Graham J. Williams / Simeon J. Simoff

Data Mining

Theory, Methodology, Techniques, and Applications
Herausgegeben:Williams, Graham J.; Simoff, Simeon J.

• Broschiertes Buch

Produktbeschreibung

This volume provides a snapshot of the current state of the art in data mining, presenting it both in terms of technical developments and industrial applications. The collection of chapters is based on works presented at the Australasian Data Mining conferences and industrial forums. Authors include some of Australia's leading researchers and practitioners in data mining. The volume also contains chaptersby regional and international authors. The original papers were initially reviewed for the workshops, conferences and forums.
The 25 articles in this state-of-the-art survey were carefully reviewed and selected from numerous contributions during at least two rounds of reviewing and improvement for inclusion in the book. They provide an interesting and broad update on current research and development in data mining. The book is divided into two parts. It starts with state-of-the-art research papers organized in topical sections on methodological advances, data linkage, text mining, and temporal and sequence mining. The second part comprises papers on state-of-the-art industrial applications from the fields ofhealth, finance and retail.

Produktdetails

• Lecture Notes in Computer Science 3755
• Verlag: Springer / Springer Berlin Heidelberg / Springer, Berlin
• Artikelnr. des Verlages: 978-3-540-32547-5
• Seitenzahl: 331
• Englisch
• Abmessung: 233mm x 155mm
• Gewicht: 540g
• ISBN-13: 9783540325475
• ISBN-10: 3540325476
• Artikelnr.: 21828513

Autorenporträt

Graham J. Williams, Togaware Data Mining, Macquarie, Australia / Simeon J. Simoff, University of Technology, Sydney, Australia

Inhaltsangabe

1: State-of-the-Art in Research.- Generality Is Predictive of Prediction Accuracy.- Visualisation and Exploration of Scientific Data Using Graphs.- A Case-Based Data Mining Platform.- Consolidated Trees: An Analysis of Structural Convergence.- K Nearest Neighbor Edition to Guide Classification Tree Learning: Motivation and Experimental Results.- Efficiently Identifying Exploratory Rules' Significance.- Mining Value-Based Item Packages - An Integer Programming Approach.- Decision Theoretic Fusion Framework for Actionability Using Data Mining on an Embedded System.- Use of Data Mining in System Development Life Cycle.- Mining MOUCLAS Patterns and Jumping MOUCLAS Patterns to Construct Classifiers.- A Probabilistic Geocoding System Utilising a Parcel Based Address File.- Decision Models for Record Linkage.- Intelligent Document Filter for the Internet.- Informing the Curious Negotiator: Automatic News Extraction from the Internet.- Text Mining for Insurance Claim Cost Prediction.- An Application of Time-Changing Feature Selection.- A Data Mining Approach to Analyze the Effect of Cognitive Style and Subjective Emotion on the Accuracy of Time-Series Forecasting.- A Multi-level Framework for the Analysis of Sequential Data.- 2: State-of-the-Art in Applications.- Hierarchical Hidden Markov Models: An Application to Health Insurance Data.- Identifying Risk Groups Associated with Colorectal Cancer.- Mining Quantitative Association Rules in Protein Sequences.- Mining X-Ray Images of SARS Patients.- The Scamseek Project - Text Mining for Financial Scams on the Internet.- A Data Mining Approach for Branch and ATM Site Evaluation.- The Effectiveness of Positive Data Sharing in Controlling the Growth of Indebtedness in Hong Kong Credit Card Industry.