Chemometrics for material analysis

Chemometric methods are described that provide effective means to enhance the information of analytical methods in material analysis and to enable knowledge engineering tools to be applied within the field. Typical applications are outlined for depth-profile analysis, for pattern recognition on ceramic systems as well as for the utilization of expert systems.     

Anwendbarkeit und Anwendung von Expertensystemen in der Analytik

Summary General prerequisites of problem definition and solving by man and machine are discussed. The advantage of expert systems over books and data base systems to have a capability of understanding problems is pointed out. The chemical knowledge needs to be restructured and cast into operational terms to prepare it for formalized reasoning. Writing program routines in artificial intelligence by a chemist means wasting time and makes bad use of his chemical expertise. The commercial empty shells TWAICE, INSIGHT 2 + and 1st CLASS have been used to build expert systems in the domains of slag analysis and Karl-Fischer titrations. Examples of method selection for routine problems, compilation of new analytical methods from computer based knowledge and of strictly user oriented trouble shooting in analytical chemistry are given.     

Problem solving by expert systems in analytical chemistry

Summary It is expected that artificial intelligence and more particularly, expert systems, will have a profound effect upon analytical chemistry. This paper provides an overview of this rapidly evolving field. General aspects of expert systems are dealt with first, such as knowledge representation, knowledge manipulation, dealing with uncertainty, and the application of software tools to facilitate the construction of expert systems. The accomplishments with respect to the application of expert systems to chemistry — particularly analytical chemistry — are discussed. Future directions of the research and applications are forecasted.

---

Expert Systems zur Problemlösung in der analytischen Chemie

Zusammenfassung Es ist zu erwarten, daß künstliche Intelligenz und besonders expert systems einen großen Einfluß auf die analytische Chemie ausüben werden. In dieser Arbeit wird ein Überblick über dieses in schneller Entwicklung befindliche Gebiet gegeben. Zunächst werden die expert systems im allgemeinen behandelt (Darstellung und Handhabung von Wissen, Behandlung von Unsicherheiten, Anwendung von Software zur Konstruktion der expert systems). Anschließend werden Anwendungen in der Chemie — speziell der analytischen Chemie — diskutiert und zukünftige Entwicklungen aufgezeigt.

     

Expert systems for method development and validation in HPLC

ESCA, Expert Systems Applied to Chemical Analysis, started its research in March 1987, with the aim of building prototype expert systems for HPLC method development. Results of this research have been published as the work has progressed. The project is now completed and this paper summarises some of the overall project conclusions. Seven different expert systems have been built which tackle problems throughout the process of method development, four stand-alone systems and three integrated systems. The object of ESCA was to evaluate the applicability of expert system technology to analytical chemistry and not all the systems were built for commercial uses. Many of the systems tackle problems specific to one or more of the partners and thus may not be useful outside this environment. However, the results of the work are still pertinent to analysts wishing to build their own systems. These results are described, however, the emphasis of the paper is on those systems developed for method validation.Method validation for HPLC is a complex task which requires many characteristics of the method to be tested, e.g. accuracy, precision, etc. The expert systems built within ESCA concern the validation of precision. Two systems were developed for repeatability testing and ruggedness testing. The method validation process can be divided into several discrete stages, these include: (1) The selection of the method feature to test, for instance which factors can influence the ruggedness of a method. (2) The definition of a test procedure, for instance an efficient statistical design. (3) The execution of experiments and the interpretation of results. (4) A diagnosis of any observed problem. This paper describes these two systems in some detail and summarises some of the results obtained from their evaluation. It concludes that expert systems can be useful in solving analytical problems and the integration of several expert systems can provide extremely powerful tools for the analyst.     

GESA--a two-dimensional processing system using knowledge base techniques

The successful analysis of two-dimensional (2-D) polyacrylamide electrophoresis gels demands considerable experience and understanding of the protein system under investigation as well as knowledge of the separation technique itself. The present work concerns the development of a computer system for analysing 2-D electrophoretic separations which incorporates concepts derived from artificial intelligence research such that non-experts can use the technique as a diagnostic or identification tool. Automatic analysis of 2-D gel separations has proved to be extremely difficult using statistical methods. Non-reproducibility of gel separations is also difficult to overcome using automatic systems. However, the human eye is extremely good at recognising patterns in images, and human intervention in semi-automatic computer systems can reduce the computational complexities of fully automatic systems. Moreover, the expertise and understanding of an "expert" is invaluable in reducing system complexity if it can be encapsulated satisfactorily in an expert system. The combination of user-intervention in the computer system together with the encapsulation of expert knowledge characterises the present system. The domain within which the system has been developed is that of wheat grain storage proteins (gliadins) which exhibit polymorphism to such an extent that cultivars can be uniquely identified by their gliadin patterns. The system can be adapted to other domains where a range of polymorpic protein sub-units exist. In its generalised form, the system can also be used for comparing more complex 2-D gel electrophoretic separations.     

