Why do people move? Enhancing human mobility prediction using local functions based on public records and SNS data

Area of Science:

• Urban Planning and Human Mobility
• Computational Social Science
• Geographic Information Science

Background:

• Improving urban quality of life necessitates accurate prediction of human mobility patterns.
• Traditional urban planning models often struggle to capture the complexities of inner-city movement.
• Gravity models, adapted from physics, offer a potential framework for understanding urban interactions.

Purpose of the Study:

• To evaluate the applicability of the urban traffic gravity model in inner-city versus intra-city regions.
• To identify and compare key variables influencing human mobility, including population, employment, and social media activity.
• To define and assess 'functional distance' as a predictor of urban human mobility.

Main Methods:

• Compared resident population, employee numbers, and social networking service (SNS) posts for gravity model mass.
• Evaluated straight-line distance, travel distance, and time as potential distance metrics.
• Utilized dimension reduction for public records and machine learning clustering for SNS data to define urban social functions and functional distance (Euclidean distance between social function vectors).

Main Results:

• The study identified key variables for mass (population, employment, SNS activity) and distance (functional distance) in urban mobility.
• Functional distance, calculated as the Euclidean distance between social function vectors, emerged as a significant factor.
• The gravity model's effectiveness was assessed across different urban scales, highlighting the importance of nuanced distance measures.

Conclusions:

• Functional distance, derived from diverse social data, is a significant predictor of urban human mobility.
• This approach enhances the gravity model's utility for understanding and planning for inner-city dynamics.
• Accurate mobility prediction through refined models can lead to improved urban planning and quality of life.