Dynamic detection of window starting positions and its implementation within an activity recognition framework

Qin Ni, Timothy Patterson, Ian Cleland, Chris Nugent

Research output: Contribution to journalArticle

15 Citations (Scopus)

Abstract

Activity recognition is an intrinsic component of many pervasive computing and ambient intelligent solutions. This has been facilitated by an explosion of technological developments in the area of wireless sensor network, wearable and mobile computing. Yet, delivering robust activity recognition, which could be deployed at scale in a real world environment, still remains an active research challenge. Much of the existing literature to date has focused on applying machine learning techniques to pre-segmented data collected in controlled laboratory environments. Whilst this approach can provide valuable ground truth information from which to build recognition models, these techniques often do not function well when implemented in near real time applications. This paper presents the application of a multivariate online change detection algorithm to dynamically detect the starting position of windows for the purposes of activity recognition.
LanguageEnglish
Pages171-180
JournalJournal of Biomedical Informatics
Volume62
Early online date5 Jul 2016
DOIs
Publication statusE-pub ahead of print - 5 Jul 2016

Fingerprint

Controlled Environment
Explosions
Mobile computing
Ubiquitous computing
Research
Learning systems
Wireless sensor networks
Machine Learning

Keywords

  • Data segmentation
  • Activities of Daily Living (ADLs)
  • Change detection
  • Feature selection
  • Activity recognition

Cite this

@article{435a12823bdb4f47b7a0d9e9231fe59b,
title = "Dynamic detection of window starting positions and its implementation within an activity recognition framework",
abstract = "Activity recognition is an intrinsic component of many pervasive computing and ambient intelligent solutions. This has been facilitated by an explosion of technological developments in the area of wireless sensor network, wearable and mobile computing. Yet, delivering robust activity recognition, which could be deployed at scale in a real world environment, still remains an active research challenge. Much of the existing literature to date has focused on applying machine learning techniques to pre-segmented data collected in controlled laboratory environments. Whilst this approach can provide valuable ground truth information from which to build recognition models, these techniques often do not function well when implemented in near real time applications. This paper presents the application of a multivariate online change detection algorithm to dynamically detect the starting position of windows for the purposes of activity recognition.",
keywords = "Data segmentation, Activities of Daily Living (ADLs), Change detection, Feature selection, Activity recognition",
author = "Qin Ni and Timothy Patterson and Ian Cleland and Chris Nugent",
year = "2016",
month = "7",
day = "5",
doi = "10.1016/j.jbi.2016.07.005",
language = "English",
volume = "62",
pages = "171--180",
journal = "Journal of Biomedical Informatics",
issn = "1532-0464",
publisher = "Elsevier",

}

Dynamic detection of window starting positions and its implementation within an activity recognition framework. / Ni, Qin; Patterson, Timothy; Cleland, Ian; Nugent, Chris.

In: Journal of Biomedical Informatics, Vol. 62, 05.07.2016, p. 171-180.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Dynamic detection of window starting positions and its implementation within an activity recognition framework

AU - Ni, Qin

AU - Patterson, Timothy

AU - Cleland, Ian

AU - Nugent, Chris

PY - 2016/7/5

Y1 - 2016/7/5

N2 - Activity recognition is an intrinsic component of many pervasive computing and ambient intelligent solutions. This has been facilitated by an explosion of technological developments in the area of wireless sensor network, wearable and mobile computing. Yet, delivering robust activity recognition, which could be deployed at scale in a real world environment, still remains an active research challenge. Much of the existing literature to date has focused on applying machine learning techniques to pre-segmented data collected in controlled laboratory environments. Whilst this approach can provide valuable ground truth information from which to build recognition models, these techniques often do not function well when implemented in near real time applications. This paper presents the application of a multivariate online change detection algorithm to dynamically detect the starting position of windows for the purposes of activity recognition.

AB - Activity recognition is an intrinsic component of many pervasive computing and ambient intelligent solutions. This has been facilitated by an explosion of technological developments in the area of wireless sensor network, wearable and mobile computing. Yet, delivering robust activity recognition, which could be deployed at scale in a real world environment, still remains an active research challenge. Much of the existing literature to date has focused on applying machine learning techniques to pre-segmented data collected in controlled laboratory environments. Whilst this approach can provide valuable ground truth information from which to build recognition models, these techniques often do not function well when implemented in near real time applications. This paper presents the application of a multivariate online change detection algorithm to dynamically detect the starting position of windows for the purposes of activity recognition.

KW - Data segmentation

KW - Activities of Daily Living (ADLs)

KW - Change detection

KW - Feature selection

KW - Activity recognition

U2 - 10.1016/j.jbi.2016.07.005

DO - 10.1016/j.jbi.2016.07.005

M3 - Article

VL - 62

SP - 171

EP - 180

JO - Journal of Biomedical Informatics

T2 - Journal of Biomedical Informatics

JF - Journal of Biomedical Informatics

SN - 1532-0464

ER -