Temporal understanding of human mobility: A multi-time scale analysis.

Area of Science:

• Mobility studies
• Data science
• Human behavior analysis

Background:

• Previous studies on human mobility relied on low-resolution cellphone call data, limiting analysis across various time scales.
• Sparse and irregular call data, especially with the rise of mobile internet, hindered detailed investigation of human mobility patterns.

Purpose of the Study:

• To introduce and utilize Mobile Flow Records (MFR) for high-resolution, large-scale human mobility analysis.
• To investigate the influence of different time scales on human mobility models and identify biases.
• To explore novel features of human mobility, such as transportation modes, motion speed, duration, and trajectory distance.

Main Methods:

• Employed Mobile Flow Records (MFR), collecting high-resolution online activity data from smartphone users.
• Analyzed mobility patterns for a large population (3,542,235 individuals) with fine granularity (7.5min).
• Investigated statistical variations and biases in mobility models across time scales ranging from 7.5 minutes to 32 hours.

Main Results:

• Demonstrated that time scale significantly influences human mobility models, indicating a strong coupling between mobility and time.
• Revealed that mobility behaviors, including transportation modes, contribute to the diversity of human movement.
• Found that previous 2-hour sampling resolutions are insufficient for detailed motion behavior studies.

Conclusions:

• Mobile Flow Records provide unprecedented macroscopic and microscopic views of spatial-temporal human mobility.
• The study highlights the limitations of previous data resolutions and introduces valuable new features for mobility research.
• Findings are beneficial for studies on phenomena driven by human mobility, offering deeper insights into human movement patterns.