Analysing Games for Health through Users' Opinion Mining
Renato Tadeu dos Santos, Joel Perdiz Arrais, Paula Alexandra Silva · 2021
Serious games are a category of games which purpose extends beyond entertainment. Among these, we find a specific type of games, exergames, which aim to promote physical activity. Despite the positive influence of exergames on their users, players often stop playing them after a short period of time, losing the positive benefits of gameplay. It is in this context, that the need to better understand the experience and the opinions of players emerges. To grasp what users feel and (dis)like is key so that games can be redesigned and improved to fit users preferences. This work proposes to analyse users' comments from YouTube, using Natural Language Processing techniques, to extract knowledge on usability, user experience and perceived impacts on health, in particular on quality of life, that could inform the redesign of the game thereafter. This paper is a work in progress that reports on preliminary work that explores users opinions about the Just Dance game. In mining users' opinions, the process of annotation of usability, user experience and quality of life dimensions is based on a pre-established vocabulary. Each extracted opinion is annotated with the concepts present in the opinion, using an approach that is based on the English dictionary lexicon in conjunction with sentiment analysis. The results obtained are then displayed on a dashboard, where the data extracted from the previously collected user comments can be viewed, analysed, and explored.