Home > Archive > 2018 > Volume 8 Number 6 (Dec. 2018) >
IJMLC 2018 Vol.8(6): 577-582 ISSN: 2010-3700
DOI: 10.18178/ijmlc.2018.8.6.748

Comparative Study of Machine Learning and Deep Learning Architecture for Human Activity Recognition Using Accelerometer Data

Sarbagya Ratna Shakya, Chaoyang Zhang, and Zhaoxian Zhou

Abstract—Human activity recognition (HAR) has been a popular fields of research in recent times. Many approaches have been implemented in literature with the aim of recognizing and analyzing human activity. Classical machine learning approaches use hand-crafted feature extraction and are based on classification technique, however of late, deep learning approaches have shown greater success in recognition accuracy with increased performance. With the current, wide popularity of mobile phones and various sensors such as accelerometers, gyroscopes, and cameras that are already installed on mobile phones, the activity recognition using the accumulating data from mobile phones has been a significant area of research in HAR. In this paper, we investigate the HAR based on the data collected through the accelerometer sensor of mobile devices. We employ different machine learning (ML) classifiers, algorithms, and deep learning (DL) models across different benchmark datasets. The experimental results from this study provide a comparative performance analysis based on accuracy, performance, and the costs of different ML algorithms and DL algorithms, based on recurrent neural network (RNN) and convolutional neural network (CNN) models for activity recognition.

Index Terms—ML, DL, CNN, RNN.

S. R. Shakya, C. Zhang, Z. Zhou are with the School of Computing Sciences and Computer Engineering, University of Southern Mississippi, Hattiesburg, MS 39406 USA (email: sarbagya.shakya@usm.edu, Chaoyang.zhang@usm.edu, zhaoxian.zhou@usm.edu).

[PDF]

Cite: Sarbagya Ratna Shakya, Chaoyang Zhang, and Zhaoxian Zhou, "Comparative Study of Machine Learning and Deep Learning Architecture for Human Activity Recognition Using Accelerometer Data," International Journal of Machine Learning and Computing vol. 8, no. 6, pp. 577-582, 2018.

General Information

  • E-ISSN: 2972-368X
  • Abbreviated Title: Int. J. Mach. Learn.
  • Frequency: Quaterly
  • DOI: 10.18178/IJML
  • Editor-in-Chief: Dr. Lin Huang
  • Executive Editor:  Ms. Cherry L. Chen
  • Abstracing/Indexing: Inspec (IET), Google Scholar, Crossref, ProQuest, Electronic Journals LibraryCNKI.
  • E-mail: ijml@ejournal.net


Article Metrics in Dimensions