Pattern-recognition techniques

Fig. 3. Types of pattern recognition techniques (a) preprocessing, (b) display, (c) unsupervised learning, and (d) supervisediearning.

P. B. Harrington, K. J. Voorhees, T. E. Street, F. Radicati di Brozolo, and R. W. Odom. Anal Chem. 61, 715, 1989. Presents a discussion of LIMS polymer analysis and pattern recognition techniques. 

Thus, they share exactly the same solution (H) and performance criteria (y ) spaces. Furthermore, since their role is simply to estimate y for a given X, no search procedures S are attached to classical pattern recognition techniques. Consequently, the only element that dilfers from one classification procedure to another is the particular mapping procedure / that is used to estimate y(x) and/ or ply = j x). The available set of (x, y) data records is used to build /, either through the construction of approximations to the decision boundaries that separate zones in the decision space leading to different y values (Fig. 2a), or through the construction of approximations to the conditional probability functions, piy =j ). 

Metabolomics studies the entire metabolism of an organism. It is possible to consider characterising the complex pattern of cellular proteins and metabolites that are excreted in urine. Pattern recognition techniques of nuclear magnetic resonance spectra have been applied to determine the dose-response using certain classical liver and kidney toxicants (Robertson et al, 2000). This could well provide a signature of the functional state of the kidney, and perturbations in the pattern as a result of exposure to a chemical could be observed. But first it would be necessary to understand how compounds with known effects on the kidney affect these processes. 

D. Coomans and D.L. Massart, Potential methods in pattern recognition. Part 2. CLUPOT an unsupervised pattern recognition technique. Anal. Chim. Acta., 133 (1981) 225-239. 

Supervised pattern recognition techniques essentially consist of the following steps. 

While principal components models are used mostly in an unsupervised or exploratory mode, models based on canonical variates are often applied in a supervisory way for the prediction of biological activities from chemical, physicochemical or other biological parameters. In this section we discuss briefly the methods of linear discriminant analysis (LDA) and canonical correlation analysis (CCA). Although there has been an early awareness of these methods in QSAR [7,50], they have not been widely accepted. More recently they have been superseded by the successful introduction of partial least squares analysis (PLS) in QSAR. Nevertheless, the early pattern recognition techniques have prepared the minds for the introduction of modem chemometric approaches. 

MIR techniques have simplified obtaining infrared spectra of many materials important in packaging. These include rubber, plastics, laminations, and components of these materials that find use in pumps, sample packages, and devices. The combination of MIR and computerized pattern recognition techniques can be used for differentiating and classification of flexible packaging polymers such as polyvinyl chloride (PVC), polyvinylidene chloride (PVdC), acrylonitrile (Barex), and CTFE (Aclar) [22]. 

H. Liu. Classifications of PYC for pharmaceutical blister packaging using pattern recognition techniques. Ph.D. dissertation, Rutgers University, New Brunswick, NJ, 1998. 

Macdonald, C.R., R.J. Norstrom, and R. Turle. 1992. Application of pattern recognition techniques to assessment of biomagnification and sources of polychlorinate multicomponent pollutants, such as PCBs, PCDDs, and PCDFs. Chemosphere 25 129-134. 

The high dimensional nature of LIBS signals can lead to several computational issues when used in conjunction with many machine learning techniques. Dimensionality reduction is the process by which the high dimensional signals are mapped into a lower dimensional space. The resulting lower dimensional space can enable more robust performance when used in conjunction with pattern recognition techniques. 

In order to establish such a correlation, however, a statistical analysis of a very large number of patterns would be necessary. This is one possible area for application for the pattern recognition techniques mentioned above. For thin single crystal substrates, any epitaxial relationship of the metal particles to the support is clearly evidenced because the patterns are superimposed in nanodiffraction. A comparison can be made of the patterns obtained with the beam on and just off the particle. 

Liddell, R. W., Jurs, P. C. Anal. Chem. 46, 1974, 2126-2130. Interpretation of infrared spectra using pattern recognition techniques. 

Landis WG, Matthews RA (1993) Development of pattern recognition techniques for the evaluation of toxicant impacts to multispecies systems. Government Reports Announcements Index (GRA I) Issue 21 NTIS/AD-A267 197/2, p 255... 

Lavine BK, Stine AB, Qin XH (1997) Application of pattern recognition techniques to problems in advanced pollution monitoring. Government Reports Announcements Index (GRA I) Issue 02 NTIS/AD-A313 960/7, p 238... 

Soft Independent Method of Class Analogy (SIMCA), a pattern recognition technique based on principal components (25) was selected to evaluate and apply to the problems of establishing similarities among sample residue profiles. The development of a laboratory data management system to assist in the calculation and organization of results greatly enhanced the feasibility of this approach (26). 

Valafar R 2002. Pattern recognition techniques in microarray data analysis a survey. Ann NY Acad Sci 980 41. 

Several statistical and pattern recognition techniques were used to unravel the relationships between chemical reactivity data and the previously described effects which influence them. 

The multivariate tools typically used for the NIR-CI analysis of pharmaceutical products fall into two main categories pattern recognition techniques and factor-based chemometric analysis methods. Pattern recognition algorithms such as spectral correlation or Euclidian distance calculations basically determine the similarity of a sample spectrum to a reference spectrum. These tools are especially useful for images where the individual pixels yield relatively unmixed spectra. These techniques can be used to quickly define spatial distributions of known materials based on external reference spectra. Alternatively, they can be used with internal references, to locate and classify regions with similar spectral response. 

N.K. Shah and P.J. Gemperhne, Comhination of the Mahalanobis distance and residual variance pattern recognition techniques for classification of near-infrared reflectance spectra, Anal. Chem., 62, 465-470 (1990). 

Y. Woo, H. Kim and J. Cho, Identification of herbal medicines nsing pattern recognition techniques with near-infrared reflectance spectra, Microchem. J., 63, 61-70 (1999). 

These applications demonstrate that pattern recognition techniques based on principal components may be effectively used to character zate complex environmental residues. In comparisons of PCBs in bird eggs collected from different regions, we demonstrated through the use of SIHCA that the profiles in samples from a relatively clean area differed in concentration and composition from profiles in samples from a more highly contaminated region. Quality control can be evaluated by the proximity of replicate analysis of samples in principal components plots. 

Evaluating Data Quality in Large Data Bases Using Pattern-Recognition Techniques... 

---