Deep Residual Bidir-LSTM for Human Activity Recognition Using Wearable Sensors

• Published in: Mathematical problems in engineering Vol. 2018; no. 2018; pp. 1 - 13
• Main Authors: Xu, Ximeng, Chevalier, Guillaume, Yang, Rennong, Zhao, Yu, Zhang, Zhenxing
• Format: Journal Article
• Language: English
• Published: Cairo, Egypt Hindawi Publishing Corporation 01.01.2018; Hindawi; John Wiley & Sons, Inc
• Subjects: Accuracy; Algorithms; Big Data; Datasets; Human activity recognition; International conferences; Machine learning; Natural language processing; Neural networks; Public domain; Sensors; Time series; Ubiquitous computing; Voice recognition
• ISSN: 1024-123X; 1563-5147
• DOI: 10.1155/2018/7316954

Human activity recognition (HAR) has become a popular topic in research because of its wide application. With the development of deep learning, new ideas have appeared to address HAR problems. Here, a deep network architecture using residual bidirectional long short-term memory (LSTM) is proposed. The advantages of the new network include that a bidirectional connection can concatenate the positive time direction (forward state) and the negative time direction (backward state). Second, residual connections between stacked cells act as shortcut for gradients, effectively avoiding the gradient vanishing problem. Generally, the proposed network shows improvements on both the temporal (using bidirectional cells) and the spatial (residual connections stacked) dimensions, aiming to enhance the recognition rate. When testing with the Opportunity dataset and the public domain UCI dataset, the accuracy is significantly improved compared with previous results.