Unsupervised Domain Adaptation for Human Activity Recognition

  • Paulo Barbosa
  • Kemilly Dearo Garcia
  • João Mendes-MoreiraEmail author
  • André C. P. L. F. de Carvalho
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 11314)


Human Activity Recognition has been primarily investigated as a machine learning classification task forcing it to handle with two main limitations. First, it must assume that the testing data has an equal distribution with the training sample. However, the inherent structure of an activity recognition systems is fertile in distribution changes over time, for instance, a specific person can perform physical activities differently from others, and even sensors are prone to misfunction. Secondly, to model the pattern of activities carried out by each user, a significant amount of data is needed. This is impractical especially in the actual era of Big Data with effortless access to public repositories. In order to deal with these problems, this paper investigates the use of Transfer Learning, specifically Unsupervised Domain Adaptation, within human activity recognition systems. The yielded experiment results reveal a useful transfer of knowledge and more importantly the convenience of transfer learning within human activity recognition. Apart from the delineated experiments, our work also contributes to the field of transfer learning in general through an exhaustive survey on transfer learning for human activity recognition based on wearables.


Human activity recognition Transfer learning Unsupervised domain adaptation 


  1. 1.
    Malbut-Shennan, K., Young, A.: The physiology of physical performance and training in old age. Coron. Artery Dis. 10(1), 37–42 (1999)CrossRefGoogle Scholar
  2. 2.
    Sun, B., Feng, J., Saenko, K.: Return of frustratingly easy domain adaptation. In: AAAI, vol. 6. no. 7 (2016)Google Scholar
  3. 3.
    Thrun, S., Pratt, L. (eds.) Learning to Learn. Springer, Heidelberg (2012)Google Scholar
  4. 4.
    Calatroni, A., Roggen, D., Tröster, G.: Automatic transfer of activity recognition capabilities between body-worn motion sensors: training newcomers to recognize locomotion. In: 8th International Conference on Networked Sensing Systems (2011)Google Scholar
  5. 5.
    Khan, M.A.A.H., Roy, N.: TransAct: transfer learning enabled activity recognition. In: 2017 IEEE International Conference on Pervasive Computing and Communications Workshops, pp. 545–550 (2017)Google Scholar
  6. 6.
    Rokni, S.A., Ghasemzadeh, H.: Autonomous training of activity recognition algorithms in mobile sensors: a transfer learning approach in context-invariant views. IEEE Trans. Mob. Comput. 17(8), 1764–1777 (2018)CrossRefGoogle Scholar
  7. 7.
    Krishnan, N.C.: A computational framework for wearable accelerometer-based. Dissertation, Ph.D. thesis, Arizona State University (2010)Google Scholar
  8. 8.
    Kurz, M., et al.: Real-time transfer and evaluation of activity recognition capabilities in an opportunistic system. In: The Third International Conference on Adaptive and Self-Adaptive Systems and Applications, pp. 73–78 (2011)Google Scholar
  9. 9.
    Roggen, D.: The adarc pattern analysis architecture for adaptive human activity recognition systems. J. Ambient. Intell. Hum. Ized Comput. 4(2), 169–186 (2013)CrossRefGoogle Scholar
  10. 10.
    Zhao, Z.: Cross-mobile elm based activity recognition. Int. J. Eng. Ind. 1(1), 30–38 (2010)Google Scholar
  11. 11.
    Deng, W.-Y., Zheng, Q.-H., Wang, Z.-M.: Cross-person activity recognition using reduced kernel extreme learning machine. Neural Netw. 53, 1–7 (2014)CrossRefGoogle Scholar
  12. 12.
    Zhao, Z., et al.: Cross-people mobile-phone based activity recognition. In: IJCAI, vol. 11, no. 3 (2011)Google Scholar
  13. 13.
    Chen, H., Cui, S., Li, S.: Application of Transfer Learning Approaches in Multimodal Wearable Human Activity Recognition. arXiv preprint arXiv:1707.02412 (2017)
  14. 14.
    Hachiya, H., Sugiyama, M., Ueda, N.: Importance-weighted least-squares probabilistic classifier for covariate shift adaptation with application to human activity recognition. Neurocomputing 80, 93–101 (2012)CrossRefGoogle Scholar
  15. 15.
    Diethe, T., Twomey, N., Flach, P.: Active transfer learning for activity recognition. In: European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning (2016)Google Scholar
  16. 16.
    Fallahzadeh, R., Ghasemzadeh, H.: Personalization without user interruption: boosting activity recognition in new subjects using unlabelled data. In: Proceedings of the 8th International Conference on Cyber-Physical Systems. ACM (2017)Google Scholar
  17. 17.
    Laasonen, K., Raento, M., Toivonen, H.: Adaptive on-device location recognition. In: Ferscha, A., Mattern, F. (eds.) Pervasive 2004. LNCS, vol. 3001, pp. 287–304. Springer, Heidelberg (2004). Scholar
  18. 18.
    Blanke, U., Schiele, B.: Remember and transfer what you have learned-recognizing composite activities based on activity spotting. In: 2010 International Symposium on Wearable Computers (ISWC). IEEE (2010)Google Scholar
  19. 19.
    Margolis, A.: A literature review of domain adaptation with unlabeled data. Technical report, pp. 1–42 (2011)Google Scholar
  20. 20.
    Pan, S.J.: Domain adaptation via transfer component analysis. IEEE Trans. Neural Netw. 22(2), 199–210 (2011)CrossRefGoogle Scholar
  21. 21.
    Pan, S.J., Kwok, J.T., Yang, Q.: Transfer learning via dimensionality reduction. In: AAAI, vol. 8 (2008)Google Scholar
  22. 22.
    Fernando, B., et al.: Subspace alignment for domain adaptation. arXiv preprint arXiv:1409.5241 (2014)
  23. 23.
    Loog, M.: Nearest neighbor-based importance weighting. In: 2012 IEEE International Workshop on Machine Learning for Signal Processing (MLSP). IEEE (2012)Google Scholar
  24. 24.
    Huang, J., et al.: Correcting sample selection bias by unlabeled data. In: Advances in Neural Information Processing Systems (2007)Google Scholar
  25. 25.
    Santoyo-Ramón, J.A., Casilari, E., Cano-García, J.M.: Analysis of a smartphone-based architecture with multiple mobility sensors for fall detection with supervised learning. Sensors 18(4), 1155 (2018)CrossRefGoogle Scholar
  26. 26.
    Reiss, A., Stricker, D.: Creating and benchmarking a new dataset for physical activity monitoring. In: Proceedings of the 5th International Conference on PErvasive Technologies Related to Assistive Environments. ACM (2012)Google Scholar
  27. 27.
    Zhu, X., Ghahramani, Z.: Learning from labeled and unlabeled data with label propagation, p. 1 (2002)Google Scholar
  28. 28.
    Zhou, D., et al.: Learning with local and global consistency. In: Advances in Neural Information Processing Systems (2004)Google Scholar
  29. 29.
    Demŝar, J.: Statistical comparisons of classifiers over multiple data sets. J. Mach. Learn. Res. 7(Jan), 1–30 (2006)Google Scholar

Copyright information

© Springer Nature Switzerland AG 2018

Authors and Affiliations

  1. 1.Faculty of EngineeringUniversity of PortoPortoPortugal
  2. 2.ICMCUniversity of São PauloSão PauloBrazil
  3. 3.LIAAD-INESC TECPortoPortugal
  4. 4.University of TwenteEnschedeNetherlands

Personalised recommendations