Skip to main content

Research Repository

See what's under the surface

Clustering and nearest neighbour based classification approach for mobile activity recognition.

Bashir, S.A; Doolan, D.C.; Petrovski, A.

Authors

S.A Bashir

D.C. Doolan

A. Petrovski



Abstract

We present a hybridized algorithm based on clustering and nearest neighbour classifier for mobile activity recognition. The algorithm transforms a training dataset into a more compact and reduced representative set that lessens the computational cost on mobile devices. This is achieved by applying clustering on the original dataset with the concept of percentage data retention to direct the operation. After clustering, we extract three reduced and transformed representation of the original dataset to serve as the reference data for nearest neighbour classification. These reduced representative sets can be used for classifying new instances using the nearest neighbour algorithm step on the mobile phone. Experimental evaluation of our proposed approach using real mobile activity recognition dataset shows improved result over the basic KNN algorithm that uses all the training dataset.

Journal Article Type Article
Publication Date Apr 8, 2016
Journal Journal of mobile multimedia
Print ISSN 1550-4646
Publisher River Publishers
Peer Reviewed Peer Reviewed
Volume 12
Issue 1-2
Pages 100-124
Institution Citation BASHIR, S.A., DOOLAN, D.C. and PETROVSKI, A. 2016. Clustering and nearest neighbour based classification approach for mobile activity recognition. Journal of mobile multimedia [online], 12(1-2), pages 100-124. Available from: http://www.rintonpress....2/jmm-12-12/110-124.pdf
Keywords Activity recognition; KNN; Smartphones; Clustering
Publisher URL http://www.rintonpress.com/xjmm12/jmm-12-12/110-124.pdf

Files





You might also like



Downloadable Citations