2012 | OriginalPaper | Buchkapitel
Multivariate Adaptive Embedding, MAE-Process
verfasst von : Gerhard Sartorius
Erschienen in: Autonomous Systems: Developments and Trends
Verlag: Springer Berlin Heidelberg
Aktivieren Sie unsere intelligente Suche, um passende Fachinhalte oder Patente zu finden.
Wählen Sie Textabschnitte aus um mit Künstlicher Intelligenz passenden Patente zu finden. powered by
Markieren Sie Textabschnitte, um KI-gestützt weitere passende Inhalte zu finden. powered by
The multivariate adaptive embedding (MAE-Process) provides an adaptive System which creates artificial neural network in the form of an appropriate model of the training-data set by using a globally optimal optimisation and acts in most cases without iterations and parameter settings. There is basically no change to input data and the training-data set is prepared by a special fitting method in order to make it treatable for spectral methods. In the working phase, new input data can be processed multivariatly by the system with good generalising properties. In combination with a Wavelet transformation (WT) for noise- and data reduction, the System performs fast and efficiently in classifying parameterised curves, such as Raman spectra.