Resident travel feature extraction method integrating space-time clustering and support vector machine

Abstract

The invention discloses a resident travel feature extraction method integrating space-time clustering and a support vector machine. The method comprises the following steps: 1, performing mobile phonesensor data acquisition and questionnaire survey filling in a travel process; 2, collecting check line data; 3, preprocessing the mobile phone sensor data in the travel process to obtain complete travel sensor data of the individual in one day; 4, performing travel feature identification: (1) identifying travel endpoints and travel time by using a space-time clustering algorithm; (2) identifyingthe travel mode of each travel by using a support vector machine algorithm; and (3) verifying the identification result by utilizing the check line data. Compared with the prior art, the method has the positive effects that the method has the advantages of high identification precision, strong dynamism, large sample size, low investigation cost and the like, all travel characteristics of the areacan be obtained through cyclic identification, good basic data is provided for four-stage method prediction, and a powerful guarantee is provided for development of urban traffic planning and construction.

Description

technical field [0001] The invention belongs to the field of sensor data information identification of traffic big data, and is especially aimed at the identification of travel endpoints and travel modes of urban residents. Background technique [0002] Traffic planning is the basis of urban traffic construction, which refers to the use of scientific methods to predict the future traffic supply and demand of the city on the basis of the overall urban planning, and to rationally organize the construction work using existing resources. With the rapid development of urbanization, urban functions are becoming more and more complex. Scientific and reasonable transportation planning can help reduce the risk in the process of urban transportation development and provide a basis for urban transportation construction. [0003] In conventional traffic forecasting, the "four-stage forecasting method" is the most commonly used and classic method at present. By dividing the traffic area,...

AI Technical Summary

Problems solved by technology

[0006] In order to overcome the above-mentioned shortcomings of the prior art, the present invention proposes a resident travel feature extraction method that integrates spatio-temporal clustering and support vector machines, aiming to solve the problem of directly obtaining travel features using mobile phone GPS, accelerometer and other sensor data in a real environment. Problems, especially travel feature recognition in the case of multiple stops and accelerations in the real travel environment

Images