Factorial Discriminating

Mouly, P.P. et ak, Differentiation of citrus juices by factorial discriminant analysis using liquid chromatography of flavanone glycosides, J. Agric. Food Chem., 42, 70, 1994. 

Figure 1. Principal Components Analysis (PCA) and Factorial Discriminating Analysis (FDA) ofRavensara aromatica Essential oils (26).

Fluorescence spectroscopy coupled with factorial discriminant analysis technique to identify sheep milk from different feeding systems. Food Chemistry, 122, pp. 1344-1350, ISSN 0308-8146... 

There have been many other methods put forth in the literature including K-nearest neighbor (Ref. 42), cluster analysis (Ref. 11), principal component analysis (PCA) factorial discriminant analysis (Refs. 43, 44), SIMCA (Ref. 36), and BEAST (Refs. 37-40). 

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. 

The result of the discriminant analysis is shown in Fig. 5. The industrial activities are represented by ellipses of inertia containing the coordinates of industrial categories on the first factorial plane. All the categories largely overlap each other and show that this set of ecotoxicological data could not allow any linkage between the type of industry and the toxicological pattern. 

Fig. 9-9. Plot of the scores of discriminant function df 2 vs. scores of discriminant function df 1 of the different areas ( surroundings of the ferrous metallurgical factory, surroundings of the chemical factory, x unpolluted area for comparison). (The circles correspond to the 5% risk of error of the MANOVA)...

Addelman, S. (1962). Symmetrical and asymmetrical fractional factorial plans. Technometrics, 4, 47-58. Allen, T. T. and Bemshteyn, M. (2003). Supersaturated designs that maximize the probability of identifying active factors. Technometrics, 45, 92-97. Bingham, D. and Li, W. (2002). A class of optimal robust parameter designs. Journal of Quality Technology, 34, 244—259. Biswas, A. and Chaudhuri, P. (2002). An efficient design for model discrimination and ... 

This type of test depends upon the testers being representative of the group in question, e.g. fudge purchasers. If the testers are recruited from the staff of a confectionery factory there is a risk that they will become more discriminating than the general public. 

Much has been published concerning toxicity of metals to aquatic life although only during the past decade have there been interpretations of the toxicity data in terms of ]jie rela ive toxicity of particular speciation forms, e.g., Cu, CuOH, CuCO The specific objective of this paper is to illustrate the use of Factor Analysis in discriminating between toxic and nontoxic species. The procedure to be followed is determination of equilibrium aqueous speciation, calculation of appropriate factory correlation of toxicity with these factors, and interpretation of the correlation analysis in terms of particular species activities. 

Sequential strategies for discrimination can be risky if applied before a thorough enumeration of plausible model forms. Important regions of experimentation may then be missed, and the kinds of models needed to describe those regions may go undiscovered. Factorial experimentation at the outset is desirable to allow inductive building of good model forms. 

It can be seen that for factory A, two out of eleven batches differ from the others, whereas for factory B three groups of paints can be discriminated three batches are characterized by the ratios of peak heights of the characteristic substance of 0.5 0.1, two batches 0.7 0.1 and six batches 0.8 0.1. Similar results could not be obtained by inorganic analysis [16]. Emission spectroscopic analysis, for example, identified 53 paint samples out of 190, as opposed to 141 identified by Py—GC [16]. 

Flavors are widely used in pharmaceutical solutions to mask drug bitterness. Zhu s group [48] has used an MOS electronic nose to perform headspace analysis of these formulations. The method was able to qualitatively distinguish six common flavors (raspberry, red berry, strawberry, pineapple, orange, and cherry) in placebo mixtures. The instrument was also able to identify unknown flavors. It was also indicated that the instrument could be used to identify different flavor raw materials. Moreover, the electronic nose was used for quantitative analysis of flavors in an oral solution. Data processing and identification were done by PCA, discriminant factorial analysis (DFA), and partial least squares. 

Taylor K, Wick C, Castada H, et al. Discrimination of Swiss cheese fiom 5 different factories by high impact volatile organic compound profiles determined by odor activity values using selected ion flow tube-mass spectrometry and odor threshold. J Food Sci. 2013 78 C1509 15. 

---