Keynote Speeches and Invited Talk -- Perspective on Data Mining from Statistical Viewpoints -- Inductive Databases and Knowledge Scouts -- Hyperlink-Aware Mining and Analysis of the Web -- Data Mining Theory -- Polynomial Time Matching Algorithms for Tree-Like Structured Patterns in Knowledge Discovery -- Fast Discovery of Interesting Rules -- Performance Controlled Data Reduction for Knowledge Discovery in Distributed Databases -- Minimum Message Length Criterion for Second-Order Polynomial Model Discovery -- Frequent Itemset Counting Across Multiple Tables -- Frequent Closures as a Concise Representation for Binary Data Mining -- An Optimization Problem in Data Cube System Design -- Exception Rule Mining with a Relative Interestingness Measure -- Feature Selection and Transformation -- Consistency Based Feature Selection -- Feature Selection for Clustering -- A Simple Dimensionality Reduction Technique for Fast Similarity Search in Large Time Series Databases -- Missing Value Estimation Based on Dynamic Attribute Selection -- On Association, Similarity and Dependency of Attributes -- Clustering -- Prototype Generation Based on Instance Filtering and Averaging -- A Visual Method of Cluster Validation with Fastmap -- COE: Clustering with Obstacles Entities A Preliminary Study -- Combining Sampling Technique with DBSCAN Algorithm for Clustering Large Spatial Databases -- Predictive Adaptive Resonance Theory and Knowledge Discovery in Databases -- Improving Generalization Ability of Self-Generating Neural Networks Through Ensemble Averaging -- Application of Data Mining -- Attribute Transformations on Numerical Databases -- Efficient Detection of Local Interactions in the Cascade Model -- Extracting Predictors of Corporate Bankruptcy: Empirical Study on Data Mining Methods -- Evaluating Hypothesis-Driven Exception-Rule Discovery with Medical Data Sets -- Discovering Protein Functional Models Using Inductive Logic Programming -- Mining Web Transaction Patterns in an Electronic Commerce Environment -- Association Rules and Related Topics -- Making Use of the Most Expressive Jumping Emerging Patterns for Classification -- Mining Structured Association Patterns from Databases -- Association Rules -- Density-Based Mining of Quantitative Association Rules -- AViz: A Visualization System for Discovering Numeric Association Rules -- Discovering Unordered and Ordered Phrase Association Patterns for Text Mining -- Using Random Walks for Mining Web Document Associations -- Induction -- A Concurrent Approach to the Key-Preserving Attribute-Oriented Induction Method -- Scaling Up a Boosting-Based Learner via Adaptive Sampling -- Adaptive Boosting for Spatial Functions with Unstable Driving Attributes -- Robust Ensemble Learning for Data Mining -- Interactive Visualization in Mining Large Decision Trees -- VQTree: Vector Quantization for Decision Tree Induction -- Making Knowledge Extraction and Reasoning Closer -- Discovery of Relevant Weights by Minimizing Cross-Validation Error -- Efficient and Comprehensible Local Regression -- Information Granules for Spatial Reasoning -- Text, Web, and Graph Mining -- Uncovering the Hierarchical Structure of Text Archives by Using an Unsupervised Neural Network with Adaptive Architecture -- Mining Access Patterns Efficiently from Web Logs -- A Comparative Study of Classification Based Personal E-mail Filtering -- Extension of Graph-Based Induction for General Graph Structured Data -- Text-Source Discovery and GlOSS Update in a Dynamic Web -- Extraction of Fuzzy Clusters from Weighted Graphs -- Text Summarization by Sentence Segment Extraction Using Machine Learning Algorithms.

The Fourth Pacific-Asia Conference on Knowledge Discovery and Data Mining (PAKDD 2000) was held at the Keihanna-Plaza, Kyoto, Japan, April 18 - 20, 2000. PAKDD 2000 provided an international forum for researchers and applica­ tion developers to share their original research results and practical development experiences. A wide range of current KDD topics were covered including ma­ chine learning, databases, statistics, knowledge acquisition, data visualization, knowledge-based systems, soft computing, and high performance computing. It followed the success of PAKDD 97 in Singapore, PAKDD 98 in Austraha, and PAKDD 99 in China by bringing together participants from universities, indus­ try, and government from all over the world to exchange problems and challenges and to disseminate the recently developed KDD techniques. This PAKDD 2000 proceedings volume addresses both current issues and novel approaches in regards to theory, methodology, and real world application. The technical sessions were organized according to subtopics such as Data Mining Theory, Feature Selection and Transformation, Clustering, Application of Data Mining, Association Rules, Induction, Text Mining, Web and Graph Mining. Of the 116 worldwide submissions, 33 regular papers and 16 short papers were accepted for presentation at the conference and included in this volume. Each submission was critically reviewed by two to four program committee members based on their relevance, originality, quality, and clarity.