科研之友
微信
新浪微博
Facebook
分享链接
摘要
Personal travel pattern is significant to transportation analysis and modeling, and the rapid development of in-depth application of location-based services makes it possible to obtain large-scale positioning data. So, it is crucial to develop proper algorithm to identify trips/trip-segments from individual positioning records. This article presents an automatic trips/trip-segment detection method based on instantaneous Global Positioning System records collected by smartphones. The method consists of a series of procedures including data cleaning and pre-processing, inferring and removing pseudo trip ends, as well as trip combination. The result of the model has been compared with the "ground truth" collected and verified by volunteers. Finally, 1954 trips from 125 volunteers were identified and the overall detection accuracy is between 97.5% and 98.7% with a 95% confidence level. Besides, purity was introduced to evaluate the performance of the proposed method. In addition, the integration of instantaneous speed over time shows an excellent performance in calculating the trip distance.
- 出版日期2017-6-6
- 单位上海交通大学
