A New Mixture Model for Concept Learning from Time Series

William H. Hsu and Sylvian R. Ray

We present an algorithm for combining data from multiple input sources (sensors, specialists with different concentrations, etc.) and a modular, recurrent, artificial neural network for time series learning. Fusion of time series classifiers showcase the strengths of our mixture model because there are many preprocessing methods that produce reformulated input. Typical applications are process monitoring, prediction, and control.


This page is copyrighted by AAAI. All rights reserved. Your use of this site constitutes acceptance of all of AAAI's terms and conditions and privacy policy.