Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Video

Updated: May 17, 2026

Artificial Intelligence-Based System for Detecting Attention Levels in Students
06:37

Artificial Intelligence-Based System for Detecting Attention Levels in Students

Published on: December 15, 2023

Comparing supervised learning techniques on the task of physical activity recognition.

A Dalton, G OLaighin

    IEEE Journal of Biomedical and Health Informatics
    |October 17, 2012
    PubMed
    Summary
    This summary is machine-generated.

    Related Concept Videos

    You might also read

    Related Articles

    Articles linked to this work by shared authors, journal, and citation graph.

    Sort by
    Same author

    AFM and Raman study of graphene deposited on silicon surfaces nanostructured by ion beam irradiation.

    Journal of microscopy·2020
    Same author

    Tracheal tube impingement during oral fibreoptic intubation. Comment on Br J Anaesth 2018; 120: 1139-40.

    British journal of anaesthesia·2018
    Same author

    De novo intracerebral arteriovenous malformations and a review of the theories of their formation.

    British journal of neurosurgery·2018
    Same author

    A revised method for the preparation of dry bone samples used in histological examination: Five simple steps.

    Homo : internationale Zeitschrift fur die vergleichende Forschung am Menschen·2017
    Same author

    Radiation exposure in pediatric patients during micturating cystourethrography procedures.

    Applied radiation and isotopes : including data, instrumentation and methods for use in agriculture, industry and medicine·2016
    Same author

    Assessment of patient dose and radiogenic risks during endoscopic retrograde cholangiopancreatography.

    Applied radiation and isotopes : including data, instrumentation and methods for use in agriculture, industry and medicine·2016

    This study shows that a meta-level classifier (AdaBoostM1 with C4.5 Graft) accurately recognizes physical activities (95%) using wireless sensors, even without user-specific training.

    Area of Science:

    • Biomedical Engineering
    • Machine Learning
    • Wearable Technology

    Background:

    • Physical activity recognition is crucial for health monitoring and rehabilitation.
    • Wireless kinematic sensors offer a non-invasive method for capturing human movement data.
    • Developing accurate and efficient classification algorithms is essential for practical applications.

    Purpose of the Study:

    • To compare the performance of base-level and meta-level classifiers for physical activity recognition.
    • To evaluate the effectiveness of different feature extraction techniques.
    • To determine the feasibility of user-independent physical activity recognition.

    Main Methods:

    • Utilized five wireless kinematic sensors on 25 subjects performing various physical activities.

    More Related Videos

    Evaluation of a Smartphone-based Human Activity Recognition System in a Daily Living Environment
    06:49

    Evaluation of a Smartphone-based Human Activity Recognition System in a Daily Living Environment

    Published on: December 11, 2015

    Related Experiment Videos

    Last Updated: May 17, 2026

    Artificial Intelligence-Based System for Detecting Attention Levels in Students
    06:37

    Artificial Intelligence-Based System for Detecting Attention Levels in Students

    Published on: December 15, 2023

    Evaluation of a Smartphone-based Human Activity Recognition System in a Daily Living Environment
    06:49

    Evaluation of a Smartphone-based Human Activity Recognition System in a Daily Living Environment

    Published on: December 11, 2015

  • Extracted time-domain and frequency-domain features from sensor data.
  • Employed a wrapper subset evaluation with linear forward search for feature selection.
  • Compared base-level and meta-level classifiers, including AdaBoostM1 with C4.5 Graft.

    • Main Results:

    • The AdaBoostM1 meta-level classifier achieved 95% overall accuracy in physical activity recognition.
    • High recognition rates were obtained using both subject-independent and subject-dependent data, negating the need for user-specific training.
    • An 88% accuracy was achieved using data solely from ankle and wrist sensors.

    Conclusions:

    • Meta-level classifiers, specifically AdaBoostM1 with C4.5 Graft, demonstrate superior performance in physical activity recognition.
    • Accurate physical activity recognition is achievable without user-specific training data, enhancing usability.
    • Sensor data from key body locations (ankle and wrist) can yield significant recognition accuracy.