色谱专家系统的应用和发展

随着科学技术的发展,色谱专家系统作为一种人工智能方法亦在不断发展,并在实际工作中发挥着越来越重要的作用。该文综述了作者的研究小组在色谱专家系统研究方面的进展,重点介绍从研究气相色谱专家系统开始,到应用色谱专家系统的思想,运用气相色谱、液相色谱和毛细管电泳新技术解决石化、环境、疾病诊断、基因分析和药物分析等领域实际问题的工作。引用相关文献64篇。     

Knowledge-based systems for analytical chemistry

Analytical chemists have made extensive use of computer technology, automating most analytical instruments, many analyses and reporting activities. We now need to expand the areas to which computers can be applied by addressing problems of greatly increased complexity. These problems fall into two general classes, the first class of problems are those thatcannot be solved using only first principle information and, the second class are those problems thatcan be solved using only first principles, but that are so complex that the traditional approach is often not cost effective.The discussion will center on how artificial intelligence technology (AI) can provide the means for using heuristics together with first principle information to solve instances of the first class of problems. The knowledge required to provide the solution is formulated as facts, rules (heuristics) and an inference engine.The same AI technology can also be used to refine specifications and provide cost effective solutions for very complex problems involving only first principle information. For this class of problems an AI work station can provide the software development conductive to rapid prototyping and specification refinement.Discussion of several expert systems will be used to describe the capabilities and features of rule-based systems. The strengths and weaknesses of one system, use of IR spectra for structure elucidation, will be examined in some detail. It is an expert system consisting of rule sets organized into logic trees, thus it will provide an opportunity to describe how the use of more advanced AI technology could further improve the program's performance.     

Fuzzy theory in analytical chemistry

The concept of fuzzy theory is described in order to provide the analyst with the means for dealing with vague statements, uncertain observations or the fuzziness of human perception and interpretation, in general. In a theoretical part, basic notions of fuzzy theory are given, such as types of membership functions, operations with fuzzy sets, definitions of fuzzy numbers, points, functions, and relations, and the use of linguistic variables. The difference between fuzziness and probability is outlined. The applications section demonstrates advantages of fuzzy theory methods compared to common mathematical methods with respect to data handling for calibration of analytical methods, to classification of Chromatographie and spectroscopic patterns, to component identification and multicomponent analysis, and to designing fuzzy expert systems for selection of analytical procedures.     

Identification of novel illicit amphetamines from vapor-phase FTIR spectra - a chemometrical solution

A computer-aided procedure automating the identification of illicit amphetamine analogs eluting from a gas chromatograph coupled to a Fourier transform infrared spectrometer is presented. The expert system discriminates novel amphetamines from other classes of drugs of abuse normally screened in illicit tablets or powders. The main analytical advantages of the system over the automated procedures dedicated to general unknown analysis are the objectivity and the accuracy in predicting the class identity of the compound (i.e. stimulant, hallucinogen) when the reference spectrum is not present in the spectral library. The expert system uses quantitative thresholds defining the similarity of the unknown to the classes of illicit amphetamines and checks the presence of the molecular skeletons associated with different psychotropic effects of amphetamines. The challenge in building the system was the fuzziness of vapor-phase Fourier transform infrared spectrometer spectra of low-weight molecules such as amphetamines. This paper emphasizes the chemometrical techniques found most appropriate for modeling such spectral behavior. An exploratory (principal component) analysis indicated the sample preparation and the feature weight function yielding the best input for the knowledge base. The class identity of a compound was assigned using Soft Independent Modeling of Class Analogy. A rule-based decision system was implemented to enhance the accuracy in identity assignment. The flow diagram optimizing the knowledge base content of each model is presented. Finally, up to 81.13% (out of 159 tested compounds) were classified with a 5% confidence level. The total correct classification rate was 93.93%, for a yield of 96.30% true positive amphetamines.     

Fractals and analytical chemistry — an old and a new look

Summary The impact of the concept of fractals on the foundations of analytical chemistry is discussed. Examples from such diverse subfields like geochemistry, reactions in turbulent flow systems or particle technology illustrate the new look, quantifying what was so far left to intuition of the expert — complexity. These developments led to philosophical problems resolvable in the unity of arts and sciences.