An ensemble machine learning-based intelligent system for human activity recognition using sensory data
Edited book (chapter)
Abdulla, Shahab, Diykh, Mohammed, Siuly, Siuly and Ali, Mumtaz. 2023. "An ensemble machine learning-based intelligent system for human activity recognition using sensory data." Sinha, G.R., Subudhi, Bidyadhar, Fan, Chih-Peng and Nisar, Humaira (ed.) Cognitive Sensing Technologies and Applications. Institution of Engineering and Technology (IET). pp. 119-130
Chapter Title | An ensemble machine learning-based intelligent system for human activity recognition using sensory data |
---|---|
Book Chapter Category | Edited book (chapter) |
ERA Publisher ID | 6656 |
Book Title | Cognitive Sensing Technologies and Applications |
Authors | Abdulla, Shahab, Diykh, Mohammed, Siuly, Siuly and Ali, Mumtaz |
Editors | Sinha, G.R., Subudhi, Bidyadhar, Fan, Chih-Peng and Nisar, Humaira |
Volume | 135 |
Page Range | 119-130 |
Series | IET Control, robotics and sensors series |
Chapter Number | 5 |
Number of Pages | 12 |
Year | 2023 |
Publisher | Institution of Engineering and Technology (IET) |
ISBN | 9781839536908 |
9781839536892 | |
Digital Object Identifier (DOI) | https://doi.org/10.1049/pbce135e_ch5 |
Web Address (URL) | https://digital-library.theiet.org/content/books/10.1049/pbce135e_ch5 |
Abstract | Human activity recognition (HAR) from sensor data has become an attractive research topic due to its application in areas such as healthcare, human-computer interaction and smart environments. Although several research works have been done on HAR, the current studies are not enough to produce optimised performance balancing efficiency and accuracy. To solve this issue, this chapter proposes a new hybrid scheme based on permutation entropy integrated with quadratic sample entropy and ensemble classifier. First, the data are segmented into intervals, then, a combination of entropy features is extracted and evaluated. The extracted features are fed to an ensemble machine learning algorithm to classify the features vector into different activities. The proposed model is tested on a publicly available sensor dataset. The results demonstrate that the proposed scheme is efficient to improve HAR while obtaining an accuracy between 95% and 96% and outperforms considerably the existing methods. © The Institution of Engineering and Technology 2023. All rights reserved. |
Keywords | Ensemble machine; HAR; Permutation entropy; Quadratic sample entropy |
Contains Sensitive Content | Does not contain sensitive content |
ANZSRC Field of Research 2020 | 461199. Machine learning not elsewhere classified |
Public Notes | Files associated with this item cannot be displayed due to copyright restrictions. |
Byline Affiliations | UniSQ College |
Victoria University |
Permalink -
https://research.usq.edu.au/item/z2779/an-ensemble-machine-learning-based-intelligent-system-for-human-activity-recognition-using-sensory-data
81
total views0
total downloads10
views this month0
downloads this month