Speaker
Description
Background
Big Data and real-time information from social media are increasingly being analyzed in research on the relationship between urban spaces, people’s behavior, and their satisfaction with the living environment. The integration of technology in urban planning, as enabled by these tools, offers enhanced insights into urban phenomena.
Topic
Big Data and real-time social media information were utilized in two investigations of urban parks conducted in Suzhou, Jiangsu Province, China. On one hand, there is growing interest in understanding the built environment’s impact on citizens’ well-being and health; on the other, the lack of public spaces is a persistent concern in China's rapid urbanization.
One case study examined how the built environment surrounding urban parks influences visitor attraction. The other study investigated the mental restorative effects of historic gardens located in Suzhou’s old town, a UNESCO World Heritage Site.
These investigations are significant for their focus—evaluating the performance of various green spaces in medium-density urban environments—and their innovative methodology for understanding how users interact with and value green spaces.
The primary research question for the first case was: How do land use and the mix of land uses within a 400-meter buffer zone around urban parks influence visitor flows? For the second case, the central question was: How can the restorative capacity of green heritage landscapes be systematically assessed by combining on-site and online sentiment analysis?
Research Method
One of the main objectives of the case studies was to define and test new investigative methods.
The first investigation analyzed large quantities of accurate and accessible data connecting users to urban spaces, including points of interest (POI) and Weibo check-in data. It evaluated the sensitivity of different types of parks to various POIs.
The second investigation proposed a novel method that combined traditional data collection tools—such as on-site observation, place description, and in-person questionnaires—with advanced techniques like social media POI GIS-based mapping and Natural Language Processing (NLP). The unstructured social media data, including geotagged posts and reviews from platforms like Weibo and Xiaohongshu, were processed and statistically analyzed to uncover patterns and relationships.
Results and Discussion
The paper discusses and compares the methods and findings of the two investigations. While both relied on location-based analysis and social media data, the first method used only online data, focusing on the relationship between visitor intensity and POI diversity. The second method combined large-scale online data with smaller, on-site data to provide a comprehensive understanding of visitors’ sentiments.
Both approaches successfully extracted useful insights and patterns from raw data and addressed the research questions. However, the mixed method demonstrated greater effectiveness in producing relevant findings and allowed for comparisons between on-site and online expressions, despite their differing formats.
Conclusion
Real-time and dynamic social media data are invaluable for studying the relationship between urban spaces and people’s behavior, preferences and satisfaction, especially through POI and geotagged posts. Although such data can be informal and subject to inaccuracies, they reveal interactions with green spaces, visitor sentiments and behavioral patterns. Urban planners and policymakers can leverage these insights to enhance the functionality of public spaces, support evidence-based decision-making, and allocate resources more effectively.
| Keywords | Big Data; social media; urban park; people behavior; public space |
|---|---|
| Best Congress Paper Award | Yes |
