Automatic Evaluation of Open-Ended Mathematics Questions with Machine Learning

This is the authors' abstract. We don't add key points for this paper.

The main purpose of this research is to develop an effective and objective scoring model for evaluating open-ended mathematics questions using machine learning algorithms predicting the academic performance of secondary school students in mathematics. Data was collected from 350 fifth-grade secondary school students, and the responses were digitized. For the classification of student answers, state-of-the-art machine learning algorithms, including Support Vector Machines (SVM), Naive Bayes (NB), Decision Tree (DT), and K-Nearest Neighbors (KNN), were employed. The study utilized TF-IDF and BoW techniques to convert the text-based answers into numerical representations. Key features were then extracted from this digitized text data using the Chi-square (Chi2) feature selection method. The classification models were evaluated using performance metrics such as accuracy, F1-score, precision, and recall. Among the algorithms tested, the SVM model combined with the TF-IDF method achieved the highest performance, with 96.57% accuracy and a 96.49% F1 score.