A methodology to characterize and compute correlation between traffic congestion and health issues via social media

Abstract

Traffic congestion is one of the most significant problems around the world. Literature shows various analyses of real time traffic incidents detection and crowd sensing. However, few researchers quantified the traffic congestion impacts on public health. To the best of our knowledge, there is no study, which determines the correlation between traffic congestion and public health issues via social media. In this paper, we propose a methodology to compute the correlation between traffic congestion and public health issues through social media analysis. To purse this task, we have used topic modeling and sentimental analysis. We mined a collection of 97 million tweets extracted from Twitter. Subsequently, different filters are applied to get the most traffic-congested locations around the world and the top health issues in the corresponding areas. Additionally, we have performed sentimental analysis to get the public perception about the initiatives taken to improve the health issues in those regions. We have found 36 most traffic congested cities around the world, such as Mexico, Bangkok, Jakarta and Chongqing etc. Apart from that, heart diseases, respiratory and psychological problems are identified as the common problems in traffic congested cities. Almost 71% public comments shows the negative sentiments. Which reflects their level of frustration about the steps taken to reduce the traffic by the higher authorities.