Inferring hybrid transportation modes from sparse GPS data using a moving window SVM classification

Bolbol, A; Cheng, Tao; Tsapakis, Ioannis; Haworth, James

doi:10.1016/j.compenvurbsys.2012.06.001

Cited by 160 publications

(105 citation statements)

References 11 publications

Supporting

Mentioning

101

Contrasting

Unclassified

Order By: Relevance

“…Studies [3], [4], and [5] have already demonstrated the possibility to use GPS data in Transportation research by capturing the characteristics of different types of trips. Later studies, such as [6] and [7] evaluated the use of processed GPS data for both trip tracking and transportationmode detection without the support of questionnaires. Their results showed that trip identification deviates slightly from the census data whereas for mode detection it was not possible to distinguish between transportation modes with similar speed, for instances bus and car trips.…”

Section: Introduction and Related Workmentioning

confidence: 99%

Travelled Distance Estimation for GPS-Based Round Trips Car-Sharing Use Case

Lopez

Šemanjski

Gillis

et al. 2016

ToMS

View full text Add to dashboard Cite

Abstract-Traditional travel survey methods have been widely used for collecting information about urban mobility although, since middle of the 90's Global Position System (GPS) has become an automatic option for collecting more precise data of the households. But how good is the collected data? many studies on mobility patterns have focused on the GPS advantages and leaving aside its issues. However, when it comes to extract the frequency of the trips and travelled distance this technology faces some gaps due to related issues, such as signal reception and time-to-first-fix location that turns out in missing observations and respectively unrecognised or over-segmented trips. In this study, we focus on two aspects of GPS data for a car-mode, (i) measurement of the gaps in the travelled distance and (ii) estimation of the travelled distance and the factors that influence the GPS gaps. To asses that, GPS tracks are compared to a ground truth source. Additionally, the trips are analysed based on the land use (e.g., urban and rural areas) and length (e.g., short, middle and long trips). Results from 170 participants and more than a year of GPS-tracking show that around 9% of the travelled distance is not captured by the GPS and it affects more to short trips than long ones. Moreover, we validate the importance of the time spent on the user activity and the land use as factors that influence the gaps on GPS.

show abstract

Section: Introduction and Related Workmentioning

confidence: 99%

Travelled Distance Estimation for GPS-Based Round Trips Car-Sharing Use Case

Lopez

Šemanjski

Gillis

et al. 2016

ToMS

View full text Add to dashboard Cite

show abstract

“…Ensemble methods are widely used in many machine learning challenges; for example, Random Forest was used in 2010 Knowledge Discovery and Data Mining (KDD) Cup to win the first prize [20], the Gradient Boosting Decision Tree (GBDT) was used in the Netflix prize [18], and XGBoost was successfully used in Kaggle competition and 2015 KDD Cup [19]. K-Nearest Neighbor [13,25], Decision Tree (DT) [7,[13][14][15]24,25] and Support Vector Machine (SVM) [12][13][14]17,23,25] have been widely used to classify transportation modes and have resulted in good performance. However, in this paper, we choose to use the tree ensemble models to classify transportation modes instead of using the above-mentioned methods.…”

Section: Model Classification and Model Evluationmentioning

confidence: 99%

Identifying Different Transportation Modes from Trajectory Data Using Tree-Based Ensemble Classifiers

Xiao

Yang

et al. 2017

IJGI

160

104

View full text Add to dashboard Cite

Abstract:Recognition of transportation modes can be used in different applications including human behavior research, transport management and traffic control. Previous work on transportation mode recognition has often relied on using multiple sensors or matching Geographic Information System (GIS) information, which is not possible in many cases. In this paper, an approach based on ensemble learning is proposed to infer hybrid transportation modes using only Global Position System (GPS) data. First, in order to distinguish between different transportation modes, we used a statistical method to generate global features and extract several local features from sub-trajectories after trajectory segmentation, before these features were combined in the classification stage. Second, to obtain a better performance, we used tree-based ensemble models (Random Forest, Gradient Boosting Decision Tree, and XGBoost) instead of traditional methods (K-Nearest Neighbor, Decision Tree, and Support Vector Machines) to classify the different transportation modes. The experiment results on the later have shown the efficacy of our proposed approach. Among them, the XGBoost model produced the best performance with a classification accuracy of 90.77% obtained on the GEOLIFE dataset, and we used a tree-based ensemble method to ensure accurate feature selection to reduce the model complexity.

show abstract

“…However, because of the limitations of earlier GPS devices, the sample size is usually small, and the sample density is usually sparse. 15 In 2004, Asakura and Hato 16 first used cell phone positioning functions to collect individual travel path data, obtained high-quality positioning data, and proved the feasibility of using smartphones for travel surveying. Thereafter, further studies were conducted, represented by Naphade, Douma, Gonzalez, and Stenneth in the United States and by Nitsche and Bierlaire in Europe, and the smartphone-based travel survey method became more mature.…”

Section: Introductionmentioning

confidence: 99%

Travel mode detection method based on big smartphone global positioning system tracking data

Zhou

Jia

Gao

et al. 2017

Advances in Mechanical Engineering

View full text Add to dashboard Cite

This article proposes a machine learning-based travel mode detection method using urban residents' travel routes as the data source, collected via smartphone global positioning system modules. A data-driven machine learning strategy was chosen in the model construction. This study performed data cleaning and mining on over 4400 pieces of urban resident travel records containing several millions of global positioning system tracking points. Series of characteristic values of speed, travel distance, and direction are calculated, which reflect the travel mode of smartphone holders. In travel mode identification, first, the transition regions of travel segments of different travel modes are effectively distinguished; then, continuous tracking points for single-mode travel are connected into single-mode travel segments. The travel mode of the surveyed subjects is identified based on the calculated features of average speed, average acceleration, and average change of direction within each single-mode segment. The random forest method is chosen as the basis model to classify travel mode. Three-quarters of the travel records were used to construct the random forest classifier, and the detection accuracy of the established model for the remaining ¼ of the travel record reached 94.4%. The proposed method uses massive smartphone global positioning system tracking points as the basis; the detection results are consistent with manually collected prompted recall survey records.

show abstract

Inferring hybrid transportation modes from sparse GPS data using a moving window SVM classification

Cited by 160 publications

References 11 publications

Travelled Distance Estimation for GPS-Based Round Trips Car-Sharing Use Case

Travelled Distance Estimation for GPS-Based Round Trips Car-Sharing Use Case

Identifying Different Transportation Modes from Trajectory Data Using Tree-Based Ensemble Classifiers

Travel mode detection method based on big smartphone global positioning system tracking data

Contact Info

Product

Resources

About