Home // SENSORDEVICES 2018, The Ninth International Conference on Sensor Device Technologies and Applications // View article
Authors:
Rafael de Pinho André
Pedro Henrique Diniz
Hugo Fuks
Keywords: Clothes-based sensors; IoT (Internet of Things) devices; Mobile sensing applications
Abstract:
In this work, we present an analysis of the relevance of different sensor types for the recognition of activities of daily living based on foot movement and position. By conducting a comprehensive experiment with 12 diverse volunteers that resulted in about 1 million data samples, and employing a machine learning HAR (Human Activity Recognition) classifier developed for a 9-activity classes model, we were able to assess the impact of sensor selection on the activity recognition accuracy. Aiming at a replicable research, we provide full hardware information, system source code and a public domain dataset.
Pages: 169 to 174
Copyright: Copyright (c) IARIA, 2018
Publication date: September 16, 2018
Published in: conference
ISSN: 2308-3514
ISBN: 978-1-61208-660-6
Location: Venice, Italy
Dates: from September 16, 2018 to September 20, 2018