Biomedical Engineering Reference
In-Depth Information
Fig. 5 Autoassociative neural network structure, indicating five layers performing unity mapping;
only a selection of network weights are illustrated to reduce the complexity of the figure
Fig. 6
Two
different
self-organising
map
arrangements
with
best
matching
unit
(B)
and
neighbouring nodes, whose weights are adjusted upon matching the input pattern to B
production through dark fermentation. However, SOMs were also used to analyse
2D fluorescence data collected during the cultivation of recombinant E. col and
Saccharomyces cerevisiae, where they successfully captured the relationships
between the spectra and process parameters [
45
].
3.4 Neural Network-Based Feature Extraction Case Study
Sections 3.2
and
3.3
concentrated on linear and non-linear MVDA methods,
respectively, although it is important to stress that these two categories of methods
are
often
used
in
combination
to
increase
the
accuracy
of
the
resulting
Search WWH ::
Custom Search