Search

By utilizing cognitive mapping and leveraging georeferenced text data, this paper aims to suggest a new visualization method that combines the advantages of both conventional and state-of-the-art research techniques to depict the collective identity of place in a single image. The study addressed two research questions: (1) Can crowd-sourced text data be utilized in representing place identity? (2) Can collective place identity be expressed in the form of a cognitive map? By confirming that text data gathered from social media effectively demonstrate people’s behaviors and perceptions related to places, we propose a novel method to create a visual representation of urban identity–a “crowd-sourced cognitive map”. In particular, to improve the conventional cognitive mapping method to depict the collective identity of a city, we draw cognitive maps of Bundang and Ilsan developed in the 1990s, as well as Songdo and Dongtan developed in the 2000s, just outside of the administrative boundaries of Seoul in Korea, through a computational method based on crowd-sourced opinions collected from social media. We open the possibility for the use of social media text data to capture the identity of cities and suggest a graphical image through which people without prior information could also easily apprehend the overall image of a city. The work in this paper is expected to provide a methodological technique for appropriate decision-making and the evaluation of urban identity to shape a more unique and imageable city. [ABSTRACT FROM AUTHOR]

In highly urbanized area where traffic condition fluctuates constantly, transportation infrastructure is one of the major contributing factors to Emergency Medical Service (EMS) availability and patient outcome. In this paper, we assess the impact of traffic fluctuation to the EMS first response availability in urban area, by evaluating the k-minute coverage under 21 traffic scenarios. The set of traffic scenarios represents the time-of-day and day-of-week effects, and is generated by combining road link speed information from multiple historical speed databases. In addition to the k-minute area coverage calculation, the k-minute population coverage is also evaluated for every 100m by 100m grid that partitions the case study area of Seoul, South Korea. In the baseline case of traveling at the speed limit, both the area and population coverage reached nearly 100% when compared to the five-minute travel time national target. Employing the proposed LoST (Loss of Serviceability due to Traffic) index, which measures coverage reduction in percentage compared to the baseline case, we find that the citywide average LoST for area and population coverage are similar at 34.2% and 33.8%. However, district-wise analysis reveals that such reduction varies significantly by district, and the magnitude of area and population coverage reduction is not always proportional. We conclude that the effect of traffic variation is significant to successful urban EMS first response performance, and regional variation is evident among local districts. Complexity in the urban environment requires a more adaptive approach in public health resource management and EMS performance target determination. [ABSTRACT FROM AUTHOR]

Solving the supply–demand imbalance is the most crucial issue for stable implementation of a public bike-sharing system. This gap can be reduced by increasing the accuracy of demand prediction by considering spatial and temporal properties of bike demand. However, only a few attempts have been made to account for both features simultaneously. Therefore, we propose a prediction framework based on graph convolutional networks. Our framework reflects not only spatial dependencies among stations, but also various temporal patterns over different periods. Additionally, we consider the influence of global variables, such as weather and weekday/weekend to reflect non-station-level changes. We compare our framework to other baseline models using the data from Seoul’s bike-sharing system. Results show that our approach has better performance than existing prediction models. [ABSTRACT FROM AUTHOR]