DOI: 10.62796/pijst.2024v1i603

  DOI URL: https://doi.org/10.62796/pijst.2024v1i603

Abstract- Human activity recognition (HAR) aims to enable computers to understand various human activities, such as walking, running, or dancing, by analyzing movement patterns. This technology has significant applications in fields like healthcare for monitoring elderly individuals and sports for tracking performance. Traditional HAR methods often struggle with complex and variable movements or large datasets. This paper explores the potential of deep learning to overcome these challenges by using neural networks that learn from examples and identify intricate patterns in data. Specifically, we investigate how convolutional neural networks (CNNs) and recurrent neural networks (RNNs) can process data from movement sensors to accurately recognize human activities. We also address the challenges of preparing data for analysis, selecting appropriate network architectures, and interpreting the results. This study highlights the transformative potential of deep learning in HAR, aiming to enhance the understanding of human movements and foster innovative applications across various domains.