Change Point Detection Using Multivariate Exponentially Weighted Moving Average (MEWMA) for Optimal Parameter in Online Activity Monitoring

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

In recent years, wearable sensors are integrating frequently and rapidly into our daily life day by day. Such smart sensors have attracted a lot of interest due to their small sizes and reasonable computational power. For example, body worn sensors are widely used to monitor daily life activities and identify meaningful events. Hence, the capability to detect, adapt and respond to change performs a key role in various domains. A change in activities is signaled by a change in the data distribution within a time window. This change marks the start of a transition from an ongoing activity to a new one. In this paper, we evaluate the proposed algorithm’s scalability on identifying multiple changes in different user activities from real sensor data collected from various subjects. The Genetic algorithm (GA) is used to identify the optimal parameter set for Multivariate Exponentially Weighted Moving Average (MEWMA) approach to detect change points in sensor data. Results have been evaluated using a real dataset of 8 different activities for five different users with a high accuracy from 99.2 % to 99.95 % and G-means from 67.26 % to 83.20 %.
LanguageEnglish
Title of host publicationUCAmI 2016: Ubiquitous Computing and Ambient Intelligence
Place of PublicationSpain
Pages156-165
Number of pages9
Volume10069
ISBN (Electronic)978-3-319-48746-5
DOIs
Publication statusPublished - 2 Nov 2016
Event10th International Conference, UCAmI 2016, San Bartolomé de Tirajana, Gran - Gran Canaria, Spain
Duration: 29 Nov 20162 Dec 2016
https://link.springer.com/book/10.1007/978-3-319-48746-5

Conference

Conference10th International Conference, UCAmI 2016, San Bartolomé de Tirajana, Gran
CitySpain
Period29/11/162/12/16
Internet address

Fingerprint

Monitoring
Sensors
Smart sensors
Scalability
Genetic algorithms
Wearable sensors

Keywords

  • Multiple change points
  • Activity monitoring
  • Genetic algorithm
  • Accelerometer

Cite this

@inproceedings{8eebd79dea544154bffcbee521fc529c,
title = "Change Point Detection Using Multivariate Exponentially Weighted Moving Average (MEWMA) for Optimal Parameter in Online Activity Monitoring",
abstract = "In recent years, wearable sensors are integrating frequently and rapidly into our daily life day by day. Such smart sensors have attracted a lot of interest due to their small sizes and reasonable computational power. For example, body worn sensors are widely used to monitor daily life activities and identify meaningful events. Hence, the capability to detect, adapt and respond to change performs a key role in various domains. A change in activities is signaled by a change in the data distribution within a time window. This change marks the start of a transition from an ongoing activity to a new one. In this paper, we evaluate the proposed algorithm’s scalability on identifying multiple changes in different user activities from real sensor data collected from various subjects. The Genetic algorithm (GA) is used to identify the optimal parameter set for Multivariate Exponentially Weighted Moving Average (MEWMA) approach to detect change points in sensor data. Results have been evaluated using a real dataset of 8 different activities for five different users with a high accuracy from 99.2 {\%} to 99.95 {\%} and G-means from 67.26 {\%} to 83.20 {\%}.",
keywords = "Multiple change points, Activity monitoring, Genetic algorithm, Accelerometer",
author = "Naveed Khan and McClean, {Sally I} and Shuai Zhang and CD Nugent",
year = "2016",
month = "11",
day = "2",
doi = "https://doi.org/10.1007/978-3-319-48746-5_16",
language = "English",
isbn = "978-3-319-48745-8",
volume = "10069",
pages = "156--165",
booktitle = "UCAmI 2016: Ubiquitous Computing and Ambient Intelligence",

}

Khan, N, McClean, SI, Zhang, S & Nugent, CD 2016, Change Point Detection Using Multivariate Exponentially Weighted Moving Average (MEWMA) for Optimal Parameter in Online Activity Monitoring. in UCAmI 2016: Ubiquitous Computing and Ambient Intelligence. vol. 10069, Spain, pp. 156-165, 10th International Conference, UCAmI 2016, San Bartolomé de Tirajana, Gran, Spain, 29/11/16. https://doi.org/10.1007/978-3-319-48746-5_16

Change Point Detection Using Multivariate Exponentially Weighted Moving Average (MEWMA) for Optimal Parameter in Online Activity Monitoring. / Khan, Naveed; McClean, Sally I; Zhang, Shuai; Nugent, CD.

UCAmI 2016: Ubiquitous Computing and Ambient Intelligence. Vol. 10069 Spain, 2016. p. 156-165.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - Change Point Detection Using Multivariate Exponentially Weighted Moving Average (MEWMA) for Optimal Parameter in Online Activity Monitoring

AU - Khan, Naveed

AU - McClean, Sally I

AU - Zhang, Shuai

AU - Nugent, CD

PY - 2016/11/2

Y1 - 2016/11/2

N2 - In recent years, wearable sensors are integrating frequently and rapidly into our daily life day by day. Such smart sensors have attracted a lot of interest due to their small sizes and reasonable computational power. For example, body worn sensors are widely used to monitor daily life activities and identify meaningful events. Hence, the capability to detect, adapt and respond to change performs a key role in various domains. A change in activities is signaled by a change in the data distribution within a time window. This change marks the start of a transition from an ongoing activity to a new one. In this paper, we evaluate the proposed algorithm’s scalability on identifying multiple changes in different user activities from real sensor data collected from various subjects. The Genetic algorithm (GA) is used to identify the optimal parameter set for Multivariate Exponentially Weighted Moving Average (MEWMA) approach to detect change points in sensor data. Results have been evaluated using a real dataset of 8 different activities for five different users with a high accuracy from 99.2 % to 99.95 % and G-means from 67.26 % to 83.20 %.

AB - In recent years, wearable sensors are integrating frequently and rapidly into our daily life day by day. Such smart sensors have attracted a lot of interest due to their small sizes and reasonable computational power. For example, body worn sensors are widely used to monitor daily life activities and identify meaningful events. Hence, the capability to detect, adapt and respond to change performs a key role in various domains. A change in activities is signaled by a change in the data distribution within a time window. This change marks the start of a transition from an ongoing activity to a new one. In this paper, we evaluate the proposed algorithm’s scalability on identifying multiple changes in different user activities from real sensor data collected from various subjects. The Genetic algorithm (GA) is used to identify the optimal parameter set for Multivariate Exponentially Weighted Moving Average (MEWMA) approach to detect change points in sensor data. Results have been evaluated using a real dataset of 8 different activities for five different users with a high accuracy from 99.2 % to 99.95 % and G-means from 67.26 % to 83.20 %.

KW - Multiple change points

KW - Activity monitoring

KW - Genetic algorithm

KW - Accelerometer

U2 - https://doi.org/10.1007/978-3-319-48746-5_16

DO - https://doi.org/10.1007/978-3-319-48746-5_16

M3 - Conference contribution

SN - 978-3-319-48745-8

VL - 10069

SP - 156

EP - 165

BT - UCAmI 2016: Ubiquitous Computing and Ambient Intelligence

CY - Spain

ER -