Obtaining the time and space features of the travel of urban residents can facilitate urban traffic optimization and urban planning. As traditional methods often have limited sample coverage and lack timeliness, the application of big data such as mobile phone data in urban studies makes it possible to rapidly acquire the features of residents’ travel. However, few studies have attempted to use them to recognize the travel modes of residents. Based on mobile phone call detail records and the Web MapAPI, the present study proposes a method to recognize the travel mode of urban residents. The main processes include: (a) using DBSCAN clustering to analyze each user’s important location points and identify their main travel trajectories; (b) using an online map API to analyze user’s means of travel; (c) comparing the two to recognize the travel mode of residents. Applying this method in a GIS platform can further help obtain the traffic flow of various means, such as walking, driving, and public transit, on different roads during peak hours on weekdays. Results are cross-checked with other data sources and are proven effective. Besides recognizing travel modes of residents, the proposed method can also be applied for studies such as travel costs, housing–job balance, and road traffic pressure. The study acquires about 6 million residents’ travel modes, working place and residence information, and analyzes the means of travel and traffic flow in the commuting of 3 million residents using the proposed method. The findings not only provide new ideas for the collection and application of urban traffic information, but also provide data support for urban planning and traffic management.
The Influence of Distance and Neighbourhood Surroundings on Children’s Travel Mode to School