A recommendation system to support the students performance in programming contests

Learning to program computers and other devices are becoming an essential skill for students of high school and college. With this knowledge, young people can master the tools and technology that are part of everyday life. However, the traditional teaching, even in computer science courses, needs to break the barrier and use new approaches to motivate students to learn programming in an effective way. Participation in programming contests can promote student engagement. Although there are systems that support training for programming contests, they don´t still promote student engagement and neither personalized and effective learning. This article aims to propose a recommendation system to support the student´s performance in programming contests, with the goal of providing personalized learning. Problems are recommended according to the profile and student´s skills. Therefore, we used ephemeral personalization, pull-delivery recommendation and content-based filtering approaches. Usability tests were conducted to validate the system. Finally, we present an application scenario where this system can be used how a pedagogical strategy for teaching programming.