Discriminant analysis definition

One major issue that causes some problems in the application of discriminant techniques, such as parameter focusing and discriminant analysis, is the definition of what is aaive and what is inactive. An approach to solving this problem is to define activity as very active, active, sli tly active, and inactive, and then to remove the compounds fitting the middle two classifications. The ultimate product of such pattern recognition techniques should be... 

Downey et al. (14) used a statistical approach to classify commercial skim milk powders according to heat treatment. They used 66 samples of commercially produced skim milk powder including high-heat, medium-heat, and low-heat powders. Principal component analysis (PCA) was applied to the normalized spectral data, with the use of wavelengths as principal variables and class values as supplementary variables. Factorial discriminant analysis (FDA) was performed on the PC scores. Ten components were needed to correctly classify all samples in the calibration development set 91% of those in the evaluation set were correctly identified. Three samples of the medium-heat class were incorrectly classified, but the authors pointed out difficulties in the exact definition of the heat treatment classes, particularly the medium-heat class. 

Such definitive classification may be achieved with the aid of multivariate pattern recognition techniques such as hierarchical clustering, linear discriminant analysis (LDA) and artificial neural network analysis. Hierarchical clustering techniques compare sets of data (e.g. individually acquired spectra or spectra acquired by mapping of tissue) and group the data according to some measure of similarity. For mapping data, the application of cluster analysis... 

When casework samples are being analysed, the problem of co-elution becomes even more evident (Figure 4.9). It is difficult to resolve all of the materials and diode-array spectra are not sufficiently discriminating to allow definitive identification. Coupled to long analysis times and short column life, this makes interpretation of seemingly near-identical chromatograms extremely difficult. 

Determination of Assay Sensitivity with Replication and Statistics. Sensitivity can be defined as the ability of a test to discriminate between adjacent levels or concentrations of test analyte. There are other definitions of sensitivity, but the one specified is sufficiently general to serve several needs in residue analysis. For example, the definition recognizes that test sensitivity can vary with the point on the standard curve. If one of the points used is zero, then the sensitivity estimate can be either the level of smallest quantitation or the level of detectability of the method. The... 

---