CBARS: Cluster based classification for activity recognition systems

Zahraa Said Abdallah; Mohamed Medhat Gaber; Bala Srinivasan; Shonali Krishnaswamy

doi:10.1007/978-3-642-35326-0

CBARS: Cluster based classification for activity recognition systems

Zahraa Said Abdallah, Mohamed Medhat Gaber, Bala Srinivasan, Shonali Krishnaswamy

Research output: Chapter in Book/Report/Conference proceeding › Conference Paper › Research › peer-review

16 Citations (Scopus)

Abstract

Activity recognition focuses on inferring current user activities by leveraging sensory data available on today’s sensor rich environment. Supervised learning has been applied pervasively for activity recognition. Typical activity recognition techniques process sensory data based on point-by-point approaches. In this paper, we propose a novel Cluster Based Classification for Activity Recognition Systems, CBARS. The novel approach processes activities as clusters to build a robust classification framework. CBARS integrates supervised, unsupervised and active learning and applies hybrid similarity measures technique for recognising activities. Extensive experimental results using real activity recognition dataset have evidenced that our new approach shows improved performance over other existing state-of-the-art learning methods.

Original language	English
Title of host publication	Advanced Machine Learning Technologies and Applications
Subtitle of host publication	First International Conference, AMLTA 2012, Cairo, Egypt, December 8-10, 2012, Proceedings
Editors	Aboul Ella Hassanien, Abdel-Badeeh M. Salem, Rabie Ramadan, Tai-hoon Kim
Place of Publication	Heidelberg [Germany]
Publisher	Springer
Pages	82 - 91
Number of pages	10
Volume	322
ISBN (Electronic)	9783642353260
ISBN (Print)	9783642353253
DOIs	https://doi.org/10.1007/978-3-642-35326-0
Publication status	Published - 2012
Event	International Conference on Advanced Machine Learning Technologies and Applications (AMLTA) 2012 - Cairo, Egypt Duration: 8 Dec 2012 → 10 Dec 2012 Conference number: 1st https://link.springer.com/book/10.1007/978-3-642-35326-0 (Conference Proceedings)

Publication series

Name	Communications in Computer and Information Science
Publisher	Springer
Volume	322
ISSN (Print)	1865-0929
ISSN (Electronic)	1865-0937

Conference

Conference	International Conference on Advanced Machine Learning Technologies and Applications (AMLTA) 2012
Abbreviated title	AMLTA 2012
Country/Territory	Egypt
City	Cairo
Period	8/12/12 → 10/12/12
Internet address	https://link.springer.com/book/10.1007/978-3-642-35326-0 (Conference Proceedings)

Keywords

Activity recognition
Cluster based classification
Hybrid similarity measure

Access to Document

10.1007/978-3-642-35326-0

Cite this

Abdallah, Z. S., Gaber, M. M., Srinivasan, B., & Krishnaswamy, S. (2012). CBARS: Cluster based classification for activity recognition systems. In A. E. Hassanien, A.-B. M. Salem, R. Ramadan, & T. Kim (Eds.), Advanced Machine Learning Technologies and Applications: First International Conference, AMLTA 2012, Cairo, Egypt, December 8-10, 2012, Proceedings (Vol. 322, pp. 82 - 91). (Communications in Computer and Information Science; Vol. 322). Springer. https://doi.org/10.1007/978-3-642-35326-0

Abdallah, Zahraa Said ; Gaber, Mohamed Medhat ; Srinivasan, Bala et al. / CBARS : Cluster based classification for activity recognition systems. Advanced Machine Learning Technologies and Applications: First International Conference, AMLTA 2012, Cairo, Egypt, December 8-10, 2012, Proceedings. editor / Aboul Ella Hassanien ; Abdel-Badeeh M. Salem ; Rabie Ramadan ; Tai-hoon Kim. Vol. 322 Heidelberg [Germany] : Springer, 2012. pp. 82 - 91 (Communications in Computer and Information Science).

@inproceedings{9d81c3ffbe7f427fae63739cb5ed75a8,

title = "CBARS: Cluster based classification for activity recognition systems",

abstract = "Activity recognition focuses on inferring current user activities by leveraging sensory data available on today{\textquoteright}s sensor rich environment. Supervised learning has been applied pervasively for activity recognition. Typical activity recognition techniques process sensory data based on point-by-point approaches. In this paper, we propose a novel Cluster Based Classification for Activity Recognition Systems, CBARS. The novel approach processes activities as clusters to build a robust classification framework. CBARS integrates supervised, unsupervised and active learning and applies hybrid similarity measures technique for recognising activities. Extensive experimental results using real activity recognition dataset have evidenced that our new approach shows improved performance over other existing state-of-the-art learning methods.",

keywords = "Activity recognition, Cluster based classification, Hybrid similarity measure",

author = "Abdallah, {Zahraa Said} and Gaber, {Mohamed Medhat} and Bala Srinivasan and Shonali Krishnaswamy",

