Data Mining: Data Mining, Text Mining and Their Business Applications(No. 35 WIT Transactions on Information and Communication Technologies)

This book contains most of the papers presented at the Sixth International Conference on Data Mining held in Skiathos, Greece. Twenty-five countries from all the continents are represented in the papers published in the book, offering a real multinational and multicultural range of experiences and ideas. There has been an explosion of interest in data mining applications to unstructured data and this is reflected in a large increase in text mining. Consequently it is not difficult to forecast that in the next months several applications will appear dedicated to analysis of content coming from the billions of available web pages, newsgroups, emails, chat lines and message boards, and that the interest in unstructured data mining and text mining will grow amongst researchers, OEM and system integrators working in sectors as information retrieval, semantic web, linguistics, and knowledge management. In business applications, the most promising areas are those regarding National Security, Competitive Intelligence and Customer Relationship Management.They can all be incorporated into the area of Intelligence Analysis, of especial interest for textual data mining and text mining applications. Contents include: Data mining; Text mining; Neural networks and decision trees; Link analysis; Clustering and categorization; Consumer and strategic intelligence; Applications in science, engineering and life sciences, and Applications in business, industry and government.

Table of Contents:
Part I: Methodological Approaches; Section 1: Data mining; Learning networks for tornado forecasting: a Bayesian perspective; Outlier detection based on projection-based ordering; On extending F-measure and G-mean metrics to multi-class problems; Multivariate interdependent discretization in discovering the best correlated attribute; Estimation and extension of the Stochastic Schemata Exploiter; Decision making on operational data: a remote approach to distributed data monitoring; A multi-strategy approach for mining multimedia data repositories; Section 2: Text mining; A multi-criteria decision making approach in feature selection for enhancing text categorization; Multilingual text mining; The protein ontology project: structured vocabularies for proteins; Text mining for stock movement predictions: a Malaysian perspective; Medical communication quality in the Italian pharmaceutical industry: measurement and analysis by 'NOOS'; A comparison of two algorithms for discovering repeated word sequences; A genetic algorithm for text mining; The process of sensemaking on the telework virtual community using text mining; Knowledge discovery in large text databases using the MST algorithm; Textual document pre-processing and feature extraction in OLEX; Naive rule induction for text classification based on key-phrases; Renovation of terms adjustment and effective model combination impact on information retrieval performance; Linguistic summaries on small screens; Rule discovery in Web-based educational systems using Grammar-Based Genetic Programming; Part II: Techniques; Section 3: Neural networks and decision trees; Multi-relational data mining in Microsoft SQL Server'; Neural network models for the development and evaluation of new fuels; CC4.5: cost-sensitive decision tree pruning; Cooling Growing Grid: an incremental self-organizing neural network for data exploration; Pleiotropic microarray gene expression data: advanced tandem multivariate data mining; A decision tree classifier for vehicle failure isolation; Section 4: Link analysis; A method for generating aggregated associations between discrete data features; X3-Miner: mining patterns from an XML database; Section 5: Clustering and categorisation; Mining association rules from qualitative and quantitative clustering; HyperClustering: from the digital divide to a GRID e-workspace; DEA implementation and clustering analysis using the K-Means algorithm; A hybrid method to categorize HTML documents; Part III: Applications; Section 6: Consumer and strategic intelligence; Application of technology prospective to business sectorial studies; Discovering common interests and problems to improve working conditions at a large company; The use of knowledge discovery techniques for behavioural scoring; Providing database encryption as a scalable enterprise infrastructure service; National Security and threat awareness; Measuring user satisfaction with intelligent agents: an exploratory study; A clustering approach for knowledge discovery in database marketing; Section 7: Applications in science, engineering and life sciences; Evaluation of clinical prediction rules using a convergence of knowledge-driven and data-driven methods: a semio-fuzzy approach; Evolving neural networks to flow cytometric data interpretation; Classification algorithms and analyzing the functionality of protein families; Mining GPS logs to augment location models; An adaptive Bayesian classification for real-time image analysis in real-time particle monitoring for polymer film manufacturing; Section 8: Applications in business, industry and government; Mining effective design solutions based on a model-driven approach; Application of fuzzy models and neural models in financial time series; E-commerce models for banks' profitability; Sarbanes-Oxley, Basel II, and data mining opportunities in compliance systems; Survival data mining in the telecommunications industries: usefulness and complications; Data mining methods in a metrics-deprived inventory transactions environment; Ecological mining - a case study on dam water quality; Improving effectiveness of Web sites using incremental data mining over clickstreams; Data mining education for external auditors