Using Gamification along with Recommender Models in Learning of Data Science
نویسندگان
1 Department, Faculty of Management and Accounting, Qazvin Islamic Azad University Branch, Qazvin, Iran
2 Department, Faculty of Management and Accounting, Qazvin Islamic Azad University Branch, Qazvin, Iran
3 School of Industrial Engineering, College of Engineering, University of Tehran, Tehran, Iran
doi
چکیده
Gamification is used in various fields as persuasive technology, especially in learning and education applications. Personal gamification changes or suggests the content of games and elements based on the specific characteristics of users. The purpose of this article is to create and implement a framework in personal gamification design in the field of data science learning, which uses recommender systems algorithms for the first time to improve data quality in these algorithms. This framework utilizes implicit and explicit voting in actual time and provides a dynamic and personalized environment for the enhancement quality of understanding data science learners. In this study, we developed a game environment to learn data science and its categories. Different elements of the game were considered for the challenges that existed in the process of learning. 680 students joined this system and were divided into 8 classes. After three months of users using the system, according to the collected logs and also the comments on the personalized gamification algorithm model, it was implemented by machine algorithms and suggestions were presented to the students on the site about the elements and content. With notice to root mean square error (RMSE) and mean square error (MSE) criteria, the singular value decomposition (SVD) algorithm had better results in recommender algorithms and was used in personalized gamification. The t-test and A/B test of this framework had positive effects
