Learning from Time Series: Supervised Aggregative Feature Extraction


Schirru, Andrea and Susto, Gian Antonio and Pampuri, Simone and McLoone, Sean (2012) Learning from Time Series: Supervised Aggregative Feature Extraction. In: 51st Annual Conference on Decision and Control (CDC). IEEE, pp. 5254-5259. ISBN 9781467320658

[img] Download (311kB)



Add this article to your Mendeley library


Abstract

Many modeling problems require to estimate a scalar output from one or more time series. Such problems are usually tackled by extracting a fixed number of features from the time series (like their statistical moments), with a consequent loss in information that leads to suboptimal predictive models. Moreover, feature extraction techniques usually make assumptions that are not met by real world settings (e.g. uniformly sampled time series of constant length), and fail to deliver a thorough methodology to deal with noisy data. In this paper a methodology based on functional learning is proposed to overcome the aforementioned problems; the proposed Supervised Aggregative Feature Extraction (SAFE) approach allows to derive continuous, smooth estimates of time series data (yielding aggregate local information), while simultaneously estimating a continuous shape function yielding optimal predictions. The SAFE paradigm enjoys several properties like closed form solution, incorporation of first and second order derivative information into the regressor matrix, interpretability of the generated functional predictor and the possibility to exploit Reproducing Kernel Hilbert Spaces setting to yield nonlinear predictive models. Simulation studies are provided to highlight the strengths of the new methodology w.r.t. standard unsupervised feature selection approaches.

Item Type: Book Section
Additional Information: The definitive version of this article is available at 10.1109/CDC.2012.6427099
Keywords: Kernel Hilbert Spaces; SAFE; feature extraction techniques; fixed number; functional learning; nonlinear predictive models; scalar output; statistical moments; suboptimal predictive models; supervised aggregative feature extraction; time series data; time series learning;
Academic Unit: Faculty of Science and Engineering > Electronic Engineering
Item ID: 4228
Depositing User: Sean McLoone
Date Deposited: 27 Feb 2013 16:14
Journal or Publication Title: 51st IEEE Conference on Decision and Control, Proceedings
Publisher: IEEE
Refereed: Yes
URI:

Repository Staff Only(login required)

View Item Item control page

Document Downloads

More statistics for this item...