A multi-scale attributes fusion model for travel mode identification using GPS trajectories

Figures & data

Table 1. Classification of trajectory characteristics.

Class	General features
Motion features	Average speed, acceleration, maximum speed, minimum speed, etc;
Geometric features	Deflection angle, distance, curvature, etc;
Frequency domain feature	Periodicity, frequency distribution, spectral energy, frequency peaks, etc;
Other features	Land use, POI, road type, age, gender, weather conditions, weekdays, etc.

Figure 1. Summary of trajectory feature structures.

Figure 2. Schematic of GPS data point sequence.

Figure 3. The network structure of MSAF.

Figure 4. The process of time window selection, multi-scale definition, and attribute calculation in feature engineering.

Figure 5. Spatio-temporal feature extraction module used in MSAF.

Figure 6. Multi-scale feature fusion module used in MSAF.

Table 2. Model structure.

Modules	Blocks	Layers
STFEM	3D Convolution Block	[3*3] conv3D
	Residual Block	[3*3] conv*2
		GlobalAvgPool + [1*1] Conv + ReLU + [1*1] Conv + Sigmoid
		[3*3] conv(Shortcut)
	Pool Block	[2*2] AvgPool2d
MSFFM	Fully connected block	Fully connected*2
	Scale attention block	Fully connected + Tanh + Softmax

Table 3. Comparison of TMI accuracy with classical model for different time intervals.

Model	GeoLife dataset(%)				SHL dataset(%)
	15s	30s	45s	60s	15s	30s	45s	60s
DT	85.11	85.37	86.69	86.44	74.01	75.76	76.25	73.02
SVM	74.92	75.39	77.43	77.07	68.79	70.28	73.26	71.43
XGBoost	85.83	86.66	86.84	86.71	77.79	79.53	82.52	80.79
RF	86.62	87.70	88.12	87.13	82.52	84.01	84.76	85.50
CNN	88.85	90.16	90.66	90.52	87.43	88.67	88.92	89.42
CNN-attention	89.84	90.90	92.79	91.89	89.07	90.21	90.45	90.32
MSAF	92.87	94.02	95.16	93.36	90.18	90.94	91.70	90.45

Table 4. Confusion matrix for MSAF and baseline CNN-attention model.

CNN-attention (MSAF)		Predicted class
		Walk	Bike	Bus	Subway	Car
Actual class	Walk	486(499)	17(7)	5(1)	0(0)	0(0)
	Bike	6(4)	208(214)	15(10)	0(0)	0(1)
	Bus	2(1)	4(5)	264(269)	6(0)	11(12)
	Subway	0(0)	0(0)	1(2)	70(70)	5(4)
	Car	0(0)	0(0)	14(12)	2(0)	104(109)

Figure 7. Recall and precision of MSAF and baseline CNN-attention model.

Table 5. Comparison of accuracy with related studies.

Model	Test accuracy(%)
(Endo et al. 2016)	67.9
(Wang et al. 2017)	74.1
(Dabiri and Heaslip 2018)	84.8
(Jiang et al. 2020)	83.9
(Li et al. 2020)	86.7
(Li et al. 2021)	91.1
(Kim, Kim, and Lee 2022)	92.1
(Zeng et al. 2023)	85.23
(Ma et al. 2023)	92.25
MSAF	95.16

Table 6. The schematic setup for verifying the effectiveness of multi-scale feature combinations.

Case	T	Segmenting methods
Case 1	t=15s	K5_15
		K5_15, K7_15
		K5_15, K7_15, K9_15
		K5_15, K7_15, K9_15, K11_15
Case 2	t=30s	K5_30
		K5_30, K7_30
		K5_30, K7_30, K9_30
		K5_30, K7_30, K9_30, K11_30
Case 3	t=45s	K5_45
		K5_45, K7_45
		K5_45, K7_45, K9_45
		K5_45, K7_45, K9_45, K11_45
Case 4	t=60s	K5_60
		K5_60, K7_60
		K5_60, K7_60, K9_60
		K5_60, K7_60, K9_60, K11_60

Figure 8. Performance comparison of different scale combinations.

Table 7. Ablation experiments with MSAF.

Module	Configure	Test accuracy(%)
STFEM	Without-RL	91.67
	Without-CBAM	92.46
MSFFM	Without-SA	93.03
	MSAF	95.16

Endo, Y., H. Toda, K. Nishida, and J. Ikedo. 2016. “Classifying Spatial Trajectories Using Representation Learning.” International Journal of Data Science and Analytics 2 (3): 107–117. https://doi.org/10.1007/s41060-016-0014-1.

 Google Scholar

Wang, H., G. Liu, J. Duan, and L. Zhang. 2017. “Detecting Transportation Modes Using Deep Neural Network.” IEICE Transactions on Information and Systems 100 (5): 1132–1135. https://doi.org/10.1587/transinf.2016EDL8252.

 Google Scholar

Dabiri, S., and K. Heaslip. 2018. “Inferring Transportation Modes from GPS Trajectories Using a Convolutional Neural Network.” Transportation Research Part C: Emerging Technologies 86:360–371. https://doi.org/10.1016/j.trc.2017.11.021.

 Web of Science ®Google Scholar

Jiang, G., S. Lam, P. He, C. Ou, and D. Ai. 2020. “A Multi-Scale Attributes Attention Model for Transport Mode Identification.” IEEE Transactions on Intelligent Transportation Systems 23 (1): 152–164. https://doi.org/10.1109/TITS.2020.3008469.

 Web of Science ®Google Scholar

Li, L., J. Zhu, H. Zhang, H. Tan, B. Du, and B. Ran. 2020. “Coupled Application of Generative Adversarial Networks and Conventional Neural Networks for Travel Mode Detection Using GPS Data.” Transportation Research Part A: Policy and Practice 136:282–292. https://doi.org/10.1016/j.tra.2020.04.005.

 Web of Science ®Google Scholar

Li, J., X. Pei, X. Wang, D. Yao, Y. Zhang, and Y. Yue. 2021. “Transportation Mode Identification with GPS Trajectory Data and GIS Information.” Tsinghua Science and Technology 26 (4): 403–416. https://doi.org/10.26599/TST.2020.9010014.

 Web of Science ®Google Scholar

Kim, J., J. H. Kim, and G. Lee. 2022. “GPS Data-Based Mobility Mode Inference Model Using Long-Term Recurrent Convolutional Networks.” Transportation Research Part C: Emerging Technologies 135:103523. https://doi.org/10.1016/j.trc.2021.103523.

 Web of Science ®Google Scholar

Zeng, J., Y. Yu, Y. Chen, D. Yang, L. Zhang, and D. Wang. 2023. “Trajectory-As-A-Sequence: A Novel Travel Mode Identification Framework.” Transportation Research Part C: Emerging Technologies 146:103957. https://doi.org/10.1016/j.trc.2022.103957.

 Web of Science ®Google Scholar

Ma, Y., X. Guan, J. Cao, and H. Wu. 2023. “A Multi-Stage Fusion Network for Transportation Mode Identification with Varied Scale Representation of GPS Trajectories.” Transportation Research Part C: Emerging Technologies 150:104088. https://doi.org/10.1016/j.trc.2023.104088.

 Web of Science ®Google Scholar

Data availability statement

The GeoLife dataset can be found at https://www.microsoft.com/en-us/research/publication/geolife-gps-trajectory-dataset-user-guide/. The Sussex-Huawei Locomotion dataset can be downloaded from http://www.shl-dataset.org/.