Clustering-Based Ensemble Learning for Activity Recognition in Smart Homes

Research output: Contribution to journalArticle

19 Citations (Scopus)

Abstract

Application of sensor-based technology within activity monitoring systems is becoming a popular technique within the smart environment paradigm. Nevertheless, the use of such an approach generates complex constructs of data, which subsequently requires the use of intricate activity recognition techniques to automatically infer the underlying activity. This paper explores a cluster-based ensemble method as a new solution for the purposes of activity recognition within smart environments. With this approach activities are modelled as collections of clusters built on different subsets of features. A classification process is performed by assigning a new instance to its closest cluster from each collection. Two different sensor data representations have been investigated, namely numeric and binary. Following the evaluation of the proposed methodology it has been demonstrated that the cluster-based ensemble method can be successfully applied as a viable option for activity recognition. Results following exposure to data collected from a range of activities indicated that the ensemble method had the ability to perform with accuracies of 94.2% and 97.5% for numeric and binary data, respectively. These results outperformed a range of single classifiers considered as benchmarks.
LanguageEnglish
Pages12285
JournalSensors
Volume14
Issue number7
DOIs
Publication statusPublished - 2014

Fingerprint

learning
Cluster Analysis
Learning
Sensors
Benchmarking
Aptitude
Classifiers
Monitoring
Technology
binary data
sensors
classifiers
Recognition (Psychology)
set theory
methodology
evaluation

Cite this

@article{88aa3f4850b74d82a3dc0ed65f66d641,
title = "Clustering-Based Ensemble Learning for Activity Recognition in Smart Homes",
abstract = "Application of sensor-based technology within activity monitoring systems is becoming a popular technique within the smart environment paradigm. Nevertheless, the use of such an approach generates complex constructs of data, which subsequently requires the use of intricate activity recognition techniques to automatically infer the underlying activity. This paper explores a cluster-based ensemble method as a new solution for the purposes of activity recognition within smart environments. With this approach activities are modelled as collections of clusters built on different subsets of features. A classification process is performed by assigning a new instance to its closest cluster from each collection. Two different sensor data representations have been investigated, namely numeric and binary. Following the evaluation of the proposed methodology it has been demonstrated that the cluster-based ensemble method can be successfully applied as a viable option for activity recognition. Results following exposure to data collected from a range of activities indicated that the ensemble method had the ability to perform with accuracies of 94.2{\%} and 97.5{\%} for numeric and binary data, respectively. These results outperformed a range of single classifiers considered as benchmarks.",
author = "Anna Jurek and CD Nugent and Yaxin Bi and Shengli Wu",
year = "2014",
doi = "10.3390/s140712285",
language = "English",
volume = "14",
pages = "12285",
journal = "Sensors",
issn = "1424-8220",
publisher = "MDPI",
number = "7",

}

Clustering-Based Ensemble Learning for Activity Recognition in Smart Homes. / Jurek, Anna; Nugent, CD; Bi, Yaxin; Wu, Shengli.

In: Sensors, Vol. 14, No. 7, 2014, p. 12285.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Clustering-Based Ensemble Learning for Activity Recognition in Smart Homes

AU - Jurek, Anna

AU - Nugent, CD

AU - Bi, Yaxin

AU - Wu, Shengli

PY - 2014

Y1 - 2014

N2 - Application of sensor-based technology within activity monitoring systems is becoming a popular technique within the smart environment paradigm. Nevertheless, the use of such an approach generates complex constructs of data, which subsequently requires the use of intricate activity recognition techniques to automatically infer the underlying activity. This paper explores a cluster-based ensemble method as a new solution for the purposes of activity recognition within smart environments. With this approach activities are modelled as collections of clusters built on different subsets of features. A classification process is performed by assigning a new instance to its closest cluster from each collection. Two different sensor data representations have been investigated, namely numeric and binary. Following the evaluation of the proposed methodology it has been demonstrated that the cluster-based ensemble method can be successfully applied as a viable option for activity recognition. Results following exposure to data collected from a range of activities indicated that the ensemble method had the ability to perform with accuracies of 94.2% and 97.5% for numeric and binary data, respectively. These results outperformed a range of single classifiers considered as benchmarks.

AB - Application of sensor-based technology within activity monitoring systems is becoming a popular technique within the smart environment paradigm. Nevertheless, the use of such an approach generates complex constructs of data, which subsequently requires the use of intricate activity recognition techniques to automatically infer the underlying activity. This paper explores a cluster-based ensemble method as a new solution for the purposes of activity recognition within smart environments. With this approach activities are modelled as collections of clusters built on different subsets of features. A classification process is performed by assigning a new instance to its closest cluster from each collection. Two different sensor data representations have been investigated, namely numeric and binary. Following the evaluation of the proposed methodology it has been demonstrated that the cluster-based ensemble method can be successfully applied as a viable option for activity recognition. Results following exposure to data collected from a range of activities indicated that the ensemble method had the ability to perform with accuracies of 94.2% and 97.5% for numeric and binary data, respectively. These results outperformed a range of single classifiers considered as benchmarks.

U2 - 10.3390/s140712285

DO - 10.3390/s140712285

M3 - Article

VL - 14

SP - 12285

JO - Sensors

T2 - Sensors

JF - Sensors

SN - 1424-8220

IS - 7

ER -