Hsin-Hung Wu 1, 2, Chun-Chin Hsu 3*, Shun-Chang Lin4, I-Chieh Lin 3, Yu-Chen Tsai 5, Ming-Min Lo 6*

1Department of Business Administration, National Changhua University of Education, Changhua City 500034, Taiwan

2Faculty of Education, State University of Malang, Malang, East Java, Indonesia

3Department of Industrial Engineering and Management, Chaoyang University of Technology, Taichung City 413310, Taiwan

4Department of Industrial Engineering and management, National Taipei University of Technology, Taipei City 10608, Taiwan

5Program of Business Administration in Industrial Development, Department of Business Administration, Chaoyang University of Technology, Taichung City 413310, Taiwan

6Department of Finance, Chaoyang University of Technology, Taichung City 413310, Taiwan

 

Download Citation: |
Download PDF


ABSTRACT

Real-time human activity recognition (HAR) is garnering attention across various fields, such as healthcare, fitness and sports, security and surveillance, occupational safety, smart environments, and more. This is largely attributed to the rapid development of mobile devices, which enable users to record human activity signals using accelerometers. In this study, we found that the recognition rates were poor when tri-axial activity signals collected from accelerometers were directly fed into classifiers, including decision trees (DT), discriminant analysis (DA), logistic regression (LR), Naïve Bayes classifiers, support vector machines (SVM), ensemble learning (EL), and neural networks (NN). The recognition rates improved from 75% to 94% when the three-axis signals were transformed into statistical signal features (SSF). Despite the improvement in accuracy, the increase in the number of input variables from 3 to 66 has burdened the computation time. Furthermore, a higher recognition rate is needed to have an effective decision making. Therefore, this study develops a novel feature engineering method by using genetic algorithm (GA) and exponentially weighted moving average (EWMA). The EWMA is not only used to capture the characteristics of time sequences derived from the activity signals but also to eliminate redundant SSFs. GA is employed to optimize EWMA weights for each SSF. The results demonstrate that the Ensemble Bagged Trees classifier, using the proposed GA-optimized EWMA features, achieves a testing recognition rate of 95.2% with a prediction time of less than 0.01 s, making it suitable for the field of real-time HAR.


Keywords: Human activity recognition, Feature engineering, Statistical signal features, Exponentially weighted moving average, Genetic algorithm.


Share this article with your colleagues

 


REFERENCES

  1. Affandi, Y. 2023. Sentiment analysis of student on online lectured during covid–19 pandemic using k–means and naïve bayes classifier. Journal of Advances in Information Systems and Technology, 5(1), 38–49.

  2. Agarwal, A., Singh, R., Verma, K. 2019. Augmented machine learning ensemble extension model for social media health trends predictions. International Journal of Recent Technology and Engineering (IJRTE), 8(2S7), 482–486.

  3. Aljameel, S., Alabbad, D., Alzahrani, N., Alqarni, S., Alamoudi, F., Babili, L., Alshamrani, F. 2020. A sentiment analysis approach to predict an individual’s awareness of the precautionary procedures to prevent covid–19 outbreaks in Saudi Arabia. International Journal of Environmental Research and Public Health, 18(1), 218.

  4. Almutairi, M., Abubakar, S., Chiroma, H. 2022. Detecting elderly behaviors based on deep learning for healthcare: recent advances, methods, real–world applications and challenges. IEEE Access, 10, 69802–69821.

  5. An, N., Ding, H., Yang, J., Au, R., Ang, T. 2020. Deep ensemble learning for Alzheimer's disease classification. Journal of Biomedical Informatics, 105, 103411.

  6. Anguita, D., Ghio, A., Oneto, L., Parra, X., Reyes–Ortiz, J.L. 2013. A public domain dataset for human activity recognition using smartphones. The European Symposium on Artificial Neural Networks, 437–442.

  7. Bayat, A., Pomplun, M., Tran, D.A. 2014. A study on human activity recognition using accelerometer data from smartphones. Procedia Computer Science, 34, 450–457.

  8. Breiman, L., Friedman, J., Olshen, R., Stone, C. 1984. Classification and Regression Trees. Monterey, CA: Wadsworth & Brooks/Cole.

  9. Chandra, P., Varma, K., Srivastava, E.S., Agarwal, P., Gupta, M., Singh, S. 2023. Development and Implementation of an Intelligent Health Monitoring System using IoT and Advanced Machine Learning Techniques. Journal of Machine and Computing, 456–464.

  10. Chen, H., Wang, Q., Shen, Y. 2011. Decision tree support vector machine based on genetic algorithm for multi–class classification. Journal of Systems Engineering and Electronics, 22(2), 322–326.

  11. Ciucă, A., Moldovan, R., Pintea, S., Dumitrascu, D., Baban, A. 2020. Screeners vs. non–screeners for colorectal cancer among people over 50 years of age: factual and psychological discriminants. Journal of Gastrointestinal and Liver Diseases.

