Data Mining

Data mining is an interdisciplinary field dealing with the discovery of hidden data and unexpected patterns and rules in large databases. In this book, containing the proceedings of the second International Conference on Data Mining, researchers and applications developers from academia and industry present research and practical experiences in the diverse areas which make up data mining. They include computer experts, statisticians, knowledge acquisition specialists, data analysts, IT consultants, data visualization experts as well as users and developers. Topics covered include: applications of data mining in science, engineering, business, industry and medicine; Internet applications; fraud detection and prevention; software; neural networks and decision trees; parallel and distributed techniques; and case studies.

Table of Contents:
Section 1: Applications of Data Mining in Science, Engineering, Business, Industry and Medicine - Web mining through the online analyst; Bayesian networks for knowledge discovery in a database from the program for genetic improvement of the Nelore breed; Using a data mining workbench for micro and macro economic modelling; Small business modeling within the financial accounting conceptual framework; Paperwork reduction by means of data mining; CRM in a real-world insurance company; A data-mining alternative to model hospital operations: clinical costs and predictions; Optimal entrocopy encoders for mining multiply resolved data; Acquisition of KANSEI based on fuzzy inference and multivaritate analysis; Data mining for database marketing at Garanti Bank; Use of equation discovery: Oscillating flow in a U-tube; The use of learning classifier systems in the direct marketing industry. Section 2: Data Warehousing and Databases - KR and model discovery from active DB with predictive logic; Metadata - based data auditing; An analysis of the integration between data mining applications and database systems; Design of a data warehouse to support the management of academic institutions. Section 3: Internet Applications - Automatic construction of ontology from text databases; Attractability: an indicator for optimising the design of a web site. Section 4: Data Mining Methodologies - A new algorithm for finding association rules; Local feature selection for heterogeneous problems; Content in context: a data-driven approach. Section 5: Knowledge Discovery and Data Mining - An architecture for ABEN-KDD- an agent-based environment for knowledge discovery in databases; Discovering graph structures in high dimensional spaces; Discovering salient data features by composing and manipulating logical equations; A qualitative spatial reasoning approach in knowledge discovery in spatial databases; A fuzzy - based conceptual KDD approach: the SaintEtiQ system; Supervised knowledge discovery from incomplete data; Data scale reduction via instances summarization using the Rough Set Theory; Data mining in temporal sequences: a technique based on MC; Data mining telecommunications network data for fault management and development testing; Higher order mining: modelling and mining the results of knowledge discovery; A computational environment for extracting rules from databases; Considerations about the effectiveness of inductive learning process in data-mining context. Section 6: Neural Networks and Decision Trees - The deterministic evolutionary learning algorithm; Wrapped feature selection for binary classification Bayesian regularisation neural networks: a database marketing application; Web text mining using a hybrid intelligent system based on KDT, expert system and neural network; Alleviating the complexity of the Combinatorial Neural Model using a committee machine; Application of decision tree classifiers to computer intrusion detection; Effects of attribute selection measures and sampling policies on functional structures of decision trees. Section 7: Genetic Algorithms and Parallel Techniques - Credit approval by a clustering genetic algorithm; Designing optimized pattern recognition systems by learning Voronoi vectors using genetic algorithms; Scalable parallel algorithms for predictive modelling. Section 8: Visualisation in Data Mining - Interactive rule-network layout with a genetic algorithm in a knowledge discovery process; Visualisation for Data Mining telecommunications network data; InfoZoom - Analysing Formula One racing results with an interactive data mining and visualisation tool. Section 9: Clustering and Classification Techniques - Evolving TSK fuzzy rules for classification tasks by Genetic Algorithms; Hierachical clustering for data mining by RBF network; Detecting visual feature importance via tree classifiers. An experience; Input dependent misclassification costs for cost-sensitive classifiers; Undirect knowledge discovery by using singular value decomposition; A clustering algorithm using the tabu search approach with simulated annealing; Cluster generation using tabu search based maximum descent algorithm; Stabilization of regression trees; Influence of lossy compression on hyperspectral image classification accuracy. Section 10: Tools for Pattern Discovery - Modeling dynamical systems by recurrent neural networks; A visual data mining tool to support cooperative learning; A new insight into the algebraic structure of the exponential smoothing algorithm of Brown. Section 11: Case Studies - A case based reasoning framework to extract knowledge from data; Mining customer preference ratings for product recommendation using the support vector machine and the latent class model; Data mining a large health insurance database; Case study of a retail bank marketing datamart development.