Enhancing the Assessment of Narrative Answers Through Ai-Driven Machine Learning Mode

The examination system handles evaluation, which is a crucial component of education. Assessing a large number of students requires a substantial amount of physical work. Not only is manual valuation a labor-intensive operation, but its quality also varies according to the examiner's temperament. If this could be controlled by a machine, a lot of the problems described above would be fixed in the contemporary world. Therefore, one approach to finding a solution is to use computers to evaluate replies. Computers still struggle to assess descriptive responses, though. As a result, it's critical to research and put automated declarative answer assessment methods into practice. This paper suggests a classifier-based machine learning (ML) method for assessing descriptive answers. at order to build the required, we experiment at our academic institution. 311 training samples and 78 test samples make up the dataset. When applied to logistic regression, J48, and Naive Bayes classifiers, it is shown that, on average, 75.049 % of the responses are scored properly.