Predicting Probabilities of Getting a High Grade in Mathematics: Probability Analysis Based on Previous Scores using Machine Learning Techniques

This paper examines the possibilities of predicting high mathematics scores by analyzing likelihood and probability, based on students' previous scores, and using advanced machine learning techniques to optimize predictions. The main purpose of the study is to create a model that can predict the likelihood that a student will achieve a high grade on future math tests, based on their performance history and other influencing factors, such as frequency of preparation, grade point average, class participation, as well as socio-economic data. The methodology used in this study is the quantitative approach with a predictive and analytical study design. In this research, data from previous test results are analyzed and statistical models and machine learning algorithms are used, such as logistic regression, random forests, and neural networks, to build a predictive model. To train and test the model, the data is divided into a training group and a test group, while the efficiency of the model is evaluated using the measures of accuracy, positive precision (precision), care (recall), and f1-score. This process helps identify the factors that most influence math outcomes and provides important insights for educators on ways to predict student success. The results show that machine learning models, based on previous performance data, have a high accuracy in predicting high grades and can be used as support tools to personalize teaching. By providing a more accurate estimate of the probability of success, this study contributes to the field of mathematics education and creates a basis for further development of methods for adapting teaching content and support to students to improve mathematics results.