Document Type

Article

Language

eng

Format of Original

6 p.

Publication Date

2011

Publisher

Institute of Electrical and Electronics Engineers (IEEE)

Source Publication

Proceeding of the IEEE International Conference on Information and Automation Shenzhen, China June 2011

Source ISSN

1063-6374

Original Item ID

doi: 10.1109/ICINFA.2011.5948989

Abstract

In this paper we present a method for detecting dynamic temporal patterns that are characteristic and predictive of significant events in a dynamic data system. We employ the Gaussian Mixture Model (GMM) to cluster the data sequence into three categories of signals, e.g. normal, patterns and events. The data sequence is then embedded into a Reconstructed Phase Space (RPS) which is topologically equivalent to the dynamics of the original system. We apply a hybrid method using Support Vector Machines (SVM) and Maximum a Posterior (MAP) to classify temporal pattern signals based on the event function. We performed two experimental applications using chaotic time series and Sludge Volume Index (SVI) series related to the Sludge Bulking problem. The proposed hybrid GMM-SVM phase space approach effectively detects temporal patterns and achieves higher predictive accuracy compared with the original RPS framework.

Comments

Accepted version. Proceedings of the IEEE International Conference on Information and Automation Shenzhen, China, (June 2010): 209-214. DOI: 10.1109/ICINFA.2011.5948989 © 2010 Institute of Electrical and Electronics Engineers. Used with permission.

ahamed_2526acc.docx (222 kB)
ADA Accessible Version

Share

COinS