Predicting academic achievement using educational data mining and machine learning techniques

Machine learning is a field of artificial intelligence that enables a computer to learn to analyze data, build models and make decisions. One of the areas where machine learning, which is widely used in many fields, is widely used is education. There are many benefits to using machine learning in education. Machine learning can offer personalized learning experiences tailored to students' learning needs. It can analyze many data such as student performance, teacher feedback and other metrics. Analysis of this data allows teachers to understand how well students learn and to develop learning strategies that help students learn better. By analyzing students' achievement levels, it can guide teachers about what students understand and what they don't. It automates many processes to monitor student performance and help teachers evaluate. In this way, cost savings can be achieved in education. It can also help teachers create better lesson plans and provide students with more effective learning materials. For these reasons, machine learning is widely used in education and is expected to become more widespread in the future. The aim of this thesis is to predict the academic success of students and to determine the parameters affecting the success by using different machine learning algorithms on the online open course dataset. Massive online open course dataset was used in the study. By making feature selection on the data set, the variables that have the most impact on academic achievement were determined and these features were used in modeling. In order to classify the target quality; 7 machine learning algorithms, namely Logistic Regression, Linear Discriminant Analysis, Random Forest, K-Nearest Neighbor, Decision Trees, Support Vector Machines and Naive Bayes, were applied and the success performances of the models were compared. According to the results of the model performance evaluation, the most successful model was determined as the Random Forest algorithm. The final score of the students has been determined as the attribute that affects the success the most.