Game-based learning in scientific literature: text mining analysis

Game-based learning has increased its influence in different fields (e.g., education or marketing) in recent years. However, it is difficult to have an overall landscape of the scientific approach that has been carried out to date around this term. The objective of this study is to identify, analyze, and provide a comprehensive map of the emerging concepts published in the scientific literature related to game-based learning, as well as to establish the most studied topics around these concepts. Method: a search was conducted in the Web of Science (WOS) database for all articles related to this topic, and they were analyzed using both the title and keywords, as well as abstracts, through data mining techniques. The frequency of occurrence of unigrams, bigrams, and trigrams was calculated, as well as the investigated topics within the subject through latent Dirichlet analysis. Results: some of the most frequent unigrams, bigrams, and trigrams were: 'student', 'education', 'virtual reality', 'video game', 'learning environment', 'autism spectrum disorder', 'design methodology approach', and 'information communication technology'. From the model based on latent Dirichlet analysis, 15 research topics related to game-based pedagogy emerged, focusing on new tools for implementing gamification in different contexts (e.g., marketing or economics), as well as the application of game-based learning in the university environment to enhance learning and motivation, which have increased their scientific presence in recent years. Conclusions: thanks to the results of this study, there is an overall understanding of the scientific approach that has been carried out around play-based pedagogy.