Geospatial and Social Media Analytics for Emotion Analysis of Theme Park Visitors using Text Mining and GIS

Abstract

Scrutinizing the emotions of customers and social media analytics are gaining popularity in the recent days. However, analysis of the emotions of visitors in theme parks are done on a lesser scale. In this paper, based on social media messages, the emotions of the visitors of a theme park is analyzed using geospatial as well as social media analytics convergence and visualization of cohesive places where expressions are gathered. Based on the Russell’s Circumplex Model of Affect, the words and emotions are analyzed in around 50,000 tweets collected of which 20,400 tweets contained one or more such words. Analysis of exploratory spatial data based on GIS and analysis of text mining represents various emotion in each quadrant based on the tweets. The visitor emotions are associated to various topics and emotions of considerable spatial variations. Based on the significant clustering of emotions in each quadrant, the areas of riding attraction in the theme park are identified and displayed using this research approach. Based on the analysis and implications of this research work, it is possible to develop ways in which the pleasant emotions of the visitors can be evoked by practitioners.