TweeVist: A Geo-Tweet Visualization System for Web based on Spatio-Temporal Events

This paper presents TweeVist, a geo-tweet visualization system to support users grasp event happens over time and space from tweets while they browse any web pages based on spatio-temporal analysis. TweeVist presents a tag cloud of tweets in different time periods are associated with web pages based on detected events. In order to detect events, the system extracts normal events (e.g., crowded restaurants, crowded facilities in Walt Disney World) happen at anytime and anywhere by utilizing machine learning algorithms, and it also extracts unusual or special events (e.g., time sales, disasters) by comparing current situation to those normal regularities. Thus, TweeVist can effectively visualize a summary of tweets in a tag cloud to help users immediately gain a quick overview of current situation or events through time and space while they browse a web page, and it can also effectively present a list of most related tweets to help users easily obtain more detailed information. Furthermore, TweeVist provides a communication function, which allows users to chat with other users who browse the same web pages, or Twitter users who follow an account of TweeVist.