Exploring and Comparing Machine Learning Approaches for Predicting Mood Over Time

  • Ward van BredaEmail author
  • Johnno Pastor
  • Mark Hoogendoorn
  • Jeroen Ruwaard
  • Joost Asselbergs
  • Heleen Riper
Conference paper
Part of the Smart Innovation, Systems and Technologies book series (SIST, volume 60)


Mental health related problems are responsible for great sorrow for patients and social surrounding involved. The costs for society are estimated to be 2.5 trillion dollar worldwide. More detailed data about the mental states and behaviour is becoming available due to technological developments, e.g. using Ecological Momentary Assessments. Unfortunately this wealth of data is not utilized: data-driven predictive models for short-term developments could contribute to more personalized interventions, but are rarely seen. In this paper we study how modern machine learning techniques can contribute to better models for predicting short-term mood in the context of depression. The models are based on data obtained from an experiment among 27 participants. During the study frequent mood assessments were performed and usage and sensor data of the mobile phone was recorded. Results show that much can be improved before fine-grained mood prediction is useful within E-health applications. Subsequently important next steps are identified.


Support Vector Machine Random Forest Support Vector Machine Model Dynamic Time Warping Random Forest Model 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


  1. 1.
    Insel, T.R., Collins, P.Y., Hyman, S.E.: Darkness invisible: The hidden global costs of mental illness. Foreign Aff. 94, 127 (2015)Google Scholar
  2. 2.
    LiKamWa, R., Liu, Y., Lane, N.D., Zhong, L.: Moodscope: building a mood sensor from smartphone usage patterns. In: Proceeding of the 11th Annual International Conference on Mobile Systems, Applications, and Services, MobiSys ’13, pp. 389–402. ACM, New York (2013)Google Scholar
  3. 3.
    Asselbergs, J., Ruwaard, J., Ejdys, M., Schrader, N., Sijbrandij, M., Riper, H.: Smartphone-based unobtrusive ecological momentary assessment of day-to-day mood: an explorative study. J. Med. Internet Res. (forthcoming)Google Scholar
  4. 4.
    Kessler, R.C., McGonagle, K.A., Zhao, S., Nelson, C.B., Hughes, M., Eshleman, S., Wittchen, H.-U., Kendler, K.S.: Lifetime and 12-month prevalence of DSM-III-R psychiatric disorders in the united states: results from the national comorbidity survey. Arch. Gen. Psychiatry 51(1), 8–19 (1994)CrossRefGoogle Scholar
  5. 5.
    Murray, C.J., Lopez, A.D.: Alternative projections of mortality and disability by cause 1990–2020: global burden of disease study. Lancet 349(9064), 1498–1504 (1997)CrossRefGoogle Scholar
  6. 6.
    R Core Team. R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna, Austria (2014)Google Scholar
  7. 7.
    Hyndman, R.J., Khandakar, Y.: Automatic time series forecasting: the forecast package for R. J. Stat. Softw. 26(3), 1–22 (2008)Google Scholar
  8. 8.
    Giorgino, T.: Computing and visualizing dynamic time warping alignments in R: the DTW package. J. Stat. Softw. 31(7), 1–24 (2009)CrossRefGoogle Scholar
  9. 9.
    Tormene, P., Giorgino, T., Quaglini, S., Stefanelli, M.: Matching incomplete time series with dynamic time warping: an algorithm and an application to post-stroke rehabilitation. Artif. Intell. Med. 45(1), 11–34 (2008)CrossRefGoogle Scholar
  10. 10.
    Montero, P., Vilar, J.A.: TSclust: an R package for time series clustering. J. Stat. Softw. 62(1), 1–43 (2014)CrossRefGoogle Scholar
  11. 11.
    Zeileis, A., Hornik, K., Smola, A., Karatzoglou, A.: Kernlab-an S4 package for kernel methods in R. J. Stat. Softw. 11(9), 1–20 (2004)Google Scholar
  12. 12.
    Liaw, A., Wiener, M.: Classification and regression by randomforest. R News 2(3), 18–22 (2002)Google Scholar
  13. 13.
    Karatzoglou, A., Smola, A., Hornik, K., Zeileis, A.: kernlab—an S4 package for kernel methods in R. J. Stat. Softw. 11(9), 1–20 (2004)CrossRefGoogle Scholar

Copyright information

© Springer International Publishing Switzerland 2016

Authors and Affiliations

  • Ward van Breda
    • 1
    Email author
  • Johnno Pastor
    • 1
  • Mark Hoogendoorn
    • 1
  • Jeroen Ruwaard
    • 2
  • Joost Asselbergs
    • 2
  • Heleen Riper
    • 2
  1. 1.Department of Computer ScienceVU University AmsterdamAmsterdamThe Netherlands
  2. 2.Department of Clinical PsychologyVU University AmsterdamAmsterdamThe Netherlands

Personalised recommendations