A Comparative Study of the Effect of Sensor Noise on Activity Recognition Models

  • Robert Ross
  • John Kelleher
Conference paper
Part of the Communications in Computer and Information Science book series (CCIS, volume 413)

Abstract

To provide a better understanding of the relative strengths of Machine Learning based Activity Recognition methods, in this paper we present a comparative analysis of the robustness of three popular methods with respect to sensor noise. Specifically we evaluate the robustness of Naive Bayes classifier, Support Vector Machine, and Random Forest based activity recognition models in three cases which span sensor errors from dead to poorly calibrated sensors. Test data is partially synthesized from a recently annotated activity recognition corpus which includes both interleaved activities and a range of both temporally long and short activities. Results demonstrate that the relative performance of Support Vector Machine classifiers over Naive Bayes classifiers reduces in noisy sensor conditions, but that overall the Random Forest classifier provides best activity recognition accuracy across all noise conditions synthesized in the corpus. Moreover, we find that activity recognition is equally robust across classification techniques with the relative performance of all models holding up under almost all sensor noise conditions considered.

Keywords

Activity Recognition Sensor Noise Support Vector Machines Naive Bayes Random Forests 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Geib, C.W., Steedman, M.: On natural language processing and plan recognition. In: International Joint Conference on Artificial Intelligence (2007)Google Scholar
  2. 2.
    Ye, J., Dobson, S., McKeever, S.: Situation identification techniques in pervasive computing: A review. Pervasive and Mobile Computing 8, 36–66 (2012)CrossRefGoogle Scholar
  3. 3.
    Bui, H., Venkatesh, S., West, G.A.W.: Policy recognition in the abstract hidden markov model. Journal of Artificial Intelligence Research 17, 451–499 (2002)MATHMathSciNetGoogle Scholar
  4. 4.
    Bui, H.: A general model for online probabilistic plan recognition. In: Proc. of the International Joint Conference on Artificial Intelligence (IJCAI), pp. 1309–1315 (2003)Google Scholar
  5. 5.
    Singla, G., Cook, D., Schmitter-Edgecombe, M.: Tracking activities in complex settings using smart environment technologies. International Journal of BioSciences, Psychiatry and Technology 1, 25–35 (2009)Google Scholar
  6. 6.
    Cook, D.: Learning setting-generalized activity models for smart spaces. IEEE Intelligent Systems 27, 32–38 (2012)CrossRefGoogle Scholar
  7. 7.
    Singla, G., Cook, D.J., Schmitter-Edgecombe, M.: Recognizing independent and joint activities among multiple residents in smart environments. Journal of Ambient Intelligence and Humanized Computing 1, 57–63 (2010)CrossRefGoogle Scholar
  8. 8.
    Charniak, E., Goldman, R.P.: A bayesian model of plan recognition. Artificial Intelligence 64, 53–79 (1993)CrossRefGoogle Scholar
  9. 9.
    Pynadath, D.V., Wellman, M.P.: Accounting for context in plan recognition, with application to traffic monitoring. In: Besnard, P., Hanks, S. (eds.) Proceedings of the 11th Conference on Uncertainty in Artificial Intelligence (UAI 1995), pp. 472–481. Morgan Kaufmann Publishers, San Francisco (1995)Google Scholar
  10. 10.
    Krishnan, N., Cook, D.: Activity recognition on streaming sensor data. In: Pervasive and Mobile Computing (2012)Google Scholar
  11. 11.
    Ross, R., Kelleher, J.D.: Accuracy and timeliness in ml based activity recognition. In: Proceedings of AAAI Workshop on Plan, Activity, and Intent Recognition 2013 (PAIR 2013), Seattle, USA (2013)Google Scholar
  12. 12.
    Chen, C., Das, B., Cook, D.J.: A data mining framework for activity recognition in smart environments. In: Proceedings of the 2010 Sixth International Conference on Intelligent Environments, IE 2010, pp. 80–83. IEEE Computer Society, Washington, DC (2010)CrossRefGoogle Scholar
  13. 13.
    Das, B., Cook, D.: Data mining challenges in automated prompting systems. In: Workshop on Interacting with Smart Objects (2011)Google Scholar
  14. 14.
    Stoia, L., Shockley, D.M., Byron, D.K., Fosler-Lussier, E.: Noun phrase generation for situated dialogs. In: Proceedings of the Fourth International Natural Language Generation Conference, pp. 81–88. Association for Computational Linguistics, Sydney (2006)CrossRefGoogle Scholar
  15. 15.
    Cook, D., Schmitter-Edgecombe, M.: Assessing the quality of activities in a smart environment. Methods of Information in Medicine 48, 480–485 (2009)CrossRefGoogle Scholar
  16. 16.
    Kipp, M.: ANVIL - A generic annotation tool for multimodal dialogue. In: Dalsgaard, P., Lindberg, B., Benner, H., Tan, Z.H. (eds.) INTERSPEECH, ISCA, pp. 1367–1370 (2001)Google Scholar

Copyright information

© Springer International Publishing 2013

Authors and Affiliations

  • Robert Ross
    • 1
  • John Kelleher
    • 1
  1. 1.Applied Intelligence Research Center, School of ComputingDublin Institute of TechnologyIreland

Personalised recommendations