Classification methods, exploratory data

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. 

A whole spectrum of statistical techniques have been applied to the analysis of DNA microarray data [26-28]. These include clustering analysis (hierarchical, K-means, self-organizing maps), dimension reduction (singular value decomposition, principal component analysis, multidimensional scaling, or correspondence analysis), and supervised classification (support vector machines, artificial neural networks, discriminant methods, or between-group analysis) methods. More recently, a number of Bayesian and other probabilistic approaches have been employed in the analysis of DNA microarray data [11], Generally, the first phase of microarray data analysis is exploratory data analysis. 

Gordon, A. E. (1981). Classification Methods for Exploratory Analysis of Multivariate Data. Chapman and Hall, New York. 

Once the features are defined and the compounds are coded according to their presence or absence a classification of the compounds regarding their activity should be performed. This is always a good approach even if continuous activity data are available. The classification into an active and inactive class will increase the clearness of the analysis and will speed up the procedure. If there are not obvious limits the methods of exploratory data analysis can be used to refine meaningful threshold values. The coded stmctures together with the classification and (if known) continuous activity data will form the input information of the EVAL procedure. 

Exploratory analysis of spectral data by PCA, PLS, cluster analysis, or Kohonen mapping tries to get an insight into the spectral data structure and into hidden factors, as well as to find clusters of similar spectra that can be interpreted in terms of similar chemical structures. Classification methods, such as LDA. PLS, SIMCA, KNN classification, and neural networks, have been used to generate spectral classifiers for an automatic recognition of structural properties from spectral data. The multivariate methods mostly used for spectra prediction (mainly NMR. rarely IR) are neural networks. Table 6 contains a summary of recent works in this field (see Infrared Data Correlations with Chemical Structure). 

The second part of the book—Chapters 9-12— presents some selected applications of chemometrics to different topics of interest in the field of food authentication and control. Chapter 9 deals with the application of chemometric methods to the analysis of hyperspectral images, that is, of those images where a complete spectrum is recorded at each of the pixels. After a description of the peculiar characteristics of images as data, a detailed discussion on the use of exploratory data analytical tools, calibration and classification methods is presented. The aim of Chapter 10 is to present an overview of the role of chemometrics in food traceability, starting from the characterisation of soils up to the classification and authentication of the final product. The discussion is accompanied by examples taken from the different ambits where chemometrics can be used for tracing and authenticating foodstuffs. Chapter 11 introduces NMR-based metabolomics as a potentially useful tool for food quality control. After a description of the bases of the metabolomics approach, examples of its application for authentication, identification of adulterations, control of the safety of use, and processing are presented and discussed. Finally, Chapter 12 introduces the concept of interval methods in chemometrics, both for data pretreatment and data analysis. The topics... 

Cluster analysis is an exploratory analysis, but in the process of classification, it is not necessary to give a classification standard, nonetheless, cluster analysis can proceed from the sample data, and classify automatically. Different cluster analysis methods often come to a different conclusion. 

---