What question did this study set out to answer?

To introduce nonlinear principal component analysis (NLPCA) using autoassociative neural networks to uncover both linear and nonlinear relationships among multivariate variables for dimensionality reduction.

February 1, 1991AIChE Journal

Nonlinear principal component analysis using autoassociative neural networks

View Full Paper

Ask AI

Bookmark

Authors

MKMark KramerBoston University

Discussion

Member takes

Implication

Computational modeling demonstrates nonlinear dimensionality reduction via autoassociative neural networks in simulated reaction data, highlighting improved extraction of system parameters.

Key Points

To introduce nonlinear principal component analysis (NLPCA) using autoassociative neural networks to uncover both linear and nonlinear relationships among multivariate variables for dimensionality reduction.
Designed a feedforward neural network architecture featuring an internal bottleneck layer and two hidden layers configured to perform identity mapping between input and output layers.
Trained and evaluated the network using time-dependent, simulated batch chemical reaction data to test feature extraction and dimensionality reduction capabilities.
NLPCA successfully extracted both linear and arbitrary nonlinear correlations among variables without constraints on the character of the data nonlinearities.
The network produced a reduced-dimensional feature space map that accurately reflected the underlying parameter distribution of the simulated reaction system.

Cite This Study

Mark Kramer (1991) studied this question.

synapsesocial.com/papers/69dc23aeea70a37eff95535a https://doi.org/10.1002/aic.690370209

View Full Paper

Ask AI

Bookmark

Implication

Computational modeling demonstrates nonlinear dimensionality reduction via autoassociative neural networks in simulated reaction data, highlighting improved extraction of system parameters.

Nonlinear principal component analysis using autoassociative neural networks

Discussion

Member takes

Implication

Key Points

Cite This Study

Nonlinear principal component analysis using autoassociative neural networks

Discussion

Member takes

Implication

Key Points

Cite This Study

Also Consider

Also Consider