Improving the Collection and Understanding the Quality of Datasets for the Aim of Human Activity Recognition

  • Angelica PoliEmail author
  • Susanna Spinsante
  • Chris Nugent
  • Ian Cleland
Part of the Computer Communications and Networks book series (CCN)


In the last few decades, life expectancy has been increasing. This has resulted in a higher proportion of older adults and increased prevalence of chronic conditions, posing challenges facing care needs. A possible solution is to foster both the prevention and health-related re-education, supporting healthier lifestyle and facilitating independent living. To facilitate this, it is crucial to measure individual’s key health metrics. For instance, human activity recognition through sensors provides valuable information about an individual’s lifestyle. Some crucial decisions, among which the quality of data collection, strengthen the methodological approach. This chapter addresses how the quality of data may affect the recognition performance. Two datasets of daily activities were collected through a triaxial accelerometer placed on the subject’s dominant wrist. The first dataset was collected by 141 users, whereas the second one comprised semi-realistic activities executed by three individuals. Specifically, outcomes were based on a comparison of activity recognition performance of six machine learning classifiers. Results show that, firstly, a higher number of features may not improve the recognition rate. Secondly, one approach may be robust in a laboratory setting but not generalizable to real-world applications. Finally, a great variability may increase the generalization of classifiers for successful activity recognition.


