Abstract
In this chapter we explore the problem of black box time series identification for the case of source switching. This amounts to unsupervised development of models for a time series which is generated by a collection of alternately activated, initially unknown sources. We present two algorithms which accomplish this task and present guidelines which can be used to develop variations of these algorithms. A concept which is central to our presentation is data allocation. Numerical experiments are presented to illustrate our point of view.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
Author information
Authors and Affiliations
Rights and permissions
Copyright information
© 1998 Springer Science+Business Media New York
About this chapter
Cite this chapter
Petridis, V., Kehagias, A. (1998). Source Identification Algorithms. In: Predictive Modular Neural Networks. The Springer International Series in Engineering and Computer Science, vol 466. Springer, Boston, MA. https://doi.org/10.1007/978-1-4615-5555-1_10
Download citation
DOI: https://doi.org/10.1007/978-1-4615-5555-1_10
Publisher Name: Springer, Boston, MA
Print ISBN: 978-1-4613-7540-1
Online ISBN: 978-1-4615-5555-1
eBook Packages: Springer Book Archive