Wireless Mobile Communication and Healthcare. 8th EAI International Conference, MobiHealth 2019, Dublin, Ireland, November 14-15, 2019, Proceedings

Research Article

Labeling of Activity Recognition Datasets: Detection of Misbehaving Users

  • @INPROCEEDINGS{10.1007/978-3-030-49289-2_25,
        author={Alessio Vecchio and Giada Anastasi and Davide Coccomini and Stefano Guazzelli and Sara Lotano and Giuliano Zara},
        title={Labeling of Activity Recognition Datasets: Detection of Misbehaving Users},
        proceedings={Wireless Mobile Communication and Healthcare. 8th  EAI International Conference, MobiHealth 2019, Dublin, Ireland, November 14-15, 2019, Proceedings},
        proceedings_a={MOBIHEALTH},
        year={2020},
        month={6},
        keywords={Activity recognition Wearable device Machine learning},
        doi={10.1007/978-3-030-49289-2_25}
    }
    
  • Alessio Vecchio
    Giada Anastasi
    Davide Coccomini
    Stefano Guazzelli
    Sara Lotano
    Giuliano Zara
    Year: 2020
    Labeling of Activity Recognition Datasets: Detection of Misbehaving Users
    MOBIHEALTH
    Springer
    DOI: 10.1007/978-3-030-49289-2_25
Alessio Vecchio1,*, Giada Anastasi1, Davide Coccomini1, Stefano Guazzelli1, Sara Lotano1, Giuliano Zara1
  • 1: University of Pisa
*Contact email: alessio.vecchio@unipi.it

Abstract

Automatic recognition of user’s activities by means of wearable devices is a key element of many e-health applications, ranging from rehabilitation to monitoring of elderly citizens. Activity recognition methods generally rely on the availability of annotated training sets, where the traces collected using sensors are labelled with the real activity carried out by the user. We propose a method useful to automatically identify misbehaving users, i.e. the users that introduce inaccuracies during the labeling phase. The method is semi-supervised and detects misbehaving users as anomalies with respect to accurate ones. Experimental results show that misbehaving users can be detected with more than 99% accuracy.