Machine learning model for early detection of higher education students that need additional attention in introductory programming courses
Not all higher education students who enroll in introductory pro-gramming course successfully finish it, in particular if their major is not computer programming or computer science. From empirical evidence collected at the University College Algebra we have concluded that such students can be given a better passing chance if they are identified early in the semester and given additional attention in form of extra hours in classroom, guided by mentors. The challenge lies in being able to distin-guish such students as early as possible so they can benefit from addition-al learning hours before the final exams. This paper proposes one such possible criterion: we develop a machine learning model based on previ-ous generations of students and use it for current students to calculate the probability of finishing the course unsuccessfully. Such students are then classified as »requiring extra attention«.
machine learning, model, early detection, logistic regression, higher educa-tion student, additional attention
- There are currently no refbacks.