Human Activity Recognition in WSN: A Comparative Study
Human activity recognition is an emerging field of ubiquitous and pervasive computing. Although recent smartphones have powerful resources, the execution of machine learning algorithms on a large amount of data is still a burden on smartphones. Three major factors including; classification algorithm...
Main Authors: | , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
Atlantis Press
2014-11-01
|
Series: | International Journal of Networked and Distributed Computing (IJNDC) |
Subjects: | |
Online Access: | https://www.atlantis-press.com/article/14324.pdf |
id |
doaj-b0ec358a69e14622b83a980d6165d358 |
---|---|
record_format |
Article |
spelling |
doaj-b0ec358a69e14622b83a980d6165d3582020-11-24T21:15:22ZengAtlantis PressInternational Journal of Networked and Distributed Computing (IJNDC)2211-79462014-11-012410.2991/ijndc.2014.2.4.3Human Activity Recognition in WSN: A Comparative StudyMuhammad Arshad AwanZheng GuangbinCheong-Ghil KimShin-Dug KimHuman activity recognition is an emerging field of ubiquitous and pervasive computing. Although recent smartphones have powerful resources, the execution of machine learning algorithms on a large amount of data is still a burden on smartphones. Three major factors including; classification algorithm, data feature, and smartphone position influence the recognition accuracy and time. In this paper, we present a comparative study of six classification algorithms, six data features, and four different positions that are most commonly used in the recognition process using smartphone accelerometer. This analysis can be used to select any specific classification algorithm, data feature, and smartphone position for human activity recognition in terms of accuracy and response time. The methodology we used is composed of two major components; a data collector, and a classifier. A set of eleven activities of daily living, four different positions for data collection and ten volunteers contributed to make it a worth-full comparative study. Results show that K-Nearest Neighbor and J48 algorithms performed well both in terms of time and accuracy irrespective of data features whereas the performance of other algorithms is dependent on the selected data features. Similarly, mean and mode features gave good results in terms of accuracy irrespective of the classification algorithm. A short version of the paper has already been presented at ICIS 2014.https://www.atlantis-press.com/article/14324.pdfActivity recognition; classification algorithm; data feature; smartphone position; ubiquitous computing |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Muhammad Arshad Awan Zheng Guangbin Cheong-Ghil Kim Shin-Dug Kim |
spellingShingle |
Muhammad Arshad Awan Zheng Guangbin Cheong-Ghil Kim Shin-Dug Kim Human Activity Recognition in WSN: A Comparative Study International Journal of Networked and Distributed Computing (IJNDC) Activity recognition; classification algorithm; data feature; smartphone position; ubiquitous computing |
author_facet |
Muhammad Arshad Awan Zheng Guangbin Cheong-Ghil Kim Shin-Dug Kim |
author_sort |
Muhammad Arshad Awan |
title |
Human Activity Recognition in WSN: A Comparative Study |
title_short |
Human Activity Recognition in WSN: A Comparative Study |
title_full |
Human Activity Recognition in WSN: A Comparative Study |
title_fullStr |
Human Activity Recognition in WSN: A Comparative Study |
title_full_unstemmed |
Human Activity Recognition in WSN: A Comparative Study |
title_sort |
human activity recognition in wsn: a comparative study |
publisher |
Atlantis Press |
series |
International Journal of Networked and Distributed Computing (IJNDC) |
issn |
2211-7946 |
publishDate |
2014-11-01 |
description |
Human activity recognition is an emerging field of ubiquitous and pervasive computing. Although recent smartphones have powerful resources, the execution of machine learning algorithms on a large amount of data is still a burden on smartphones. Three major factors including; classification algorithm, data feature, and smartphone position influence the recognition accuracy and time. In this paper, we present a comparative study of six classification algorithms, six data features, and four different positions that are most commonly used in the recognition process using smartphone accelerometer. This analysis can be used to select any specific classification algorithm, data feature, and smartphone position for human activity recognition in terms of accuracy and response time. The methodology we used is composed of two major components; a data collector, and a classifier. A set of eleven activities of daily living, four different positions for data collection and ten volunteers contributed to make it a worth-full comparative study. Results show that K-Nearest Neighbor and J48 algorithms performed well both in terms of time and accuracy irrespective of data features whereas the performance of other algorithms is dependent on the selected data features. Similarly, mean and mode features gave good results in terms of accuracy irrespective of the classification algorithm. A short version of the paper has already been presented at ICIS 2014. |
topic |
Activity recognition; classification algorithm; data feature; smartphone position; ubiquitous computing |
url |
https://www.atlantis-press.com/article/14324.pdf |
work_keys_str_mv |
AT muhammadarshadawan humanactivityrecognitioninwsnacomparativestudy AT zhengguangbin humanactivityrecognitioninwsnacomparativestudy AT cheongghilkim humanactivityrecognitioninwsnacomparativestudy AT shindugkim humanactivityrecognitioninwsnacomparativestudy |
_version_ |
1716745523329761280 |