Showing 1–1 of 1 results for author: Ngo, T M T

SOS: A Shuffle Order Strategy for Data Augmentation in Industrial Human Activity Recognition

Authors: Anh Tuan Ha, Hoang Khang Phan, Thai Minh Tien Ngo, Anh Phan Truong, Nhat Tan Le

Abstract: In the realm of Human Activity Recognition (HAR), obtaining high quality and variance data is still a persistent challenge due to high costs and the inherent variability of real-world activities. This study introduces a generation dataset by deep learning approaches (Attention Autoencoder and conditional Generative Adversarial Networks). Another problem that data heterogeneity is a critical challe… ▽ More In the realm of Human Activity Recognition (HAR), obtaining high quality and variance data is still a persistent challenge due to high costs and the inherent variability of real-world activities. This study introduces a generation dataset by deep learning approaches (Attention Autoencoder and conditional Generative Adversarial Networks). Another problem that data heterogeneity is a critical challenge, one of the solutions is to shuffle the data to homogenize the distribution. Experimental results demonstrate that the random sequence strategy significantly improves classification performance, achieving an accuracy of up to 0.70 $\pm$ 0.03 and a macro F1 score of 0.64 $\pm$ 0.01. For that, disrupting temporal dependencies through random sequence reordering compels the model to focus on instantaneous recognition, thereby improving robustness against activity transitions. This approach not only broadens the effective training dataset but also offers promising avenues for enhancing HAR systems in complex, real-world scenarios. △ Less

Submitted 15 May, 2025; originally announced May 2025.