The adaptable buffer algorithm for high quantile estimation in non-stationary data streams

Arandjelović, Ognjen, Pham, Duc-Son and Venkatesh, Svetha 2015, The adaptable buffer algorithm for high quantile estimation in non-stationary data streams, in IJCNN 2015: Proceedings of the International Joint Conference on Neural Networks, IEEE, Piscataway, N.J., pp. 1-7, doi: 10.1109/IJCNN.2015.7280314.

Attached Files
Name Description MIMEType Size Downloads

Title The adaptable buffer algorithm for high quantile estimation in non-stationary data streams
Author(s) Arandjelović, Ognjen
Pham, Duc-Son
Venkatesh, SvethaORCID iD for Venkatesh, Svetha orcid.org/0000-0001-8675-6631
Conference name International Joint Conference on Neural Networks (2015 : Killarney, Ireland)
Conference location Killarney, Ireland
Conference dates 12-17 Jul. 2015
Title of proceedings IJCNN 2015: Proceedings of the International Joint Conference on Neural Networks
Publication date 2015
Start page 1
End page 7
Total pages 7
Publisher IEEE
Place of publication Piscataway, N.J.
Summary The need to estimate a particular quantile of a distribution is an important problem which frequently arises in many computer vision and signal processing applications. For example, our work was motivated by the requirements of many semi-automatic surveillance analytics systems which detect abnormalities in close-circuit television (CCTV) footage using statistical models of low-level motion features. In this paper we specifically address the problem of estimating the running quantile of a data stream with non-stationary stochasticity when the memory for storing observations is limited. We make several major contributions: (i) we derive an important theoretical result which shows that the change in the quantile of a stream is constrained regardless of the stochastic properties of data, (ii) we describe a set of high-level design goals for an effective estimation algorithm that emerge as a consequence of our theoretical findings, (iii) we introduce a novel algorithm which implements the aforementioned design goals by retaining a sample of data values in a manner adaptive to changes in the distribution of data and progressively narrowing down its focus in the periods of quasi-stationary stochasticity, and (iv) we present a comprehensive evaluation of the proposed algorithm and compare it with the existing methods in the literature on both synthetic data sets and three large 'real-world' streams acquired in the course of operation of an existing commercial surveillance system. Our findings convincingly demonstrate that the proposed method is highly successful and vastly outperforms the existing alternatives, especially when the target quantile is high valued and the available buffer capacity severely limited.
ISBN 9781479919604
Language eng
DOI 10.1109/IJCNN.2015.7280314
Field of Research 080109 Pattern Recognition and Data Mining
Socio Economic Objective 970108 Expanding Knowledge in the Information and Computing Sciences
HERDC Research category E1 Full written paper - refereed
ERA Research output type E Conference publication
Copyright notice ©2015, IEEE
Persistent URL http://hdl.handle.net/10536/DRO/DU:30081885

Connect to link resolver
 
Unless expressly stated otherwise, the copyright for items in DRO is owned by the author, with all rights reserved.

Versions
Version Filter Type
Citation counts: TR Web of Science Citation Count  Cited 2 times in TR Web of Science
Scopus Citation Count Cited 2 times in Scopus
Google Scholar Search Google Scholar
Access Statistics: 457 Abstract Views, 4 File Downloads  -  Detailed Statistics
Created: Fri, 04 Mar 2016, 14:13:40 EST

Every reasonable effort has been made to ensure that permission has been obtained for items included in DRO. If you believe that your rights have been infringed by this repository, please contact drosupport@deakin.edu.au.