Foundations and Novel Approaches in Data Mining

ISBN-13: 9783540283157 / Angielski / Twarda / 2005 / 378 str.

T. y. Lin; Tsau Young Lin; Setsuo Ohsuga

Opis

Data-mining has become a popular research topic in recent years for the treatment of the "data rich and information poor" syndrome. Currently, application oriented engineers are only concerned with their immediate problems, which results in an ad hoc method of problem solving. Researchers, on the other hand, lack an understanding of the practical issues of data-mining for real-world problems and often concentrate on issues that are of no significance to the practitioners. In this volume, we hope to remedy problems by (1) presenting a theoretical foundation of data-mining, and (2) providing important new directions for data-mining research. A set of well respected data mining theoreticians were invited to present their views on the fundamental science of data mining. We have also called on researchers with practical data mining experiences to present new important data-mining topics.

Spis treści

From the contents Part I: Theoretical Foundations. Commonsense Causal Modeling in the Data Mining Context. Definability of Association Rules in Predicate Calculus. A Measurement-Theoretic Foundation of Rule Interestingness Evaluation. Statistical Independence as Linear Dependence in a Contingency Table. Foundations of Classification.- Part II: Novel Approaches. SVM-OD: SVM Method to Detect Outliers. Extracting Rules from Incomplete Decision Systems: System ERID. Mining for Patterns Based on Contingency Tables by KL-Miner – First Experience. Knowledge Discovery in Fuzzy Databases Using Attribute-Oriented Induction. Rough Set Strategies to Data with Missing Attribute Values. Privacy-Preserving Collaborative Data Mining.- Part III: Novel Applications. Research Issues in Web Structural Delta Mining. Workflow Reduction for Reachable-path Rediscovery in Workflow Mining. Principal Component-based Anomaly Detection Scheme. Making Better Sense of the Demographic Data Value in the Data Mining Procedure.