Training
During the training phase of the system all the parameters that are necessary to distinguish the spectra in the given classes are
automatically determined and optimized. The result is a "Recognition Module. Due to this, new and unknown spectra can be classified,
that is means in particular the spectra can be assigned to their appropriate class.
The training is divided into two steps: Learning and Classification.
1. Learning
In the process of learnig all spectra are taught at the same time. During this phase, the program determines the characteristics of the
individual spectra of the respective classes.
A number of small spectral regions is determined, which might be
useful for solving the given classification problem. These spectral intervals are called "Region Of Interest (ROI)".
For each ROI, one number is calculated per spectrum (serving as feature value).
In addition, depending on the previous selected
method, special measures for the distances between spectra and class mean curves might be added. According to this calculation
airspect SRS will cull automatically the most important features.
2. Classification
In this step the actual classifier is built up, optimized, and trained with the given known spectra. After that, it can be used to
determine the classes for unknown spectra. The user may choose the simple and quick "Nearest Centroid Classifier", or an "Artificial
Neural Network" with or without hidden units, which usually achieves better prediction performance.
The user can choose between 3 different classification methods:
- General classification
- Clusters
- Regression
The training steps described above yield as a by-product valuable information about your spectra's behavior during the various processes.
Inspect the assignment tables produced during the training steps in order to identify those spectra, which have been misassigned by
the system exceptionally many times. Recheck the concerned spectra. In many cases, they turn out to be of low quality, possibly noisy,
or even erroneously assigned to an incorrect class by the user.
Spectroscopy Techniques 
