Abstract:

Predicting student’s learning difficulties and student’s performance is a significant research area. Because it can help teachers stop students from quitting before end-of-semester exams, predicting students' learning challenges and performance is an important research subject. The goal of this study is to forecast the learning challenges that students will face in current and upcoming courses. Some kids with learning difficulties could have trouble paying attention in class, reading, writing, or doing math. Universities are increasingly predicting student success using machine learning and big data analytics. Based on cognitive and academic data, researchers have categorized students and forecasted their future outcomes using cutting-edge statistical approaches and machine learning algorithms. In addition to classification tasks, machine learning algorithms like SVM (Support Vector Machine) can be used to forecast learning problems. It can be used to anticipate learning issues, which was done in this study to anticipate the difficulties that particular students would face. Students' learning challenges can also be predicted using the VARK analysis. These results suggest that the SVM algorithm can be an effective tool for identifying students' learning issues. However, the data's quality, feature choices, and applied settings all affect how accurate the performance is. Hence, before making any forecasts, the data must be thoroughly gathered, pre-processed, and analyzed. In order to improve student learning, academic instructors can take proactive steps by using prediction of student academic results to gain a better idea of how actively or poorly students fared in their classes. In order to predict students' final academic performance, this study used undergraduate computer science students for categorization using the SVM method and prediction using linear regression. Some of the key causes of learning difficulties were discovered

Keywords: