Classification, Data Analysis, and Knowledge Organization

In science, industry, public administration and documentation centers large amounts of data and information are collected which must be analyzed, ordered, visualized, classified and stored efficiently in order to be useful for practical applications. This volume contains 50 selected theoretical and applied papers presenting a wealth of new and innovative ideas, methods, models and systems which can be used for this purpose. It combines papers and strategies from two main streams of research in an interdisciplinary, dynamic and exciting way: On the one hand, mathematical and statistical methods are described which allow a quantitative analysis of data, provide strategies for classifying objects or making exploratory searches for interesting structures, and give ways to make comprehensive graphical displays of large arrays of data. On the other hand, papers related to information sciences, informatics and data bank systems provide powerful tools for representing, modelling, storing and retrieving facts, data and knowledge characterized by qualitative descriptors, semantic relations, or linguistic concepts. The integration of both fields and a special part on applied problems from biology, medicine, archeology, industry and administration assure that this volume will be informative and useful for theory and practice.

Contenu

I: Mathematical and statistical methods for classification and data analysis.- Classification and clustering methods.- An agglomerative method for two-mode hierarchical clustering.- Selection from overlapping classifications.- On cluster methods for qualitative data II.- A regression analytic modification of Ward's method: A contribution to the relation between cluster analysis and factor analysis.- The "partition with a structure" concept in biological data analysis.- Classification with neural networks.- Statistical and probabilistic aspects of clustering and classifications.- Multigraphs for the uncovering and testing of structures.- Estimators and relative efficiencies in models of overlapping samples.- Lower bounds for the tail probabilities of the scan statistic.- Poisson approximations of image processes in computer tomography.- Statistical, geometrical and algebraic methods for data analysis.- Some recent developments in linear models: A short survey (plenary lecture).- Causal analysis in marketing research with LISREL or a combination of traditional multivariate methods?.- Analysis of data measured on a lattice.- Dual algorithms in multidimensional scaling.- Comparison of biplot analysis and formal concept analysis in the case of a repertory grid.- Convexity in ordinal data.- Classification and seriation by iterative reordering of a data matrix.- Data analysis based on a conceptual file.- II: Knowledge organization, data bases, and information retrieval.- Modelling, representation and organization of conceptual knowledge.- Decentralized modelling of data and relationships in enterprises.- A contribution to the examination of semantic relations between lexemes.- A mathematical model for conceptual knowledge systems.- Compositional semantics and concept representation 163.- Data bases, expert systems, information retrieval, and library systems.- Small and beautiful? Some remarks on evaluating microcomputer based library systems.- A tool for validating PROLOG programs.- On the database component in the knowledge-based system WIMDAS.- Information retrieval techniques in rule-based expert systems.- Object databases and thesauri for small museums (plenary lecture).- Terminology and classification.- The structure and role of specialized information in scientific and technical terminologies.- Terminology work in the World Health Organization: EUROTERM abbreviations.- HyperTerm - A proposal for a user-friendly termbank.- The role of classification in terminology documentation (plenary lecture).- III: Applications and methods for special subject fields.- Classification, systematics, and evolution in biology.- The hierarchy of organisms: Systematics and classification in biology (plenary lecture).- Estimating phylogenies with invariant functions of data (plenary lecture).- Statistical analysis of genetic distance data.- Variance estimation in the additive tree model.- Classification and documentation in medicine.- Semi-automated classification of medical phrases using a personal computer.- Structure of informations on medical trials.- Recent problems and longterm activities in the classification of medical concepts.- Exploring three-dimensional image data with classification methods.- Data analysis in the archeological and historical sciences.- The reconstruction of "genetic kinship" in prehistoric burial complexes - Problems and statistics.- An approach to a formal statistical analysis of historical data based on the town of Bamberg.- Automatic syntax analysis of meroitic funeral inscriptions.- Application of computers in historical-topographical research: A database for travel reports on Greece (18th and 19th century).- The use of multivariate statistics in Scandinavian archeology.- The application of correspondence analysis: some examples in archeology.- An analysis of beads found in the Merovingian cemetery of Weingarten.- Classification in industry: Coding systems and commodity description.- Bank code numbers as defining arguments and controlling tools in automated payments.- From commodity description to expert systems.- Tabular layouts of article characteristics and formal concept analysis (in German).- The postcode, a local and routing code for the transport of mail items.