Social media knows what road it is: quantifying road characteristics with geo-tagged posts

Abstract

Determining traveling routes that provide opportunities to satisfy the various requirements of users in urban areas is still an open problem. This is because it is virtually impossible to manually quantify the characteristics of each street or road and there are few web-based or semantic resources of subjective requirements that describe streets and roads directly. Thus, it is difficult to satisfy all the needs and desires of users that may arise, such as for finding boulevards that are ''fashionable'' or ''beautiful.'' The goal of this study is to automatically quantify the characteristics of streets and roads in relation to requirements that can be described using keywords such as ''fashionable.'' To achieve this goal, we propose a twostage method that analyzes social media and road networks. First, in estimating the topic distribution (i.e., the characteristics) of each point of interest (POI), our method uses the latent Dirichlet allocation model to analyze geo-tagged texts while considering which users posted useful information for estimating road characteristics. Next, it uses a Markov random field model to estimate the characteristics of each street or road on the basis of those of the POIs and the road networks associated with the POIs. Experiments on real datasets demonstrate that our method achieves statistically significant improvements over baseline methods in terms of ranking quality in information retrieval for 300 roads in three urban areas in Tokyo when given 25 keywords.

Beyond the Shortest Route: A Survey on Quality-Aware Route Navigation for Pedestrians

“…Considering safety for example, the initially recommended route may need to be dynamically updated after acquiring information about a public safety risk (a group of hooligans moving towards the area) or based on the latest crime report in the area. Social media data (such as Twitter and Flickr) could be particularly useful for this purpose and while a number of studies have used such data to represent specific features on the route ( [14], [15], [66], etc. ), they are used in a static manner, namely to precalculate a specific quality on the route.…”

“…Additionally, unique dynamic events (such as public demonstrations or irregular sporting events) may influence the nature of conversations in geo-tagged tweets, the characteristics of photos from Flickr and visit numbers on Foursquare Swarm. Hence, when constructing machine learning models to predict route quality from such data (e.g., object and color detection for street quality estimation and semantic or sentiment analysis from tweet data [14], [15], [39], etc. ), the outlier data points would need to be removed and the data would need to be normalized.…”

“…In later studies, the authors developed an urban smell dictionary which they hinted could be used to identify routes which are pleasing in the olfactory sense [66]. Instead of crowd-sourced data, other systems used information collected from Google Street View, Twitter and OpenStreetMap to determine which routes were bright, contained beautiful views [39] were fashionable [15] or passed through locations that provide a high level of sociability, quietness and are close to nature [67]. Another study in this category proposed a route recommendation algorithm aimed at providing hikers with a sense of solitude while hiking, by generating routes which would allow them to avoid meeting other people along the trail [68].…”

“…), modern systems could draw upon data about the opinions, activities and behaviors of various population groups at different locations through online social media services such as Twitter and Foursquare [13] and such data have been used to represent the subjective attributes (i.e., perceived beauty, perceptions of safety, etc.) of a path [14], [15].…”