The Impact of Transformed Data Representations on Sensor-Based Human Activity Recognition

• Laura Sthefany Colombo UNESP
• Yan Lucas Dias UNESP
• Caetano Mazzoni Ranieri UNESP

DOI: https://doi.org/10.5753/eniac.2026.30706

Resumo

Human Activity Recognition (HAR) aims to identify human activities from sensor data and has been extensively studied in recent years. However, the impact of different data representation strategies for inertial and ambient sensor signals remains insufficiently understood. This work investigates the use of feature engineering and transformation techniques, including MiniRocket and recurrence plots, in combination with machine learning models for HAR. The results show that MiniRocket consistently achieved the best performance on inertial datasets, and tree-based models benefited the most from feature engineering. Overall, the findings highlight the role of data representation in HAR and indicate that multimodal and highly imbalanced datasets remain challenging.

Referências

Akiba, T., Sano, S., Yanase, T., Ohta, T., and Koyama, M. (2019). Optuna: A next-generation hyperparameter optimization framework.

Anguita, D., Ghio, A., Oneto, L., Parra, X., Reyes-Ortiz, J. L., et al. (2013). A public domain dataset for human activity recognition using smartphones. In Esann, volume 3, pages 3–4.

Bertini, L., Montesh, T., and Rodrigues, D. (2020). Human activity recognition: An approach 2d cnn-lstm to sequential image representation. Journal of Intelligent and Robotic Systems.

Brery, A. F., Gallardo-Antolín, A., Fakhry, M., and González-Carrasco, I. (2026). Deep learning for human activity recognition: A comprehensive review of architectures, performance, and challenges across five sensory datasets. Expert Systems with Applications, 323:132415.

Chavarriaga, R., Sagha, H., Calatroni, A., Digumarti, S. T., Tröster, G., Millán, J. d. R., and Roggen, D. (2013). The opportunity challenge: A benchmark database for on-body sensor-based activity recognition. Pattern Recognition Letters, 34(15):2033–2042.

Chen, C., Jafari, R., and Kehtarnavaz, N. (2015). Utd-mhad: A multimodal dataset for human action recognition utilizing a depth camera and a wearable inertial sensor. In 2015 IEEE International conference on image processing (ICIP), pages 168–172. IEEE.

Chen, J., Sun, Y., and Sun, S. (2021a). Improving human activity recognition performance by data fusion and feature engineering. Sensors, 21(3):692.

Chen, K., Zhang, D., Yao, L., Guo, B., Yu, Z., and Liu, Y. (2021b). Deep learning for sensor-based human activity recognition: Overview, challenges, and opportunities. ACM Comput. Surv., 54(4).

Chen, T. and Guestrin, C. (2016). Xgboost: A scalable tree boosting system. In Proceedings of the 22nd acm sigkdd international conference on knowledge discovery and data mining, pages 785–794.

Dempster, A., Petitjean, F., and Webb, G. I. (2020). Rocket: Exceptionally fast and accurate time series classification using random convolutional kernels. Data Mining and Knowledge Discovery, 34:1454–1495.

Dempster, A., Schmidt, D. F., and Webb, G. I. (2021). Minirocket: A very fast (almost) deterministic transform for time series classification. In Proceedings of the 27th ACM SIGKDD conference on knowledge discovery & data mining, pages 248–257.

Demrozi, F., Pravadelli, G., Bihorac, A., and Rashidi, P. (2020). Human activity recognition using inertial, physiological and environmental sensors: A comprehensive survey. IEEE Access, 8:210816–210836.

Eckmann, J.-P., Kamphorst, S. O., and Ruelle, D. (1987). Recurrence plots of dynamical systems. Europhysics Letters, 4(9):973–977.

Hammerla, N. Y., Halloran, S., and Plötz, T. (2016). Deep, convolutional, and recurrent models for human activity recognition using wearables. In IJCAI, pages 1533–1540.

Hatami, N., Gavet, Y., and Debayle, J. (2018). Classification of time-series images using deep convolutional neural networks. In ICMV.

Kulsoom, F., Narejo, S., Mehmood, Z., Chaudhry, H. N., Butt, A., and Bashir, A. K. (2022). A review of machine learning-based human activity recognition for diverse applications. Neural Computing and Applications, 34(21):18289–18324.

Lew, C. H., Lim, K. M., Lee, C. P., and Lim, J. Y. (2023). Human activity classification using recurrence plot and residual network. In 2023 IEEE 11th Conference on Systems, Process & Control (ICSPC), pages 78–83.

Penatti, O. A. B. and Santos, M. F. S. (2017). Human activity recognition from mobile inertial sensors using recurrence plots. In ICIP, pages 2214–2218.

Ranieri, C. M., MacLeod, S., Dragone, M., Vargas, P. A., and Romero, R. A. F. (2021). Activity recognition for ambient assisted living with videos, inertial units and ambient sensors. Sensors, 21(3):768.

Reiss, A. and Stricker, D. (2012). Introducing a new benchmarked dataset for activity monitoring. In 2012 16th international symposium on wearable computers, pages 108–109. IEEE.

Wang, J., Chen, Y., Hao, S., Peng, X., and Hu, L. (2019). Deep learning for sensor-based activity recognition: A survey. Pattern recognition letters, 119:3–11.

Wang, Z., Yan, W., and Oates, T. (2017). Time series classification from scratch with deep neural networks: A strong baseline. In 2017 International joint conference on neural networks (IJCNN), pages 1578–1585. IEEE.

Zeng, M. et al. (2014). Convolutional neural networks for human activity recognition using mobile sensors. In MobiCASE, pages 197–205.