Neural network Multivariate analysis

Other methods consist of algorithms based on multivariate classification techniques or neural networks they are constructed for automatic recognition of structural properties from spectral data, or for simulation of spectra from structural properties [83]. Multivariate data analysis for spectrum interpretation is based on the characterization of spectra by a set of spectral features. A spectrum can be considered as a point in a multidimensional space with the coordinates defined by spectral features. Exploratory data analysis and cluster analysis are used to investigate the multidimensional space and to evaluate rules to distinguish structure classes. 

Perhaps the most interesting aspect of this set of studies is the question posed in the recent paper by Schmidt et al. (2004) and deals with the reality of the patterns they observed. Is the polymorphism observed a result of the calculation methods used in the study, neural network (NN), and multivariate statistical analysis (MVA) Would increased sampling result in a greater number of chemo-types It is entirely possible, of course, that the numbers obtained in this study are a true reflection of the biosynthetic capacities of the plants studied. The authors concluded—and this is a point made elsewhere in this review—that ... for a correct interpretation a good knowledge of the biosynthetic background of the components is needed. ... 

Wu et al. [46] used the approach of an artificial neural network and applied it to drug release from osmotic pump tablets based on several coating parameters. Gabrielsson et al. [47] applied several different multivariate methods for both screening and optimization applied to the general topic of tablet formulation they included principal component analysis and... 

Goodacre, R. Neal, M. J. Kell, D. B. Rapid and quantitative analysis of the pyrolysis mass spectra of complex binary and tertiary mixtures using multivariate calibration and artificial neural networks. Anal. Chem. 1994, 66,1070-1085. 

Goodacre, R. Neal, M. I Kell, D. B. Quantitative analysis of multivariate data using artificial neural networks A tutorial review and applications to the deconvolution of pyrolysis mass spectra. Zbl. Bakt. 1996,284, 516-539. 

Timmins, E. M. Goodacre, R. Rapid quantitative analysis of binary mixtures of Escherichia coli strains using pyrolysis mass spectrometry with multivariate calibration and artificial neural networks J. Appl. Microbiol. 1997,83, 208-218. 

Nilsson,T. Bassani, M. R. Larsen,T. O. Montanarella, L. Classification of species in the genus Penicillium by Curie point pyrolysis/mass spectrometry followed by multivariate analysis and artificial neural networks. J. Mass Spectrom. 1996, 31, 1422-1428. 

Neural networks are helpful tools for chemists, with a high classification and interpretation capacity. ANNs can improve and supplement data arrangements obtained by common multivariate methods of data analysis as shown by an example of classification of wine (Li-Xian Sun et al. [1997]). 

Use of multivariate approaches based on classification modelling based on cluster analysis, factor analysis and the SIMCA technique [98,99], and the Kohonen artificial neural network [100]. All these methods, though rarely implemented, lead to very good results not achievable with classical strategies (comparisons, amino acid ratios, flow charts) and, moreover it is possible to know the confidence level of the classification carried out. 

The four-volume Handbook of Chemoinformatics—From Data to Knowledge (Gasteiger 2003) contains a number of introductions and reviews that are relevant to chemometrics Partial Least Squares (PLS) in Cheminformatics (Eriksson et al. 2003), Inductive Learning Methods (Rose 1998), Evolutionary Algorithms and their Applications (von Homeyer 2003), Multivariate Data Analysis in Chemistry (Varmuza 2003), and Neural Networks (Zupan 2003). 

A more common use of informatics for data analysis is the development of (quantitative) structure-property relationships (QSPR) for the prediction of materials properties and thus ultimately the design of polymers. Quantitative structure-property relationships are multivariate statistical correlations between the property of a polymer and a number of variables, which are either physical properties themselves or descriptors, which hold information about a polymer in a more abstract way. The simplest QSPR models are usually linear regression-type models but complex neural networks and numerous other machine-learning techniques have also been used. 

There is a long history of efforts to find simple and interpretable /i and fi functions for various activities and properties (29, 30). The quest for predictive QSAR models started with Hammett s pioneer work to correlate molecular structures with chemical reactivities (30-32). However, the widespread applications of modern predictive QSAR and QSPR actually started with the seminal work of Hansch and coworkers on pesticides (29, 33, 34) and the developments of various powerful analysis tools, such as PLS (partial least squares) and neural networks, for multivariate analysis have fueled these widespread applications. Nowadays, numerous publications on guidelines, workflows, and... 

Reasonable noise in the spectral data does not affect the clustering process. In this respect, cluster analysis is much more stable than other methods of multivariate analysis, such as principal component analysis (PCA), in which an increasing amount of noise is accumulated in the less relevant clusters. The mean cluster spectra can be extracted and used for the interpretation of the chemical or biochemical differences between clusters. HCA, per se, is ill-suited for a diagnostic algorithm. We have used the spectra from clusters to train artificial neural networks (ANNs), which may serve as supervised methods for final analysis. This process, which requires hundreds or thousands of spectra from each spectral class, is presently ongoing, and validated and blinded analyses, based on these efforts, will be reported. 

The methods of data analysis depend on the nature of the final output. If the problem is one of classification, a number of multivariate classifiers are available such as those based on principal components analysis (SIMCA), cluster analysis and discriminant analysis, or non-linear artificial neural networks. If the required output is a continuous variable, such as a concentration, then partial least squares regression or principal component regression are often used [20]. 

---