ADVANCED ENSEMBLE CLASSIFICATION MODEL FOR HUMAN PHYSIOLOGICAL CONDITION PREDICTION
DOI:
https://doi.org/10.29121/shodhkosh.v5.i1.2024.2754Keywords:
Activity Recognition, Classification, Feature Reduction, HAR, Machine Learning, Predictive Analytics, Support Vector MachineAbstract [English]
Human Activity Recognition (HAR) is an essential area of research, often approached through time series classification. Traditional HAR studies focus on basic behaviors, such as walking, sitting, and running. In contrast, this work seeks to predict more complex human physiological states—emotional, mental, physical, and neutral—based on sensor data obtained from healthcare devices, including ECG, TEB, and EDA. We employed three classifiers—Support Vector Machine (SVM), Naïve Bayes (NB), and k-Nearest Neighbors (k-NN)—to predict these conditions, with SVM and k-NN yielding the most accurate results. To enhance accuracy, an Optimized Ensemble Classifier (OEC) combining SVM and k-NN is proposed, achieving a 93% accuracy rate.
References
A. Badawi, A. Al-Kabbany and H. Shaban, "Daily Activity Recognition using Wearable Sensors via Machine Learning and Feature Selection," 2018 13th International Conference on Computer Engineering and Systems (ICCES), Cairo, Egypt, 2018, pp. 75-79, doi: 10.1109/ICCES.2018.8639309. DOI: https://doi.org/10.1109/ICCES.2018.8639309
Wang, Aiguo, et al. “Evaluation of Random Forest for Complex Human Activity Recognition Using Wearable Sensors.” 2020 International Conference on Networking and Network Applications (NaNA), IEEE, Dec. 2020. Crossref, https://doi.org/10.1109/nana51271.2020.00060. DOI: https://doi.org/10.1109/NaNA51271.2020.00060
Badawi, Abeer A., et al. “Sensor Type, Axis, and Position-Based Fusion and Feature Selection for Multimodal Human Daily Activity Recognition in Wearable Body Sensor Networks.” Journal of Healthcare Engineering, vol. 2020, Hindawi Limited, June 2020, pp. 1–14. Crossref, https://doi.org/10.1155/2020/7914649. DOI: https://doi.org/10.1155/2020/7914649
Ahmed, Nadeem, et al. “Enhanced Human Activity Recognition Based on Smartphone Sensor Data Using Hybrid Feature Selection Model.” Sensors, vol. 20, no. 1, MDPI AG, Jan. 2020, p. 317. Crossref, https://doi.org/10.3390/s20010317. DOI: https://doi.org/10.3390/s20010317
Mohino-Herranz, Inma, et al. “Activity Recognition Using Wearable Physiological Measurements: Selection of Features From a Comprehensive Literature Study.” Sensors, vol. 19, no. 24, MDPI AG, Dec. 2019, p. 5524. Crossref, https://doi.org/10.3390/s19245524. DOI: https://doi.org/10.3390/s19245524
Badshah, Mustafa. Sensor - Based Human Activity Recognition Using Smartphones. San Jose State University Library. Crossref, https://doi.org/10.31979/etd.8fjc-drpn. DOI: https://doi.org/10.31979/etd.8fjc-drpn
Othman, N.A., Aydin, I. (2021). Challenges and limitations in human action recognition on unmanned aerial vehicles: A comprehensive survey. Traitement du Signal, Vol. 38, No. 5, pp. 1403-1411. https://doi.org/10.18280/ts.380515 DOI: https://doi.org/10.18280/ts.380515
Available from: https://www.analyticsvidhya.com/blog/2018/03/introduction-k-neighbours-algorith-clustering
Available from: https://www.javatpoint.com/machine-learning-naive-bayes-classifier
Available from: https://www.geeksforgeeks.org/support-vector-machine-algorithm
Available from: https://www.analyticsvidhya.com/blog/2017/09/understating-support-vector-machine-example-code
Available from: https://www.javatpoint.com/machine-learning-support-vector-machine-algorithm
Activity recognition using wearable physiological measurements. (2019). UCI Machine Learning Repository. https://doi.org/10.24432/C5RK6V
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2024 Thangapriya, Nancy Jasmine Goldena
This work is licensed under a Creative Commons Attribution 4.0 International License.
With the licence CC-BY, authors retain the copyright, allowing anyone to download, reuse, re-print, modify, distribute, and/or copy their contribution. The work must be properly attributed to its author.
It is not necessary to ask for further permission from the author or journal board.
This journal provides immediate open access to its content on the principle that making research freely available to the public supports a greater global exchange of knowledge.
