Predicting students academic performance at the end of the semester by using machine learning algorithms with the data in moodle system

The transformative power of technology in measuring and assessing academic performance is particularly striking when analysing the impact of technology in education. In this area, artificial intelligence and machine learning algorithms stand out for their ability to predict students' future academic success. By subjecting students' past performance, learning processes and various factors to big data analysis, it enables personalised educational approaches and improves the strategic planning capabilities of educational institutions. In this study, the data obtained from the MOODLE learning management system were analysed using five different machine learning algorithms (Random Forest, Decision Trees, K-Nearest Neighbors, Support Vector Machines and Naive Bayes). These analyses, conducted with binary and multiple classification techniques, revealed the high success of the algorithms in predicting student achievement. The study shows the potential of machine learning algorithms in education and how effective they can be in critical processes such as performance prediction. Enriched with a literature review, this thesis provides a detailed analysis of similar studies and evaluates the applicability of existing algorithms on training data. The processing of the MOODLE dataset includes pre-processing steps such as data cleaning, feature selection and normalisation. The performances of the algorithms are evaluated by measures such as accuracy, sensitivity, specificity and f1 score, and the best performing methods are identified. In conclusion, in addition to making a significant contribution to the educational technology literature, this study also sheds light on the processes of educators and administrators in monitoring and guiding student achievement. The effective use of machine learning algorithms can enable strategic decisions to be made to improve the quality of education and maximise student achievement.