Abstract
Probe person survey by smartphone is expected to serve as a research method that gathers large-scale travel behavior of individuals in a flexible manner. In this study, we developed a method to identify transportation mode from movement history data such as GPS data that is automatically collected, with the aim of improving research methods by automatic transportation mode identification. We propose a simple discrimination technique for practical application and verify its discrimination accuracy by using the movement history data collected by the probe person survey by smartphones in practice. It is shown that more than 80% of transportation mode is identified accurately by the technique even using various movement history data from various participants. We consider the utility of the method from the verification results and describe the direction of practical use.