Activity recognition Dataset quality Features selection Accelerometry 


  1. 1.
    Anguita D, Ghio A, Oneto L et al (2013) A public domain dataset for human activity recognition using smartphones, pp 24–26Google Scholar
  2. 2.
    Ariani A, Koesoema AP, Soegijoko S (2017) Innovative healthcare systems for the 21st centuryGoogle Scholar
  3. 3.
    Banos O, Galvez JM, Damas M et al (2014) Window size impact in human activity recognition. Sensors (Switzerland) 14:6474–6499. Scholar
  4. 4.
    Bao L, Intille SS (2004) Activity recognition from user-annotated acceleration data, pp 1–17. Scholar
  5. 5.
    Brox Ó (2013) The value of health care information exchange and interoperability. Atalante 74–83. Scholar
  6. 6.
    Bulling A, Blanke U, Schiele B (2014) A tutorial on human activity recognition using body-worn inertial sensors. ACM Comput Surv 1:1–33. Scholar
  7. 7.
    Chen L, Khalil I (2010) Activity recognition: approaches, practices and trends (Chap. 3)Google Scholar
  8. 8.
    Cleland I, Kikhia B, Nugent C, Boytsov A, Josef H, Synnes K et al (2013) Optimal placement of accelerometers for the detection of everyday activities, pp 9183–9200CrossRefGoogle Scholar
  9. 9.
    Cleland I, Donnelly MP, Nugent CD et al (2018) Collection of a diverse, naturalistic and annotated dataset for wearable activity recognition, pp 674–679.
  10. 10.
    Cook DJ, Krishnan NC (2015) Activity learning: discovering, recognizing, and predicting human behavior from sensor dataCrossRefGoogle Scholar
  11. 11.
    Guiry JJ, P Van De Ven, Nelson J (2014) Multi-sensor fusion for enhanced contextual awareness of everyday activities with ubiquitous devices. 5687–5701. Scholar
  12. 12.
    Gupta P, Dallas T (2014) Feature selection and activity recognition system using a single triaxial accelerometer. IEEE Trans Biomed Eng 61:1780–1786. Scholar
  13. 13.
    Hall AM (1999) Correlation-based feature selection for machine learning. University of WaikatoGoogle Scholar
  14. 14.
    John G, Kohavi R, Pfleger K (1994) Irrelevant features and the subset selection problem. Icml 121–129.
  15. 15.
    Khusainov R, Azzi D, Achumba IE, Bersch SD (2013) Real-time human ambulation, activity, and physiological monitoring: taxonomy of issues, techniques, applications, challenges and limitations. Sensors (Switzerland) 13:12852–12902. Scholar
  16. 16.
    Kim E, Helal S, Cook D (2010) Human activity recognition and pattern discovery. IEEE Pervasive Comput 9CrossRefGoogle Scholar
  17. 17.
    Kong Y, Fu Y (2018) Action recognition and human interaction. Hum Act Recognit Predict A Surv 13. Scholar
  18. 18.
    Kononenko I (1994) Estimating attributes: analysis and extensions of RELIEF. Mach Learn ECML-94 784:171–182. Scholar
  19. 19.
    Kou Z, Wu C (2018) Smartphone based operating behaviour modelling of agricultural machinery. IFAC-PapersOnLine 51:521–525. Scholar
  20. 20.
    Lara OD, Labrador MA (2013) A survey on human activity recognition using wearable sensors. IEEE Commun Surv Tutor 15:1192–1209. Scholar
  21. 21.
    Lester J, Choudhury T, Borriello G (2006) A practical approach to recognizing physical activities, pp 1–16Google Scholar
  22. 22.
    Mannini A, Intille SS, Rosenberger M et al (2013) Activity recognition using a single accelerometer placed at the wrist or ankle. 2193–2203. Scholar
  23. 23.
    Morales J, Akopian D (2017) Physical activity recognition by smartphones, a survey. Biocybern Biomed Eng 37:388–400. Scholar
  24. 24.
    Morales J, Akopian D, Agaian S (2014) Human activity recognition by smartphones regardless of device orientation human activity recognition by smartphones regardless of device orientation.
  25. 25.
    Nakamoto K, Konishi Y, Kondo K, Ishigaki H (1999) Detection of posture and motion by accelerometry: a validation study in ambulatory monitoring. IDC 1999—1999 Information, Decis Control Data Inf Fusion Symp Signal Process Commun Symp Decis Control Symp Proc 15:283–288.
  26. 26.
    Nugent C et al (2016) Improving the quality of user generated data sets for activity recognition. In: García C, Caballero-Gil P, Burmester M, Quesada-Arencibia A (eds) Ubiquitous computing and ambient intelligence. IWAAL 2016, AmIHEALTH 2016, UCAmI 2016. Lecture Notes in Computer Science, vol 10070. Springer, ChamGoogle Scholar
  27. 27.
    Preece SJ, Goulermas JY, Kenney LPJ et al (2009) Activity identification using body-mounted sensors—a review of classification techniques. Physiol Meas 30. Scholar
  28. 28.
    Randell C, Muller H (2000) Context awareness by analysing accelerometer data, pp 175–176Google Scholar
  29. 29.
    Twomey N, Diethe T, Fafoutis X et al (2018) A comprehensive study of activity recognition using accelerometers. Informatics 5:27. Scholar
  30. 30.
    Vaizman Y, Weibel NG, Lanckriet N (2017) Context recognition in-the-wild: unified model for multi-modal sensors and multi-label classification. PACM Interact Mob Wearable Ubiquitous Technol 1:1–22. Scholar
  31. 31.
    Vavilis S, Petković M, Zannone N (2012) Impact of ICT on home healthcare. IFIP Adv Inf Commun Technol 386 AICT:111–122. Scholar
  32. 32.
    Viet VQ, Thang HM, Choi D (2012) Balancing precision and battery drain in activity recognition on mobile phone. 1–2.
  33. 33.
    Vrigkas M, Nikou C, Kakadiaris IA (2015) A review of human activity recognition methods. Front Robot AI 2.
  34. 34.
    Weber DM, Kauffman RJ (2011) What drives global ICT adoption? Analysis and research directions. Electron Commer Res Appl 10:683–701. Scholar
  35. 35.
    Yang JY, Wang JS, Chen YP (2008) Using acceleration measurements for activity recognition: an effective learning algorithm for constructing neural classifiers. Pattern Recognit Lett 29(16):2213–2220. Scholar
  36. 36.
    Yiyan L, Fang Z, Wenhua S (2016) An hidden markov model based complex walking pattern recognition algorithm. In: 2016 fourth international conference on ubiquitous positioning, indoor navigation and location based services (UPINLBS), pp 223–229.

Copyright information

© Springer Nature Switzerland AG 2020

Authors and Affiliations

  • Angelica Poli
    • 1
    Email author
  • Susanna Spinsante
    • 1
  • Chris Nugent
    • 2
  • Ian Cleland
    • 2
  1. 1.Dipartimento di Ingegneria dell’InformazioneUniversità Politecnica delle MarcheAnconaItaly
  2. 2.School of ComputingUlster UniversityBelfastUK

Personalised recommendations