Relationship between Number of Shopping Visitors to Roadside Rest Area and Tweets

Abstract

The utilization of time series analysis has seen a significant surge, owing to the remarkable increase in the temporal granularity of data, primarily propelled by the widespread adoption of the Internet of Things (IoT) and the continuous enhancement of analytical capabilities. The primary objective of this research endeavor is to gain a comprehensive understanding of the intricate relationship between transportation demand and public interest through a meticulous regression analysis of various time-series datasets. In this context, the term "demand" pertains to the influx of visitors at roadside rest areas, and our dataset spans a duration of three years, with an additional one-year dataset provided by two distinct operators. The measure of "interest" is derived from the quantification of Twitter posts and newspaper reports related to these rest areas. The outcomes of our regression analysis unveiled a noteworthy association, indicating that a single tweet is linked with an approximate increase of either 5 or 25 visitors. However, it is imperative to underscore that our causal analysis has illuminated a unidirectional relationship, suggesting that the number of visitors exerts an influence on the volume of tweets, rather than the reverse scenario where tweets impact visitor numbers.