  12. Cortes, C., Vapnik, V. 1995. Support–vector networks. Machine Learning, 20(3), 273–297.

  13. Cover, T.M., Hart, P.E. 1967. Nearest neighbor pattern classification. IEEE Transactions on Information Theory, 13(1), 21–27.

  14. Cox, D.R. 1958. The regression analysis of binary sequences. Journal of the Royal Statistical Society: Series B (Methodological), 20(2), 215–242.

  15. Devarakonda, P., Božić, B. 2022. Particle swarm optimization of convolutional neural networks for human activity prediction.

  16. Devikanniga, D., Ramu, A., Haldorai, A. 2018. Efficient diagnosis of liver disease using support vector machine optimized with crows search algorithm. EAI Endorsed Transactions on Energy Web, 164177.

  17. Dengel, A., Sefen, B., Baumbach, S., Abdennadher, S. 2016. Human activity recognition: using sensor data of smartphones and smart watches. Eighth International Conference on Agents and Artificial Intelligence (ICAART), Rome, 2, 488–493.

  18. Dietterich, T.G. 2000. Ensemble methods in machine learning. International Workshop on Multiple Classifier Systems, 1857, 1–15.

  19. Domeniconi, C., Gunopulos, D., Peng, J. 2005. Large margin nearest neighbor classifiers. IEEE Transactions on Neural Networks, 16(4), 899–909.

  20. Duda, R.O., Hart, P.E. 1973. Pattern Classification and Scene Analysis. Wiley.

  21. Figo, D., Diniz, P.C., Ferreira, D.R., Cardoso, J.M.P. 2010. Preprocessing techniques for context recognition from accelerometer data. Personal and Ubiquitous Computing, 14(7), 645–662.

  22. Fisher, R.A. 1936. The use of multiple measurements in taxonomic problems. Annals of Eugenics, 7(2), 179–188.

  23. Gumaei, A., Hassan, M., Alelaiwi, A., AlSalman, H. 2019. A hybrid deep learning model for human activity recognition using multimodal body sensing data. IEEE Access, 7, 99152–99160.

  24. Hendriks, M., Tönis, T. M., Hoogendoorn, M., Scherder, E., van der Cammen, T. J. M., & Jaspers, M. W. M. 2019. Sensor-based human activity recognition using temporal decision trees. Sensors, 19(15), 3311. https://doi.org/10.3390/s19153311

  25. Haq, A., Li, J., Kumar, R., Ali, Z., Khan, I., Uddin, M., Agbley, B. 2022. Mcnn: a multi–level CNN model for the classification of brain tumors in IoT–healthcare system. Journal of Ambient Intelligence and Humanized Computing, 14(5), 4695–4706.

  26. Hossain, M., Muhammad, G. 2016. Healthcare big data voice pathology assessment framework. IEEE Access, 4, 7806–7815.

  27. Hosmer, D.W., Lemeshow, S., Sturdivant, R.X. 2013. Applied Logistic Regression. John Wiley & Sons.

  28. Hsu, H.H., Chu, C.T., Zhou, Y., Cheng, Z. 2015. Two phase activity recognition with smartphone sensors. International Conference on Network–Based Information Systems (NBiS), IEEE, Taipei, 611–615.

  29. Irfan, S., Anjum, N., Masood, N., Khattak, A., Ramzan, N. 2021. A novel hybrid deep learning model for human activity recognition based on transitional activities. Sensors, 21(24), 8227.

  30. Kabir, M. 2021. Adopting Andersen’s behavior model to identify factors influencing maternal healthcare service utilization in Bangladesh. PLoS ONE, 16(11), e0260502.

  31. Kaddi, S. 2023. Ensemble learning based health care claim fraud detection in an imbalance data environment. Indonesian Journal of Electrical Engineering and Computer Science, 32(3), 1686–1694.

  32. Khan, I., Afzal, S., Lee, J. 2022. Human activity recognition via hybrid deep learning based model. Sensors, 22(1), 323.

  33. Kose, M., Incel, O.D., Ersoy, C. 2012. Online human activity recognition on smartphones. Second International Workshop on Mobile Sensing: From Smartphones and Wearables to Big Data, ACM, Beijing, 11–15.

  34. Kwapisz, J.R., Weiss, G.M., Moore, S.A. 2011. Activity recognition using cell phone accelerometers. ACM SIGKDD Explorations Newsletter, 12(2), 72–82.

  35. Lenka, S., Bisoy, S., Priyadarshini, R. 2023. Multiple optimized ensemble learning for high–dimensional imbalanced credit scoring datasets.

  36. Li, D., Fan, S. 2014. A modified decision tree algorithm based on genetic algorithm for mobile user classification problem. The Scientific World Journal, 1–11.

  37. McCulloch, W.S., Pitts, W. 1943. A logical calculus of the ideas immanent in nervous activity. The Bulletin of Mathematical Biophysics, 5(4), 115–133.

