Pattern recognition levels

As with all scientific writing there are various levels that can be presented. For example, infrared spectroscopy could be used on simply the pattern recognition level or at the more sophisticated level of quantum mechanics. So it is with physical adsorption. One can use the data from physical adsorption measurements as a simple control device, i.e. Does this powder have the right adsorption isotherm to meet production requirements , or on a different level What is the meaning of the isotherm in terms of surface and pore structure and chemical attractions For most applications, the level of sophistication is somewhat intermediate. 

Often the goal of a data analysis problem requites more than simple classification of samples into known categories. It is very often desirable to have a means to detect oudiers and to derive an estimate of the level of confidence in a classification result. These ate things that go beyond sttictiy nonparametric pattern recognition procedures. Also of interest is the abiUty to empirically model each category so that it is possible to make quantitative correlations and predictions with external continuous properties. As a result, a modeling and classification method called SIMCA has been developed to provide these capabihties (29—31). 

For less well defined incidents however, these detection systems may be inadequate. Portable chemical detectors may not be able to be deployed to the site, not detect the agen, or give inconclusive results. Clinical findings may be non-specific, present in an atypical manner, or for example in the case of sulphur mustard, have a latency period that delays firm pattern recognition. Due to the physico-chemical properties of the agent or the time between release and collection, environmental samples may have low agent levels or sufficiently high contaminants to prevent adequate results. 

Advanced mathematical and statistical techniques used in analytical chemistry are often referred to under the umbrella term of chemometrics. This is a loose definition, and chemometrics are not readily distinguished from the more rudimentary techniques discussed in the earlier parts of this chapter, except in terms of sophistication. The techniques are applied to the development and assessment of analytical methods as well as to the assessment and interpretation of results. Once the province of the mathematician, the computational powers of the personal computer now make such techniques routinely accessible to analysts. Hence, although it would be inappropriate to consider the detail of the methods in a book at this level, it is nevertheless important to introduce some of the salient features to give an indication of their value. Two important applications in analytical chemistry are in method optimization and pattern recognition of results. 

Stateful pattern recognition This method examines and compares the contents of certain key parts of an information packet against a database of acceptable information. Information traveling from inside the firewall to the outside is monitored for specific defining characteristics, then incoming information is compared to these characteristics. If the comparison yields a reasonable match, the information is allowed through. If not, the information is discarded. Provides a limited time window to allow pockets of information to be sent does not allow any direct connections between internal and external hosts supports user-level authentication. Slower than packet filtering does not support all types of connections. 

Chim. Acta. 103, 1978, 429443. Four levels of pattern recognition. 

Skill-based At the lowest level, these are routine skills of observation, hand-eye coordination, and control skills. Skills also include pattern recognition and actions that are manual, well known, and... 

Four levels of pattern recognition have been defined by Albano (2). Levels I and II are most frequently used to determine the similarity of objects, or to characterize clusters of samples and to classify unknown objects. Level III takes advantage of the reduction of data dimensions resulting from SIMCA and seeks to establish a correlation of sample scores with independent variables... 

The SIMCA approach can be applied in all of the four levels of pattern recognition. We focus on its use to describe complex mixtures graphically, and on its utility in quality control. This approach was selected for the tasks of developing a quality control program and evaluating similarities in samples of various types. Principal components analysis has proven to be well suited for evaluating data from capillary gas chromatographic (GC) analyses (6-8). 

Data have been collected since 1970 on the prevalence and levels of various chemicals in human adipose (fat) tissue. These data are stored on a mainframe computer and have undergone routine quality assurance/quality control checks using univariate statistical methods. Upon completion of the development of a new analysis file, multivariate statistical techniques are applied to the data. The purpose of this analysis is to determine the utility of pattern recognition techniques in assessing the quality of the data and its ability to assist in their interpretation. 

Feature selection is the process by which the data or variables liq>or-tant for class assignment are determined. In this step of a pattern recognition study the various methods differ considerably. In the hyperplane methods, the strategy is to begin with a block of variables for the classes, calculate a classification function, and test it for classification of the training set. In this initial phase, generally many more variables are included than are necessary. Variables are then detected in a stepwise process and a new rule is derived and tested. This process is repeated until a set of variables is obtained that will give an acceptable level of classification. 

Several conceptual and technical orderings based on the analytical information level can also be established (see Fig. 1.2). Reports contain information of the highest level in fact, in addition to the results, they provide an interpretation that is facilitated by chemometric techniques (e.g. those based on pattern recognition) and cooperation with other scientific and technical areas. In fact, reports provide answers to the problems addressed. Results are referred to samples and analytes, and are arrived at by chemo-... 

Chen, Q., Zhao, J., and Vittayapadung, S. (2008). Identification of the green tea grade level using electronic tongue and pattern recognition. Food Res. Int. 41(5), 500-504. 

Recent developments in the field of sensing airborne chemicals using electrochemical sensors and sensor arrays are reviewed. Such systems detect, Identify, and quantify potential chemical hazards to protect the health and safety of workers and citizens. The application discussed In this review article Is single chemicals at part-per-million levels in air. The sensor system consists of an array of sensors used In four modes of operation, and the data are Interpreted by a computer algorithm. Pattern recognition techniques are being used to understand the information content of the arrays and to focus future experimental work. 

Microelectrode arrays containing AChE were also utilised within a flow injection system [40]. A system was developed where a sample was separated and flushed simultaneously through eight cells, each containing a screen-printed electrode and fitted with a separate bespoke mini-potentiostat (Fig. 15.3). This allowed multiple measurements to be made on a single water sample using multiple electrodes, each specific for a different pesticide due to inclusions of different AChE mutants in each of the electrodes. Pattern-recognition software could then be utilised to deduce the pesticide levels in a potentially complex sample. 

CHERNOFF [1973] created an unusual graphical representation of multivariate data. Fie made the assumption that the human pattern recognition ability is best trained with human faces. Faces can be described with parameters like face width, ear level, half-face height, eccentric upper face, eccentric lower face, nose length, mouth centering, etc. 

Nestrick, T.J., Lamparski, L.L., Townsend D.I. (1980) Identification of tetrachlorodibenzo-p-dioxin isomers at the 1-ng level by photolytic degradation and pattern recognition techniques. Anal. Chem. 52, 1865-1874. 

---