Machine Learning Models for Basic Skills Identification in University Students

After the pandemic, remote teaching became a challenge, accentuating inequalities and affecting student performance. Thus, this work addresses the use of Machine Learning techniques to assistant the teaching and learning process. The focus is on identifying weaknesses in the basic skills of university students and recommending content for improvement, helping to cover learning gaps and support the educational community. Machine Learning techniques were used to predict the student performance in basic skills, from a previous mapping of students' psychometric profile. The student performance in this work was based on autoperception and, among used machine learning techiques, Neural Networks and Support Vector Machine demonstrated better accuracy in classifying student performance. The results achieved reinforce the importance of innovative approaches in education, especially in times of distance learning and post-pandemic challenges, allowing for a significant contribution to academic success.