Deep Convolutional Bidirectional LSTM Based Transportation Mode Recognition

Traditional machine learning approaches for recognizing modes of transportation rely heavily on hand-crafted feature extraction methods which require domain knowledge. So, we propose a hybrid deep learning model: Deep Convolutional Bidirectional-LSTM (DCBL) which combines convolutional and bidirectional LSTM layers and is trained directly on raw sensor data to predict the transportation modes. We compare our model to the traditional machine learning approaches of training Support Vector Machines and Multilayer Perceptron models on extracted features. In our experiments, DCBL performs better than the feature selection methods in terms of accuracy and simplifies the data processing pipeline. The models are trained on the Sussex-Huawei Locomotion-Transportation (SHL) dataset. The submission of our team, Vahan, to SHL recognition challenge uses an ensemble of DCBL models trained on raw data using the different combination of sensors and window sizes and achieved an F1-score of 0.96 on our test data.

[1]  Zhezhuang Xu,et al.  Learning Transportation Modes From Smartphone Sensors Based on Deep Neural Network , 2017, IEEE Sensors Journal.

[2]  Mani Srivastava,et al.  MiLift: Efficient Smartwatch-Based Workout Tracking Using Automatic Segmentation , 2018, IEEE Transactions on Mobile Computing.

[3]  Xiaohui Peng,et al.  Deep Learning for Sensor-based Activity Recognition: A Survey , 2017, Pattern Recognit. Lett..

[4]  Arash Jahangiri,et al.  Developing a Support Vector Machine (SVM) Classifier for Transportation Mode Identification by Using Mobile Phone Sensor Data , 2014 .

[5]  Yang Wang,et al.  Identifying Different Transportation Modes from Trajectory Data Using Tree-Based Ensemble Classifiers , 2017, ISPRS Int. J. Geo Inf..

[6]  Héctor Pomares,et al.  Window Size Impact in Human Activity Recognition , 2014, Sensors.

[7]  Hesham A. Rakha,et al.  Applying Machine Learning Techniques to Transportation Mode Recognition Using Mobile Phone Sensor Data , 2015, IEEE Transactions on Intelligent Transportation Systems.

[8]  Gary M. Weiss,et al.  Activity recognition using cell phone accelerometers , 2011, SKDD.

[9]  Ling Bao,et al.  Activity Recognition from User-Annotated Acceleration Data , 2004, Pervasive.

[10]  Deborah Estrin,et al.  Using mobile phones to determine transportation modes , 2010, TOSN.

[11]  Xuan Song,et al.  DeepTransport: Prediction and Simulation of Human Mobility and Transportation Mode at a Citywide Level , 2016, IJCAI.

[12]  Yuan Yu,et al.  TensorFlow: A system for large-scale machine learning , 2016, OSDI.

[13]  Gaël Varoquaux,et al.  Scikit-learn: Machine Learning in Python , 2011, J. Mach. Learn. Res..

[14]  Jae-Young Pyun,et al.  Deep Recurrent Neural Networks for Human Activity Recognition , 2017, Sensors.

[15]  Nathan Srebro,et al.  Exploring Generalization in Deep Learning , 2017, NIPS.

[16]  Shih-Hau Fang,et al.  Transportation Modes Classification Using Sensors on Smartphones , 2016, Sensors.

[17]  Philip S. Yu,et al.  Transportation mode detection using mobile phones and GIS information , 2011, GIS.

[18]  Xiaoli Li,et al.  Deep Convolutional Neural Networks on Multichannel Time Series for Human Activity Recognition , 2015, IJCAI.

[19]  Lin Wang,et al.  Summary of the Sussex-Huawei Locomotion-Transportation Recognition Challenge , 2018, UbiComp/ISWC Adjunct.

[20]  Geoffrey E. Hinton,et al.  Deep Learning , 2015, Nature.

[21]  Hasan Ogul,et al.  Integrating Features for Accelerometer-based Activity Recognition , 2016, EUSPN/ICTH.

[22]  Bernt Schiele,et al.  A tutorial on human activity recognition using body-worn inertial sensors , 2014, CSUR.

[23]  Lin Wang,et al.  The University of Sussex-Huawei Locomotion and Transportation Dataset for Multimodal Analytics With Mobile Devices , 2018, IEEE Access.

[24]  Stefan Valentin,et al.  High reliability Android application for multidevice multimodal mobile data acquisition and annotation , 2017, SenSys.

[25]  Sozo Inoue,et al.  Recognition of multiple overlapping activities using compositional CNN-LSTM model , 2017, UbiComp/ISWC Adjunct.

[26]  Luo Haiyong,et al.  A convolutional neural networks based transportation mode identification algorithm , 2017, 2017 International Conference on Indoor Positioning and Indoor Navigation (IPIN).

[27]  Daniel Roggen,et al.  Deep Convolutional and LSTM Recurrent Neural Networks for Multimodal Wearable Activity Recognition , 2016, Sensors.

[28]  Wei-Ying Ma,et al.  Understanding mobility based on GPS data , 2008, UbiComp.

[29]  Thomas Plötz,et al.  Deep, Convolutional, and Recurrent Models for Human Activity Recognition Using Wearables , 2016, IJCAI.

[30]  Bo Yu,et al.  Convolutional Neural Networks for human activity recognition using mobile sensors , 2014, 6th International Conference on Mobile Computing, Applications and Services.

[31]  Stefan Valentin,et al.  A Versatile Annotated Dataset for Multimodal Locomotion Analytics with Mobile Devices , 2017, SenSys.

[32]  M. Gams,et al.  Comparing Deep and Classical Machine Learning Methods for Human Activity Recognition using Wrist Accelerometer , 2016 .

[33]  Guigang Zhang,et al.  Deep Learning , 2016, Int. J. Semantic Comput..

[34]  Wei Guo,et al.  Analyzing Urban Human Mobility Patterns through a Thematic Model at a Finer Scale , 2016, ISPRS Int. J. Geo Inf..