year = "2012",

doi = "10.1007/978-3-642-35326-0",

language = "English",

isbn = "9783642353253",

volume = "322",

series = "Communications in Computer and Information Science",

publisher = "Springer",

pages = "82 -- 91",

editor = "Hassanien, {Aboul Ella} and Salem, {Abdel-Badeeh M.} and Rabie Ramadan and Tai-hoon Kim",

booktitle = "Advanced Machine Learning Technologies and Applications",

note = "International Conference on Advanced Machine Learning Technologies and Applications (AMLTA) 2012, AMLTA 2012 ; Conference date: 08-12-2012 Through 10-12-2012",

url = "https://link.springer.com/book/10.1007/978-3-642-35326-0",

}

Abdallah, ZS, Gaber, MM, Srinivasan, B & Krishnaswamy, S 2012, CBARS: Cluster based classification for activity recognition systems. in AE Hassanien, A-BM Salem, R Ramadan & T Kim (eds), Advanced Machine Learning Technologies and Applications: First International Conference, AMLTA 2012, Cairo, Egypt, December 8-10, 2012, Proceedings. vol. 322, Communications in Computer and Information Science, vol. 322, Springer, Heidelberg [Germany], pp. 82 - 91, International Conference on Advanced Machine Learning Technologies and Applications (AMLTA) 2012, Cairo, Egypt, 8/12/12. https://doi.org/10.1007/978-3-642-35326-0

CBARS: Cluster based classification for activity recognition systems. / Abdallah, Zahraa Said; Gaber, Mohamed Medhat; Srinivasan, Bala et al.
Advanced Machine Learning Technologies and Applications: First International Conference, AMLTA 2012, Cairo, Egypt, December 8-10, 2012, Proceedings. ed. / Aboul Ella Hassanien; Abdel-Badeeh M. Salem; Rabie Ramadan; Tai-hoon Kim. Vol. 322 Heidelberg [Germany]: Springer, 2012. p. 82 - 91 (Communications in Computer and Information Science; Vol. 322).

Research output: Chapter in Book/Report/Conference proceeding › Conference Paper › Research › peer-review

TY - GEN

T1 - CBARS

T2 - International Conference on Advanced Machine Learning Technologies and Applications (AMLTA) 2012

AU - Abdallah, Zahraa Said

AU - Gaber, Mohamed Medhat

AU - Srinivasan, Bala

AU - Krishnaswamy, Shonali

N1 - Conference code: 1st

PY - 2012

Y1 - 2012

N2 - Activity recognition focuses on inferring current user activities by leveraging sensory data available on today’s sensor rich environment. Supervised learning has been applied pervasively for activity recognition. Typical activity recognition techniques process sensory data based on point-by-point approaches. In this paper, we propose a novel Cluster Based Classification for Activity Recognition Systems, CBARS. The novel approach processes activities as clusters to build a robust classification framework. CBARS integrates supervised, unsupervised and active learning and applies hybrid similarity measures technique for recognising activities. Extensive experimental results using real activity recognition dataset have evidenced that our new approach shows improved performance over other existing state-of-the-art learning methods.

AB - Activity recognition focuses on inferring current user activities by leveraging sensory data available on today’s sensor rich environment. Supervised learning has been applied pervasively for activity recognition. Typical activity recognition techniques process sensory data based on point-by-point approaches. In this paper, we propose a novel Cluster Based Classification for Activity Recognition Systems, CBARS. The novel approach processes activities as clusters to build a robust classification framework. CBARS integrates supervised, unsupervised and active learning and applies hybrid similarity measures technique for recognising activities. Extensive experimental results using real activity recognition dataset have evidenced that our new approach shows improved performance over other existing state-of-the-art learning methods.

KW - Activity recognition

KW - Cluster based classification

KW - Hybrid similarity measure

UR - http://www.scopus.com/inward/record.url?scp=84880355940&partnerID=8YFLogxK

UR - https://link.springer.com/book/10.1007%2F978-3-642-35326-0

U2 - 10.1007/978-3-642-35326-0

DO - 10.1007/978-3-642-35326-0

M3 - Conference Paper

SN - 9783642353253

VL - 322

T3 - Communications in Computer and Information Science

SP - 82

EP - 91

BT - Advanced Machine Learning Technologies and Applications

A2 - Hassanien, Aboul Ella

A2 - Salem, Abdel-Badeeh M.

A2 - Ramadan, Rabie

A2 - Kim, Tai-hoon

PB - Springer

CY - Heidelberg [Germany]

Y2 - 8 December 2012 through 10 December 2012

ER -

Abdallah ZS, Gaber MM, Srinivasan B, Krishnaswamy S. CBARS: Cluster based classification for activity recognition systems. In Hassanien AE, Salem ABM, Ramadan R, Kim T, editors, Advanced Machine Learning Technologies and Applications: First International Conference, AMLTA 2012, Cairo, Egypt, December 8-10, 2012, Proceedings. Vol. 322. Heidelberg [Germany]: Springer. 2012. p. 82 - 91. (Communications in Computer and Information Science). doi: 10.1007/978-3-642-35326-0

CBARS: Cluster based classification for activity recognition systems

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Cite this