Assessing Malaysian University English Test (MUET) Essay on Language and Semantic Features Using Intelligent Essay Grader (IEG)

Automated Essay Scoring (AES) refers to the Artificial Intelligence (AI) application with “intelligence” in assessing and scoring essays. There are several well-known commercial AES adopted by western countries, as well many research works conducted investigating automated essay scoring. However, most of products not related Malaysian English test context. The tend score essays based on rubrics a particular text context (e.g., TOEFL, GMAT) employing their proprietary algorithm that is accessible users. In Malaysia, development scarce. This paper intends formulate Malaysia-based AES, namely Intelligent Grader (IEG), for environment using our collection two University Test (MUET) dataset. We proposed rubric its language semantic features. analyzed correlation features grade Pearson Correlation Coefficient. Furthermore, we constructed an model predict grades. result, found featured such vocabulary count advanced part speech were highly correlated grades, showed greater influence grades than From prediction model, observed yielded better accuracy results selected high-correlated features, followed