Tourists Preferred Streets Visualization Using Articulated GPS Trajectories Driven by Mobile Sensors

Abstract

Although there has been a lot of research on GPS big data analysis to discover the preferred spots and destinations for tourists, discovering the preferred streets also plays an important role in improving tourism content and optimizing transportation systems. In general, GPS data has inaccuracies and redundancies, and its density is biased to specific locations. This causes difficulty making tourists preferred streets stand out by using density-based visualizations such as heatmaps. In this study, we attempted to apply a mobile sensor-based data cleaning method, which is executed at the end-user device, and thereby equalize the bias of each tourist’s trajectory data. This paper examines how to generate a heatmap from such lightweight and non-unstable data and whether it can effectively visualize tourists preferred streets using real tourism data. A heat map was created by plotting GPS data on a geographic grid square and focusing on the quartiles of density values. The results were almost consistent with the tendency of actual tourist routes. The contribution of this research is that it does not require any preparation such as a road network, and does not require a large amount of computation compared to conventional approaches.