Authors: Chao Yang*, China University of Geosciences(Wuhan), Wenwen Tian, China University of Geosciences(Wuhan)
Topics: Social Geography, China, Behavioral Geography
Keywords: Human appearance; Social media data; AI; sociology study
Session Type: Paper
Start / End Time: 1:10 PM / 2:50 PM
Room: Coolidge, Marriott, Mezzanine Level
Presentation File: No File Uploaded
Appearance, as one of the external attributes and identity characteristics of human beings, has very important significance in social and economic activities, such as affecting employment and wage levels, finding spouses, fairness and justice.
In this study, we explore the role of appearance in human social activities from a geographical perspective by exploring the differences in behavioral patterns and mobility exhibited by groups of people with different appearances.
Our research data comes from young social media users in Wuhan and Hangzhou, China, they posted a series of self-portrait photos and check-in data. First, we use deep learning and data mining method to extract social media user's photos and evaluate their appearance grade. Then, we analyze the trajectory and footprint of the user's check-in data to analyze the mobility and behavior patterns of different appearance groups. The study shows that there are differences between mobility and behavioral patterns between groups with different appearance, including user's active time, activity preferences, and even consumption levels. The significance of this paper is to explore sociology data in social media content through Artificial Intelligence methods (including deep learning and data mining technologies), and explore new ideas, data sources and methods of social research through geographic analysis.