Advertisement

User-Centred Evaluation for Machine Learning

Chapter
Part of the Human–Computer Interaction Series book series (HCIS)

Abstract

Activity tracking wearables like Fitbit or mobile applications like Moves have seen immense growth in recent years. However, users often experience errors that occur in unexpected and inconsistent ways making it difficult for them to find a workaround and ultimately leading them to abandon the system. This is not too surprising given that intelligent systems typically design the modelling algorithm independent of the overall user experience. Furthermore, the user experience often takes a seamless design approach which hides nuanced aspects of the model leaving only the model’s prediction for the user to see. This prediction is presented optimistically meaning that the user is expected to assume that it is correct. To better align the design of the user experience with the development of the underlying algorithms we propose a validation pipeline based on user-centred design principles and usability standards for use in model optimisation, selection and validation. Specifically, we show how available user experience research can highlight the need for new evaluation criteria for models of activity and we demonstrate the use of a user-centred validation pipeline to select a modelling approach which best addresses the user experience as a whole.

References

  1. 1.
    Abdallah, Z.S., Gaber, M.M., Srinivasan, B., Krishnaswamy, S.: StreamAR: incremental and active learning with evolving sensory data for activity recognition. In: 2012 IEEE 24th International Conference on Tools with Artificial Intelligence, vol. 1, pp. 1163–1170 (2012)Google Scholar
  2. 2.
    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
  3. 3.
    Alemdar, H., van Kasteren, T., Ersoy, C.: Using active learning to allow activity recognition on a large scale. In: Ambient Intelligence, pp. 105–114 (2011)Google Scholar
  4. 4.
    Bao, L., Intille, S.: Activity recognition from user-annotated acceleration data. In: Pervasive Computing, pp. 1–17 (2004)Google Scholar
  5. 5.
    Chalmers, M.: Seamful design: showing the seams in wearable computing. In: Proceedings of IEE Eurowearable’03, vol. 2003, pp. 11–16. IEE (2003)Google Scholar
  6. 6.
    Chalmers, M., MacColl, I.: Seamful and seamless design in ubiquitous computing. In: Workshop at the Crossroads: The Interaction of HCI and Systems Issues in UbiComp, vol. 8 (2003)Google Scholar
  7. 7.
    Choe, E.K., Abdullah, S., Rabbi, M., Thomaz, E., Epstein, D.A., Cordeiro, F., Kay, M., Abowd, G.D., Choudhury, T., Fogarty, J., Lee, B., Matthews, M., Kientz, J.A.: Semi-automated tracking: a balanced approach for self-monitoring applications. IEEE Pervasive Comput. 16(1), 74–84 (2017)CrossRefGoogle Scholar
  8. 8.
    Cook, D., Feuz, K.D., Krishnan, N.C.: Transfer learning for activity recognition: a survey. Knowl. Inf. Syst. 36(3), 537–556 (2013)CrossRefGoogle Scholar
  9. 9.
    Garcia-Ceja, E., Brena, R.: Building personalized activity recognition models with scarce labeled data based on class similarities. Ubiquitous Computing and Ambient Intelligence. Sensing, Processing, and Using Environmental Information. Lecture Notes in Computer Science, pp. 265–276. Springer, Cham (2015)CrossRefGoogle Scholar
  10. 10.
    Harrison, D., Marshall, P., Bianchi-Berthouze, N., Bird, J.: Activity tracking: barriers, workarounds and customisation. In: Proceedings of the 2015 ACM International Joint Conference on Pervasive and Ubiquitous Computing, UbiComp ’15, pp. 617–621. New York, NY, USA (2015)Google Scholar
  11. 11.
    Kulesza, T., Burnett, M., Wong, W.K., Stumpf, S.: Principles of explanatory debugging to personalize interactive machine learning. In: Proceedings of the 20th International Conference on Intelligent User Interfaces, IUI ’15, pp. 126–137. ACM Press, New York (2015)Google Scholar
  12. 12.
    Kwapisz, J.R., Weiss, G.M., Moore, S.A.: Activity recognition using cell phone accelerometers. ACM SigKDD Explor. Newsl. 12(2), 74–82 (2011)CrossRefGoogle Scholar
  13. 13.
