An Information Gain-Based Model and an Attention-Based RNN for Wearable Human Activity Recognition.

Leyuan Liu,Ruihai Dong,Yibin Hou,Jian He,Jonathan Lungu,Keyan Ren

doi:10.3390/e23121635

Abstract

Wearable sensor-based HAR (human activity recognition) is a popular human activity perception method. However, due to the lack of a unified human activity model, the number and positions of sensors in the existing wearable HAR systems are not the same, which affects the promotion and application. In this paper, an information gain-based human activity model is established, and an attention-based recurrent neural network (namely Attention-RNN) for human activity recognition is designed. Besides, the attention-RNN, which combines bidirectional long short-term memory (BiLSTM) with attention mechanism, was tested on the UCI opportunity challenge dataset. Experiments prove that the proposed human activity model provides guidance for the deployment location of sensors and provides a basis for the selection of the number of sensors, which can reduce the number of sensors used to achieve the same classification effect. In addition, experiments show that the proposed Attention-RNN achieves F1 scores of 0.898 and 0.911 in the ML (Modes of Locomotion) task and GR (Gesture Recognition) task, respectively.

Highlights

Based on the past research and the human skeleton model proposed by Microsoft, this paper proposes the information gain-based human activity model
Attention-RNN was 0.898, which was over 3% higher than Random Forest [43] and was
Attention-RNN was 0.911, which was higher than Random Forest and CNN [44], but slightly lower than DeepConvLSTM

Summary

Introduction

Academic Editors: Felipe Ortega and Emilio López Cano. Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. Human activity recognition (HAR) technology [1] has been widely used in various areas, such as security monitoring [2], human-machine interaction [3], sports analysis [4], medical treatment [5], and health care [6], etc. According to the types of sensors used, HAR systems can be mainly divided into environmental sensor-based HAR, video-based. HAR, and wearable sensor-based HAR [7]. Environmental sensor-based HAR requires placing sensors in a fixed environment, which may cause certain limitations [8,9]

Methods

Results

Conclusion