The spatial pattern of POIs density (i.e. the number of POIs in 500 m grids) collected from Google Maps (left) and OSM (right) in large (a), middle (b), and small (c) size cities.
Newswise — In a pioneering study published in Geo-spatial Information Science, a research group led by Yuyu Zhou from The University of Hong Kong developed an integrated framework that achieves 94% accuracy in mapping building functions across 50 U.S. cities using multi-source web-based geospatial data, offering potential for worldwide application. The team utilized TripAdvisor.com and Google Maps as primary data sources due to their extensive coverage and dynamic updating features.
Highlights
1. Broad Applicability: With the universal presence of Google Maps and TripAdvisor.com, this framework can potentially be applied to cities globally, offering a standardized method for urban building function mapping.
2. Resource Efficiency: The study circumvents the expensive and tedious process of gathering social sensing datasets. Instead, it capitalizes on readily available online geospatial data, ensuring scalability and cost-effectiveness.
3. Potential Extensions: The framework can further be utilized for a myriad of urban studies, from evaluating green spaces to assessing pedestrian exposure to environmental pollutants.
The research findings are promising. The proposed framework demonstrated an impressive average overall accuracy of 94% with a kappa coefficient of 0.63. While the study centered on U.S. cities, its potential for global application, especially in regions with data accessibility challenges, is immense. Furthermore, as the digital footprint of cities grows, the researchers see opportunities to refine this approach. They recommend future endeavors to enhance mapping accuracy, especially in multi-functional buildings and in regions less covered by TripAdvisor.com and Google Maps. This research provides a powerful tool for city planners, policy makers, and researchers, with the hope that the integrated framework becomes an essential asset in the global pursuit of sustainable, efficient, and inclusive urban spaces.
###
References
This press release is written by Shujuan Zhang from Wuhan University.
10.1080/10095020.2023.2264342
https://doi.org/10.1080/10095020.2023.2264342
About Geo-spatial Information Science
Geo-spatial Information Science is an open access journal that publishes research on the application and development of surveying and mapping technology. Geo-spatial Information Science was founded by Wuhan University and is now published in partnership with Taylor & Francis.The journal particularly encourages papers on innovative applications and theories in the fields above, or of an interdisciplinary nature.Geo-spatial Information Science’s editorial committee consists of 21 professors and research scientists from different regions and countries, such as America, Germany, Switzerland, Austria, and China. All articles are made freely and permanently available online through gold open access publication.
RSS