Publications

• Sorted by Date • Classified by Publication Type • Classified by Research Category •

Sensor Data for Human Activity Recognition: Feature Representation and Benchmarking

Flavia Alves, Martin Gairing, Frans A. Oliehoek, and Thanh-Toan Do. Sensor Data for Human Activity Recognition: Feature Representation and Benchmarking. In International Joint Conference on Neural Networks (IJCNN), pp. 1–8, 2020.
Also see arXiv version.

Download

pdf [521.6kB]  ps.gz ps HTML 

Abstract

The field of Human Activity Recognition (HAR) focuses on obtaining and analysing data captured from monitoring devices (e.g. sensors). There is a wide range of applications within the field; for instance, assisted living, security surveillance, and intelligent transportation. In HAR, the development of Activity Recognition models is dependent upon the data captured by these devices and the methods used to analyse them, which directly affect performance metrics. In this work, we address the issue of accurately recognising human activities using different Machine Learning (ML) techniques. We propose a new feature representation based on consecutive occurring observations and compare it against previously used feature representations using a wide range of classification methods. Experimental results demonstrate that techniques based on the proposed representation outperform the baselines and a better accuracy was achieved for both highly and less frequent actions. We also investigate how the addition of further features and their pre-processing techniques affect performance results leading to state-of-the-art accuracy on a Human Activity Recognition dataset.

BibTeX Entry

@inproceedings{Alves20IJCNN,
    title =     {Sensor Data for Human Activity Recognition: Feature Representation and Benchmarking},
    author =    {Fl{a}via Alves and Martin Gairing and Frans A. Oliehoek and Thanh-Toan Do},
    booktitle = {International Joint Conference on Neural Networks (IJCNN)},
    year =      {2020},
    wwwnote =   {Also see <a href="https://arxiv.org/abs/2005.07308">arXiv version</a>.},
    pages =     {1--8},
    doi =       {10.1109/IJCNN48605.2020.9207068},
    url =       {https://research.tudelft.nl/files/84381506/09207068.pdf},
    keywords =   {refereed},
    abstract =  {
    The field of Human Activity Recognition (HAR) focuses on obtaining and
    analysing data captured from monitoring devices (e.g. sensors). There is
    a wide range of applications within the field; for instance, assisted
    living, security surveillance, and intelligent transportation. In HAR,
    the development of Activity Recognition models is dependent upon the data
    captured by these devices and the methods used to analyse them, which
    directly affect performance metrics. In this work, we address the issue of
    accurately recognising human activities using different Machine Learning
    (ML) techniques. We propose a new feature representation based on
    consecutive occurring observations and compare it against previously used
    feature representations using a wide range of classification methods.
    Experimental results demonstrate that techniques based on the proposed
    representation outperform the baselines and a better accuracy was achieved
    for both highly and less frequent actions. We also investigate how the
    addition of further features and their pre-processing techniques affect
    performance results leading to state-of-the-art accuracy on a Human
    Activity Recognition dataset.    
    }
}

Generated by bib2html.pl (written by Patrick Riley) on Tue Nov 05, 2024 16:13:37 UTC