  38. Miluzzo, E., Lane, N.D., Fodor, K., Peterson, R., Lu, H., Musolesi, M., Eisenman, S.B., Zheng, X., Cambel, A.T. 2008. Sensing meets mobile social networks: the design, implementation and evaluation of the CenceMe application. ACM Conference on Embedded Network Sensor Systems (SenSys), ACM, Raleigh, 337–350.

  39. Ortiz, R., Luis, J. 2015. Smartphone–based human activity recognition. Springer, London, 59–77.

  40. Park, J., Lin, C., Delaney, C., Westra, B. 2019. Knowledge discovery with machine learning for hospital–acquired catheter–associated urinary tract infections. CIN Computers Informatics Nursing, 38(1), 28–35.

  41. Purushotham, S., Meng, C., Che, Z., Liu, Y. 2018. Benchmarking deep learning models on large healthcare datasets. Journal of Biomedical Informatics, 83, 112–134.

  42. Punn, N. 2024. Ensemble meta–learning using SVM for improving cardiovascular disease risk prediction.

  43. Quinlan, J.R. 1986. Induction of decision trees. Machine Learning, 1(1), 81–106.

  44. Ramadhan, P., Tugiono, T., Nurarif, S. 2019. Expert system of detection defisiensi imun uses k–nearest neighbor method. The IJICS (International Journal of Informatics and Computer Science), 3(2), 41.

  45. Rivenbark, J., Ichou, M. 2020. Discrimination in healthcare as a barrier to care: experiences of socially disadvantaged populations in France from a nationally representative survey. BMC Public Health, 20(1).

  46. Sánchez, V., Skeie, N. 2018. Decision trees for human activity recognition in smart house environments.

  47. Sav, S., Bossuat, J., Troncoso–Pastoriza, J., Claassen, M., Hubaux, J. 2022. Privacy–preserving federated neural network learning for disease–associated cell classification.

  48. Sherchan, J., Fernandez, J., Qiao, S., Kruglanski, A., Forde, A. 2022. Perceived covid–19 threat, perceived healthcare system inequities, personal experiences of healthcare discrimination and their associations with covid–19 preventive behavioral intentions among college students in the U.S. BMC Public Health, 22(1).

  49. Shoaib, M., Scholten, H., Havinga, P.J.M. 2013. Towards physical activity recognition using smartphone sensors. Tenth IEEE International Conference on Ubiquitous Intelligence and Computing (UIC), IEEE, Vietri sul Mare, 80–87.

  50. Siirtola, P., Röning, H. 2012. Recognizing human activities user–independently on smartphones based on accelerometer data. International Journal of Interactive Multimedia and Artificial Intelligence, 1(5), 38–45.

  51. Siddiqui, T. 2024. Chronic obstructive pulmonary disease diagnosis with bagging ensemble learning and ANN classifiers. Engineering Technology & Applied Science Research, 14(3), 14741–14746.

  52. Susilawati, D.S., Riana, D. 2021. Optimization the Naive Bayes Classifier Method to diagnose diabetes Mellitus. IAIC Transactions on Sustainable Digital Innovation (ITSDI), 1(1), 78–86.

  53. Tintorer, D., Beneyto, S., Manresa, J., Torán–Monserrat, P., Jiménez–Zarco, A., Torrent–Sellens, J., Saigí–Rubió, F. 2015. Understanding the discriminant factors that influence the adoption and use of clinical communities of practice: the ECOPIH case. BMC Health Services Research, 15(1)

  54. Tiwari, V. 2023. Optimizing antibiotic prescriptions and infectious disease management in hospitals using neural networks. Journal of Advanced Zoology, 44(S–5), 1895–1903.

  55. Xiao, Z., Shi, Y., Xue, Y., Hu, F., Wu, Y. 2013. Research on the taxonomy of activity recognition based on inertial sensors. Advanced Materials Research, 823, 107–110. 7.

  56. Yin, X.Z. 2016. Leveraging smartphone sensor data for human activity recognition. Dissertation, The University of Western Ontario, Ontario.

  57. Zhang, M., Sawchuk, A.A. 2012. A feature selection based framework for human activity recognition using wearable multimodal sensors. Sixth International Conference on Body Area Networks (BodyNets), ACM, Beijing, 1036–1043.

  58. Zhou, Y., Cheng, Z. 2015. Two–phase activity recognition with smartphone sensors. International Conference on Network–Based Information Systems (NBiS), IEEE, Taipei, 611–615.


ARTICLE INFORMATION

Received: 2024-10-11
Revised: 2024-12-23
Accepted: 2025-06-30
Available Online: 2025-07-21


Cite this article:

Wu, H.H., Hsu, C.C., Lin, S.C., Lin, I.C., Tsai, Y.C., Lo, M.M. 2025. Developing a GA-optimized EWMA feature engineering method for real-time human activity recognition. International Journal of Applied Science and Engineering, 22, 2024366. https://doi.org/10.6703/IJASE.202506_22(2).004

  Copyright The Author(s). This is an open access article distributed under the terms of the Creative Commons Attribution License (CC BY 4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are cited.

Other people also read ...