2015
Feng, T.; Timmermans, H. J. P.
Detecting activity type from gps traces using spatial and temporal information Journal Article
In: European Journal of Transport and Infrastructure Research, vol. 15, iss. 4, 2015, ISSN: 15677141.
Abstract | Links | BibTeX | Tags: Activity type, GPS, Random forest
@article{Feng2015,
title = {Detecting activity type from gps traces using spatial and temporal information},
author = {T. Feng and H. J. P. Timmermans},
url = {https://journals.open.tudelft.nl/ejtir/article/view/3103},
doi = {10.18757/ejtir.2015.15.4.3103},
issn = {15677141},
year = {2015},
date = {2015-01-01},
urldate = {2015-01-01},
journal = {European Journal of Transport and Infrastructure Research},
volume = {15},
issue = {4},
abstract = {Detecting activity types from GPS traces has been important topic in travel surveys. Compared to inferring transport mode, existing methods are still relatively inaccurate in detecting activity types due to the simplicity of their assumptions and/or lack of background information. To reduce this gap, this paper reports the results of an endeavour to infer activity type by incorporating both spatial information and aggregated temporal information. Three machine learning algorithms, Bayesian belief network, decision tree and random forest, are used to investigate the performance of these approaches in detecting activity types. The test is based on GPS traces and prompted recall data, collected in the Rijnmond region, The Netherlands. Results show that the random forest model has the highest accuracy. The model incorporating spatial and temporal information can predict activity types with an accuracy of 96.8% for the used dataset. These findings are expected to benefit research on the use of GPS technology to collect activity-travel diary data.},
keywords = {Activity type, GPS, Random forest},
pubstate = {published},
tppubtype = {article}
}
Detecting activity types from GPS traces has been important topic in travel surveys. Compared to inferring transport mode, existing methods are still relatively inaccurate in detecting activity types due to the simplicity of their assumptions and/or lack of background information. To reduce this gap, this paper reports the results of an endeavour to infer activity type by incorporating both spatial information and aggregated temporal information. Three machine learning algorithms, Bayesian belief network, decision tree and random forest, are used to investigate the performance of these approaches in detecting activity types. The test is based on GPS traces and prompted recall data, collected in the Rijnmond region, The Netherlands. Results show that the random forest model has the highest accuracy. The model incorporating spatial and temporal information can predict activity types with an accuracy of 96.8% for the used dataset. These findings are expected to benefit research on the use of GPS technology to collect activity-travel diary data.
2013
Feng, T.; Timmermans, H. J. P.
Transportation mode recognition using GPS and accelerometer data Journal Article
In: Transportation Research Part C: Emerging Technologies, vol. 37, 2013, ISSN: 0968090X.
Abstract | Links | BibTeX | Tags: Accelerometer, Activity type, Bayesian belief network, GPS, Transportation mode
@article{Feng2013b,
title = {Transportation mode recognition using GPS and accelerometer data},
author = {T. Feng and H. J. P. Timmermans},
doi = {10.1016/j.trc.2013.09.014},
issn = {0968090X},
year = {2013},
date = {2013-01-01},
journal = {Transportation Research Part C: Emerging Technologies},
volume = {37},
abstract = {Potential advantages of global positioning systems (GPS) in collecting travel behavior data have been discussed in several publications and evidenced in many recent studies. Most applications depend on GPS information only. However, transportation mode detection that relies only on GPS information may be erroneous due to variance in device performance and settings, and the environment in which measurements are made. Accelerometers, being used mainly for identifying peoples' physical activities, may offer new opportunities as these devices record data independent of exterior contexts. The purpose of this paper is therefore to examine the merits of employing accelerometer data in combination with GPS data in transportation mode identification. Three approaches (GPS data only, accelerometer data only and a combination of both accelerometer and GPS data) are examined. A Bayesian Belief Network model is used to infer transportation modes and activity episodes simultaneously. Results show that the use of accelerometer data can make a substantial contribution to successful imputation of transportation mode. The accelerometer only approach outperforms the GPS only approach in terms of the predictive accuracy. The approach which combines GPS and accelerometer data yields the best performance. © 2013 Elsevier Ltd.},
keywords = {Accelerometer, Activity type, Bayesian belief network, GPS, Transportation mode},
pubstate = {published},
tppubtype = {article}
}
Potential advantages of global positioning systems (GPS) in collecting travel behavior data have been discussed in several publications and evidenced in many recent studies. Most applications depend on GPS information only. However, transportation mode detection that relies only on GPS information may be erroneous due to variance in device performance and settings, and the environment in which measurements are made. Accelerometers, being used mainly for identifying peoples’ physical activities, may offer new opportunities as these devices record data independent of exterior contexts. The purpose of this paper is therefore to examine the merits of employing accelerometer data in combination with GPS data in transportation mode identification. Three approaches (GPS data only, accelerometer data only and a combination of both accelerometer and GPS data) are examined. A Bayesian Belief Network model is used to infer transportation modes and activity episodes simultaneously. Results show that the use of accelerometer data can make a substantial contribution to successful imputation of transportation mode. The accelerometer only approach outperforms the GPS only approach in terms of the predictive accuracy. The approach which combines GPS and accelerometer data yields the best performance. © 2013 Elsevier Ltd.