Using Open Data to Create Smart Auditory based Pervasive Game Environments

In this paper, we discuss an on-going research carried out to create smart auditory based pervasive game environments by using open street map and social media data. Two examples are provided, the first showing how data from Twitter could be used to generate talking NPC (Non Player Character) as a way to represent the social presence of people who frequent different areas. In the second prototype, we show how a context aware storyteller could be developed to tell stories from books based on the characteristics of the physical world locations where the user is located. Natural Language processing and Machine learning techniques such as clustering and word embedding based modeling were used to develop the prototypes. The approaches discussed in this study would be particularly useful in developing pervasive games with serious purposes and we hope to apply these techniques to create pervasive games for mental healthcare and education.