Advertisement

Comparative Study of Batch and Stream Learning for Online Smartphone-based Human Activity Recognition

  • Ilham AmezzaneEmail author
  • Youssef Fakhri
  • Mohamed El Aroussi
  • Mohamed Bakhouya
Conference paper
Part of the Lecture Notes in Intelligent Transportation and Infrastructure book series (LNITI)

Abstract

The availability of diverse embedded sensors in modern smartphones has created exciting opportunities for developing context-aware services and applications, such as Human activity recognition (HAR) in healthcare and smart buildings. However, recognizing human activities using smartphones remains a challenging task and requires efficient data mining approaches due to the limited resources of the device. For example, the training process is usually performed offline but rarely online on the mobile device itself, because traditional batch learning usually needs a large dataset of many users. Therefore, building models using complex multiclass algorithms is generally very time-consuming. In this paper, we present a comparison study using two approaches in order to reduce training time and memory usage while maintaining significant performance. In the first approach, we conducted experiments using batch learning on a GPU platform. Results showed that High Performance Extreme Learning Machine (HPELM) offers the best compromise: accuracy, memory usage and training time. Moreover, it achieved better performance on two dynamic activities, outperforming the best SVM model obtained in our previous study. In the second approach, we conducted experiments using online stream learning on the MOA platform. Unlike the first approach, experiments were performed using accelerometer data only. We also studied the effects of user/device dependency and feature engineering on the classification performance and memory usage by comparing five constructed real data streams. Simulation results showed that Hoeffding Adaptive Tree has comparable performance to batch learning, especially for user and device dependent data streams.

