examiners’ In this paper, we especially address short answer questions. Most research done English language. On other side, few works conducted Arabic. Moreover, Arabic considered rare resource languages. This paper aimed build an Automatic Short Answer Grading (AASAG) using semantic similarity approaches. It used measure between student answer. proposed applied scarce publicly available datasets which called (AR-ASAG). contains 2133 pairs models several versions such as txt, xml, db. efficiency was evaluated through two experiments weighting schemas local, hybrid local global schema. developed approach with weight-based LSA achieved better than (82.82%) F1-score value, 0.798 RMSE (Root-Mean-Square Error) value LSA.