Home // ADVCOMP 2012, The Sixth International Conference on Advanced Engineering Computing and Applications in Sciences // View article

The use of Bioinformatics Techniques for Time-Series Motif-Matching: A Case Study

Authors:
Mark Transell
Carl Sandrock

Keywords: Dynamic Time Warping; BLAST; motif-matching; time series; PAA

Abstract:
Process engineers have more access to historical plant data than ever before. Finding recurring patterns in process data, also referred to as motif-matching, may reveal diagnostic information to engineers and operators. Dynamic Time Warping (DTW) is one of the most widely used techniques for performing these motif matches. Sequence matching is also an important part of bioinformatics; a field which has received a marked increase in research funding and attention in recent times. Therefore, the techniques developed in bioinformatics may be beneficial to the field of time-series motif matching. In this study, a combination of the Symbolic Aggregate Approximation (SAX) algorithm and the PSI-BLAST bioinformatics algorithm is compared to DTW as a potential method to perform time-series matches. Preliminary results suggest that this combination may be faster than global DTW techniques for large datasets. Details of the implementation are given, along with preliminary results confirming that this method is feasible. Due to implementation difficulties, accuracy and robustness remain uninvestigated. More research is recommended into the potential for this technique as an alternative to Dynamic Time Warping techniques.

Pages: 114 to 117

Copyright: Copyright (c) IARIA, 2012

Publication date: September 23, 2012

Published in: conference

ISSN: 2308-4499

ISBN: 978-1-61208-237-0

Location: Barcelona, Spain

Dates: from September 23, 2012 to September 28, 2012