Keywords

Human activity recognition GPU Online stream learning 

References

  1. 1.
    Kharboucha, A., El Khoukhi, H., Nait Malek, Y., Bakhouya, M., De Florioa, V., El Ouadghiri, D., Latre S., Blondia, C.: Towards an IoT and big data analytics platform for the definition of diabetes telecare services. In: The 2nd International Conference on Smart Applications and Data Analysis for Smart Cities (SADASC’18), Casablanca, Morocco (2018)Google Scholar
  2. 2.
    De Florio, V., Bakhouya, M., Eloudghiri, D., Blondia, C.: Towards a Smarter organization for a Self-servicing Society. DSAI 2016, pp. 254–260, Vila Real, PortugalGoogle Scholar
  3. 3.
    Elkhoukhi, A., NaitMalek, Y., Berouine, A., Bakhouya, M., Elouadghiri, D., Essaaidi, M.: Towards a Real-time Occupancy Detection Approach for Smart Buildings, in The 15th International Conference on Mobile Systems and Pervasive Computing, Gran Canaria, Spain, Procedia Computer Science, Elsevier (2018)Google Scholar
  4. 4.
    Krawczyk, B.: Active and adaptive ensemble learning for online activity recognition from data streams. Knowl.-Based Syst. 138, 69–78 (2017)CrossRefGoogle Scholar
  5. 5.
    Shoaib, M., et al.: A survey of online activity recognition using mobile phones. Sensors 15(1), 2059–2085 (2015)MathSciNetCrossRefGoogle Scholar
  6. 6.
    Kose, M., Incel, O.D., Ersoy, C.: Online human activity recognition on smart phones. In: Proceedings of the Workshop on Mobile Sensing: From Smartphones and Wearables to Big Data, Beijing, China, pp. 11–15, 16 April 2012Google Scholar
  7. 7.
    Gomes, J.B., Krishnaswamy, S., Gaber, M.M., Sousa, P.A., Ruiz, E.M.: MARS: A personalised mobile activity recognition system. 2012 IEEE 13th International Conference on Mobile Data Management, pp. 316–319 (2012)Google Scholar
  8. 8.
    Lu, D.N., Nguyen, T.T., Ngo, T.T., Nguyen, T.H., Nguyen, H.N.: Mobile online activity recognition system based on smartphone sensors. Adv. Inf. Commun. Technol. 538, 357–366 (2016)CrossRefGoogle Scholar
  9. 9.
    Spinsante, S. et al.: A mobile application for easy design and testing of algorithms to monitor physical activity in the workplace. Mobile Information Systems, 5126816:1–5126816:17 (2016)CrossRefGoogle Scholar
  10. 10.
    Chen, Z., Zhu, Q., Soh, Y.C., Zhang, L.: Robust Human Activity Recognition Using Smartphone Sensors via CT-PCA and Online SVM. IEEE Trans. Industr. Inf. 13, 3070–3080 (2017)CrossRefGoogle Scholar
  11. 11.
    Abdallah, Z.S., Gaber, M.M., Srinivasan, B., Krishnaswamy, S.: Adaptive mobile activity recognition system with evolving data streams. Neurocomputing. 150, 304–317 (2015)CrossRefGoogle Scholar
  12. 12.
    Abadi, M., Barham, P., Chen, J., Chen, Z., Davis, A., Dean, J., Devin, M., Ghemawat, S., Irving, G., Isard, M., Kudlur, M.: Tensorflow: large-scale machine learning on heterogeneous distributed systems. arXiv preprint arXiv:1603.04467 (2016)
  13. 13.
    Inoue, M., Inoue, S., Nishida, T.: Deep recurrent neural network for mobile human activity recognition with high throughput. Artif. Life. Rob. 23, 173–185 (2017)CrossRefGoogle Scholar
  14. 14.
    Li, P., Wang, Y., Tian, Y., Zhou, T.S., Li, J.S.: An automatic user-adapted physical activity classification method using smartphones. IEEE Trans. Biomed. Eng. 64(3), 706–714 (2017)Google Scholar
  15. 15.
    Amezzane, I., Fakhri, Y.El, Aroussi, M., Bakhouya, M.: Towards an efficient implementation of human activity recognition for mobile devices. EAI Endorsed Trans. Context-aware Syst. Appl. 18(13), e3 (2018)Google Scholar
  16. 16.
    Introduction to Machine Learning with Tensorflow Homepage.: https://www.slideshare.net/PTomeo1/introduction-to-machine-learning-with-tensorflow. Last accessed 17 July 2018
  17. 17.
    Colab Homepage. https://colab.research.google.com/notebooks/welcome.ipynb. Last Accessed 15 May 2018
  18. 18.
    Albadra, M.A.A., Tiuna, S.: Extreme learning machine: a review. Int. J. Appl. Eng. Res. 12(14), 4610–4623 (2017)Google Scholar
  19. 19.
    López-Fandiño, J., Quesada-Barriuso, P., Heras, D.B., Argüello, F.: Efficient ELM-based techniques for the classification of hyperspectral remote sensing images on commodity GPUs. IEEE J. Sel. Top. Appl. Earth Obs. Remote Sens. 8(6), 2884–2893 (2015)CrossRefGoogle Scholar
  20. 20.
    Akusok, A., Björk, K.M., Miche, Y., Lendasse, A.: High-performance extreme learning machines: a complete toolbox for big data applications. IEEE Access 3, 1011–1025 (2015)CrossRefGoogle Scholar
  21. 21.
    Krawczyk, B., Minku, L.L., Gama, J., Stefanowski, J., Woźniak, M.: Ensemble learning for data stream analysis: a survey. Inf. Fusion 37, 132–156 (2017)CrossRefGoogle Scholar
  22. 22.
    Domingos, P., Hulten, G.: Mining high-speed data streams. In: Knowledge Discovery and Data Mining, pp. 71–80 (2000)Google Scholar
  23. 23.
    Bifet, A., Holmes, G., Kirkby, R.: Data Stream Mining: A Practical Approach, May 2011. http://jwijffels.github.io/RMOA/MOA_2014_04/doc/pdf/StreamMining.pdf
  24. 24.
    Stefanowski, J., Brzezinski, D.: Stream Classification. In: Sammut, C., Webb, G.I. (eds.) Encyclopedia of Machine Learning and Data Mining. Springer, Boston, MA (2017)Google Scholar
  25. 25.
    Stisen, A., Blunck, H., Bhattacharya, S., Prentow, T.S., Kjærgaard, M.B., Dey, A., Sonne, T., Jensen, M.M.: Smart devices are different: Assessing and mitigating mobile sensing heterogeneities for activity recognition. In: Proceedings of the 13th ACM Conference on Embedded Networked Sensor Systems. ACM (2015)Google Scholar
  26. 26.
    Bifet, A., Zhang, J., Fan, W., He, C., Zhang, J., Qian, J., Holmes, G., Pfahringer, B.: Extremely fast decision tree mining for evolving data streams. In Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, pp. 1733–1742. ACM (2017)Google Scholar
  27. 27.
    Bifet, A., Gavaldà, R.: Adaptive learning from evolving data streams. In 8th International Symposium on Intelligent Data Analysis, pp. 249–260 (2009)CrossRefGoogle Scholar
  28. 28.
    MOA Tutorial 1. https://moa.cms.waikato.ac.nz/tutorial-1-introduction-to-moa/. Last Accessed 15 May 2018
  29. 29.
    Bifet, A., Zhang, J., Fan, W., He, C., Zhang, J., Qian, J., Holmes, G., Pfahringer, B.: Extremely fast decision tree mining for evolving data streams. In Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, pp. 1733–1742. ACM (2017)Google Scholar
  30. 30.
    Bifet, A., Gavaldà, R.: Adaptive learning from evolving data streams. In 8th International Symposium on Intelligent Data Analysis, pp. 249–260 (2009)CrossRefGoogle Scholar
  31. 31.
    Nakano K., Chakraborty B.: Effect of dynamic feature for human activity recognition using smartphone sensors. In: IEEE 8th International Conference on Awareness Science and Technology (iCAST), pp. 539–543, 8 November 2017. IEEEGoogle Scholar
  32. 32.
    He, K., Sun, J.: Convolutional neural networks at constrained time cost. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 5353–5360 (2015)Google Scholar

Copyright information

© Springer Nature Switzerland AG 2019

Authors and Affiliations

  • Ilham Amezzane
    • 1
    Email author
  • Youssef Fakhri
    • 1
  • Mohamed El Aroussi
    • 1
  • Mohamed Bakhouya
    • 2
  1. 1.LaRIT Lab, Faculty of SciencesIbn Tofail UniversityKenitraMorocco
  2. 2.Faculty of Computer and Logistics, TIC Lab Sala AljadidaInternational University of RabatRabatMorocco

Personalised recommendations