A Knowledge Engineering-Based Approach to Detect Gaming the System in Novice Programmers

• Hemilis Joyse Barbosa Rocha UFPE / IFAL
• Evandro de Barros Costa UFAL
• Patricia Cabral de Azevedo Restelli Tedesco UFPE

Resumo

In recent years, the behavior called “gaming the system” has received increasing attention, especially in online learning environments, where students seek to advance in the system without actually engaging in genuine learning processes. Although previous studies have made progress in identifying and understanding this behavior, making use of machine learning models and knowledge engineering, previously validated detectors have not fully captured the gaming attitudes observed in our novice programming learning context. This challenge may be related to the transferability of motivations for gaming between different contexts, as well as to detection. Therefore, this article adopts an approach focused on the exploration and specific detection of “gaming the system” attitudes manifested by novice programmers in interaction with a learning environment. We propose an approach to detecting this behavior, combining knowledge engineering and machine learning algorithms. We describe the procedures for generating our database, followed by the elaboration of a model based on knowledge engineering and the development of a model based on machine learning. As a result, we identified two new game attitudes and incorporated them into the knowledge-based model, using cognitive task analysis to understand how experts encode these attitudes.