Human activity recognition in WSN

A comparative study

Muhammad Arshad Awan, Zheng Guangbin, Cheong Ghil Kim, Shin-Dug Kim

Research output: Contribution to journalArticle

2 Citations (Scopus)

Abstract

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.

Original languageEnglish
Pages (from-to)221-230
Number of pages10
JournalInternational Journal of Networked and Distributed Computing
Volume2
Issue number4
DOIs
Publication statusPublished - 2014 Nov 1

Fingerprint

Smartphones
Ubiquitous computing
Accelerometers
Learning algorithms
Learning systems
Classifiers

All Science Journal Classification (ASJC) codes

  • Computer Science Applications
  • Computer Networks and Communications

Cite this

Awan, Muhammad Arshad ; Guangbin, Zheng ; Kim, Cheong Ghil ; Kim, Shin-Dug. / Human activity recognition in WSN : A comparative study. In: International Journal of Networked and Distributed Computing. 2014 ; Vol. 2, No. 4. pp. 221-230.
@article{112545cfcc2c4bbf8087f1c04b8d1c07,
title = "Human activity recognition in WSN: A comparative study",
abstract = "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.",
author = "Awan, {Muhammad Arshad} and Zheng Guangbin and Kim, {Cheong Ghil} and Shin-Dug Kim",
year = "2014",
month = "11",
day = "1",
doi = "10.2991/ijndc.2014.2.4.3",
language = "English",
volume = "2",
pages = "221--230",
journal = "International Journal of Networked and Distributed Computing",
issn = "2211-7938",
publisher = "Atlantis Press",
number = "4",

}

Human activity recognition in WSN : A comparative study. / Awan, Muhammad Arshad; Guangbin, Zheng; Kim, Cheong Ghil; Kim, Shin-Dug.

In: International Journal of Networked and Distributed Computing, Vol. 2, No. 4, 01.11.2014, p. 221-230.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Human activity recognition in WSN

T2 - A comparative study

AU - Awan, Muhammad Arshad

AU - Guangbin, Zheng

AU - Kim, Cheong Ghil

AU - Kim, Shin-Dug

PY - 2014/11/1

Y1 - 2014/11/1

N2 - 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.

AB - 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.

UR - http://www.scopus.com/inward/record.url?scp=85018878744&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85018878744&partnerID=8YFLogxK

U2 - 10.2991/ijndc.2014.2.4.3

DO - 10.2991/ijndc.2014.2.4.3

M3 - Article

VL - 2

SP - 221

EP - 230

JO - International Journal of Networked and Distributed Computing

JF - International Journal of Networked and Distributed Computing

SN - 2211-7938

IS - 4

ER -