Mining Insights from Virtual Reality Exergame Reviews with Latent Dirichlet Allocation Method
Shuhan Dong, Yijun Lü, Yu Wang, Jiong Dong · 2023
The increasing popularity of VR games has sparked a notable surge in interest toward VR exergames, particularly in the context of the COVID-19 pandemic. To uphold the quality of VR exergames and provide valuable feedback to developers, it is imperative to consider the perspectives and concerns of players regarding VR exergames and conduct a comprehensive analysis of their reviews. This study employs the Latent Dirichlet Allocation (LDA) algorithm to effectively discern various topics within reviews, thus elucidating players’ reservations concerning VR exergames. Empirical investigations were conducted using a dataset comprising 15,917 English-language reviews of 16 VR exergames on the Steam platform. The experimental findings unequivocally indicate that players exhibit a favorable disposition toward VR exergames (88.31% vs. 11.69%). The analysis encompasses eight distinct topics including user experience, graphics, music, and gameplay.