    Lane, N.D., Xu, Y., Lu, H., Hu, S., Choudhury, T., Campbell, A.T., Zhao, F.: Enabling large-scale human activity inference on smartphones using community similarity networks (csn). In: Proceedings of the 13th International Conference on Ubiquitous Computing, pp. 355–364. ACM, New York (2011)Google Scholar
  14. 14.
    Liu, R., Chen, T., Huang, L.: Research on human activity recognition based on active learning. In: 2010 International Conference on Machine Learning and Cybernetics, vol. 1, pp. 285–290 (2010)Google Scholar
  15. 15.
    Lockhart, J.W., Weiss, G.M.: The benefits of personalized smartphone-based activity recognition models. In: Proceedings of the 2014 SIAM International Conference on Data Mining, pp. 614–622. SIAM (2014)CrossRefGoogle Scholar
  16. 16.
    Lockhart, J.W., Weiss, G.M., Xue, J.C., Gallagher, S.T., Grosner, A.B., Pulickal, T.T.: Design considerations for the WISDM smart phone-based sensor mining architecture. In: Proceedings of the Fifth International Workshop on Knowledge Discovery from Sensor Data, pp. 25–33 (2011)Google Scholar
  17. 17.
    Longstaff, B., Reddy, S., Estrin, D.: Improving activity classification for health applications on mobile devices using active and semi-supervised learning. In: 2010 4th International Conference on Pervasive Computing Technologies for Healthcare, pp. 1–7 (2010)Google Scholar
  18. 18.
    Miu, T., Missier, P., Pltz, T.: Bootstrapping personalised human activity recognition models using online active learning. In: 2015 IEEE International Conference on Computer and Information Technology; Ubiquitous Computing and Communications; Dependable, Autonomic and Secure Computing; Pervasive Intelligence and Computing (CIT/IUCC/DASC/PICOM), pp. 1138–1147. IEEE (2015)Google Scholar
  19. 19.
    Patel, M.S., Asch, D.A., Volpp, K.G.: Wearable devices as facilitators, not drivers, of health behavior change. JAMA 313(5), 459–460 (2015)CrossRefGoogle Scholar
  20. 20.
    Pedregosa, F., Varoquaux, G., Gramfort, A., Michel, V., Thirion, B., Grisel, O., Blondel, M., Prettenhofer, P., Weiss, R., Dubourg, V., others: Scikit-learn: machine learning in Python. J. Mach. Learn. Res. 12, 2825–2830 (2011)Google Scholar
  21. 21.
    Settles, B.: Active learning literature survey. University of Wisconsin, Madison, vol. 52(55–66), p. 11 (2010)Google Scholar
  22. 22.
    Sharma, M., Bilgic, M.: Evidence-based uncertainty sampling for active learning. Data Min. Knowl. Discov. 31(1), 164–202 (2017)MathSciNetCrossRefGoogle Scholar
  23. 23.
    Shih, P.C., Han, K., Poole, E.S., Rosson, M.B., Carroll, J.M.: Use and adoption challenges of wearable activity trackers. In: iConference 2015 Proceedings (2015)Google Scholar
  24. 24.
    Stikic, M., Van Laerhoven, K., Schiele, B.: Exploring semi-supervised and active learning for activity recognition. In: 12th IEEE International Symposium on Wearable Computers (ISWC2008), pp. 81–88 (2008)Google Scholar
  25. 25.
    Weiser, M.: Some computer science issues in ubiquitous computing. Commun. ACM 36(7), 75–84 (1993)CrossRefGoogle Scholar
  26. 26.
    Weiss, G.M., Lockhart, J.W.: The impact of personalization on smartphone-based activity recognition. In: AAAI Workshop on Activity Context Representation: Techniques and Languages, pp. 98–104 (2012)Google Scholar
  27. 27.
    Yang, R., Shin, E., Newman, M.W., Ackerman, M.S.: When fitness trackers don’t ’fit’: end-user difficulties in the assessment of personal tracking device accuracy. In: Proceedings of the 2015 ACM International Joint Conference on Pervasive and Ubiquitous Computing, UbiComp ’15, pp. 623–634. New York, NY, USA (2015)Google Scholar
  28. 28.
    Yosinski, J., Clune, J., Bengio, Y., Lipson, H.: How transferable are features in deep neural networks? In: Advances in Neural Information Processing Systems, pp. 3320–3328 (2014)Google Scholar
  29. 29.
    Zhu, X., Goldberg, A.B.: Introduction to semi-supervised learning. Synth. Lect. Artif. Intell. Mach. Learn. 3(1), 1–130 (2009)CrossRefGoogle Scholar

Copyright information

© Springer International Publishing AG, part of Springer Nature 2018

Authors and Affiliations

  1. 1.Northwestern UniversityEvanstonUSA

Personalised recommendations