Transportation modes identification from mobile phone data using probabilistic models

Authors:
Dafeng Xu;Guojie Song;Peng Gao;Rongzeng Cao;Xinwei Nie;Kunqing Xie
Affiliations:
Key Laboratory of Machine Perception, Ministry of Education, Peking University, China;Key Laboratory of Machine Perception, Ministry of Education, Peking University, China;IBM Research, Beijing, China;IBM Research, Beijing, China;Key Laboratory of Machine Perception, Ministry of Education, Peking University, China;Key Laboratory of Machine Perception, Ministry of Education, Peking University, China
Venue:
ADMA'11 Proceedings of the 7th international conference on Advanced Data Mining and Applications - Volume Part II
Year:
2011

Citing 6
Cited 0

Information Retrieval

Information Retrieval
Learning transportation mode from raw gps data for geographic applications on the web

Proceedings of the 17th international conference on World Wide Web
Understanding mobility based on GPS data

UbiComp '08 Proceedings of the 10th international conference on Ubiquitous computing
UbiGreen: investigating a mobile tool for tracking and supporting green transportation habits

Proceedings of the SIGCHI Conference on Human Factors in Computing Systems
Determining transportation mode on mobile phones

ISWC '08 Proceedings of the 2008 12th IEEE International Symposium on Wearable Computers
Mobile phone location determination and its impact on intelligent transportation systems

IEEE Transactions on Intelligent Transportation Systems

Quantified Score

Hi-index	0.00

Visualization

Abstract

Transportation modes identification is an important transportation research problem with wide applications. Traditional methods are mainly done based on GPS, WiFi and some other electronic devices, which are actually not in adequately widespread use. The popularity of mobile phones makes the work of identification by mobile phone data valuable. In this paper, based on mobile phone data without other equipment for assistance, we design a probabilistic method to identify transportation modes. The method consists of a Hidden Markov Model with two sub-models for different traffic conditions. The Speed Distribution Law (SDL) based approach is used under the normal condition; to improve the performance of our method under the congested condition, the Cumulative Prospect Theory (CPT) based approach is adopted as a supplementary way to do identification. Experiments on real data show that our method can reach high accuracy in the normal and congested condition alike.