Advances in Knowledge Discovery And Data Mining: 9th Pacific-Asia Conference, PAKDD 2005, Hanoi, Vietnam, May 18-20, 2005 Proceedings

The Pacific-Asia Conference on Knowledge Discovery and Data Mining (PAKDD) is a leading international conference in the area of data mining and knowledge discovery. It provides an international forum for researchers and industry practitioners to share their new ideas, original research results and practical development experiences from all KDD-related areas including data mining, data warehousing, machine learning, databases, statistics, knowledge acquisition and automatic scientific discovery, data visualization, causality induction, and knowledge-based systems. This year’s conference (PAKDD 2005) was the ninth of the PAKDD series, and carried the tradition in providing high-quality technical programs to facilitate research in knowledge discovery and data mining. It was held in Hanoi, Vietnam at the Melia Hotel, 18–20 May 2005. We are pleased to provide some statistics about PAKDD 2005. This year we received 327 submissions (a 37% increase over PAKDD 2004), which is the highest number of submissions since the first PAKDD in 1997) from 28 countries/regions: Australia (33), Austria (1), Belgium (2), Canada (11), China (91), Switzerland (2), France (9), Finland (1), Germany (5), Hong Kong (11), Indonesia (1), India (2), Italy (2), Japan (21), Korea (51), Malaysia (1), Macau (1), New Zealand (3), Poland (4), Pakistan (1), Portugal (3), Singapore (12), Taiwan (19), Thailand (7), Tunisia (2), UK (5), USA (31), and Vietnam (9). The submitted papers went through a rigorous reviewing process. Each submission was reviewed by at least two reviewers, and most of them by three or four reviewers.

Keynote Speech and Invited Talks.- Machine Learning for Analyzing Human Brain Function.- Subgroup Discovery Techniques and Applications.- IT Development in the 21st Century and Its Implications.- Theoretic Foundations.- Data Mining of Gene Expression Microarray via Weighted Prefix Trees.- Automatic Extraction of Low Frequency Bilingual Word Pairs from Parallel Corpora with Various Languages.- A Kernel Function Method in Clustering.- Performance Measurements for Privacy Preserving Data Mining.- Extraction of Frequent Few-Overlapped Monotone DNF Formulas with Depth-First Pruning.- Association Rules.- Rule Extraction from Trained Support Vector Machines.- Pruning Derivative Partial Rules During Impact Rule Discovery.- : A New Informative Generic Base of Association Rules.- A Divide and Conquer Approach for Deriving Partially Ordered Sub-structures.- Finding Sporadic Rules Using Apriori-Inverse.- Automatic View Selection: An Application to Image Mining.- Pushing Tougher Constraints in Frequent Pattern Mining.- An Efficient Compression Technique for Frequent Itemset Generation in Association Rule Mining.- Mining Time-Profiled Associations: An Extended Abstract.- Online Algorithms for Mining Inter-stream Associations from Large Sensor Networks.- Mining Frequent Ordered Patterns.- Biomedical Domains.- Conditional Random Fields for Transmembrane Helix Prediction.- A DNA Index Structure Using Frequency and Position Information of Genetic Alphabet.- An Automatic Unsupervised Querying Algorithm for Efficient Information Extraction in Biomedical Domain.- Voting Fuzzy k-NN to Predict Protein Subcellular Localization from Normalized Amino Acid Pair Compositions.- Comparison of Tree Based Methods on Mammography Data.- Bayesian Sequence Learning for Predicting Protein Cleavage Points.- A Novel Indexing Method for Efficient Sequence Matching in Large DNA Database Environment.- Classification and Ranking.- Threshold Tuning for Improved Classification Association Rule Mining.- Using Rough Set in Feature Selection and Reduction in Face Recognition Problem.- Analysis of Company Growth Data Using Genetic Algorithms on Binary Trees.- Considering Re-occurring Features in Associative Classifiers.- A New Evolutionary Neural Network Classifier.- A Privacy-Preserving Classification Mining Algorithm.- Combining Classifiers with Multi-representation of Context in Word Sense Disambiguation.- Automatic Occupation Coding with Combination of Machine Learning and Hand-Crafted Rules.- Retrieval Based on Language Model with Relative Entropy and Feedback.- Text Classification for DAG-Structured Categories.- Sentiment Classification Using Word Sub-sequences and Dependency Sub-trees.- Improving Rough Classifiers Using Concept Ontology.- QED: An Efficient Framework for Temporal Region Query Processing.- Clustering.- A MPAA-Based Iterative Clustering Algorithm Augmented by Nearest Neighbors Search for Time-Series Data Streams.- Locating Motifs in Time-Series Data.- Stochastic Local Clustering for Massive Graphs.- A Neighborhood-Based Clustering Algorithm.- Improved Self-splitting Competitive Learning Algorithm.- Speeding-Up Hierarchical Agglomerative Clustering in Presence of Expensive Metrics.- Dynamic Cluster Formation Using Level Set Methods.- A Vector Field Visualization Technique for Self-organizing Maps.- Visualization of Cluster Changes by Comparing Self-organizing Maps.- An Incremental Data Stream Clustering Algorithm Based on Dense Units Detection.- Visual Interactive Evolutionary Algorithm for High Dimensional Data Clustering and Outlier Detection.- Approximated Clustering of Distributed High-Dimensional Data.- Dynamic Data Mining.- Improvements of IncSpan: Incremental Mining of Sequential Patterns in Large Database.- Efficient Sampling: Application to Image Data.- Cluster-Based Rough Set Construction.- Graphic Model Discovery.- Learning Bayesian Networks Structures from Incomplete Data: An Efficient Approach Based on Extended Evolutionary Programming.- Dynamic Fuzzy Clustering for Recommender Systems.- Improving Mining Quality by Exploiting Data Dependency.- High Dimensional Data.- Feature Selection for High Dimensional Face Image Using Self-organizing Maps.- Progressive Sampling for Association Rules Based on Sampling Error Estimation.- CLe Ver: A Feature Subset Selection Technique for Multivariate Time Series.- Covariance and PCA for Categorical Variables.- Integration of Data Warehousing.- ADenTS: An Adaptive Density-Based Tree Structure for Approximating Aggregate Queries over Real Attributes.- Frequent Itemset Mining with Parallel RDBMS.- Knowledge Management.- Using Consensus Susceptibility and Consistency Measures for Inconsistent Knowledge Management.- WLPMiner: Weighted Frequent Pattern Mining with Length-Decreasing Support Constraints.- Machine Learning Methods.- A Framework for Incorporating Class Priors into Discriminative Classification.- Increasing Classification Accuracy by Combining Adaptive Sampling and Convex Pseudo-Data.- Kernels over Relational Algebra Structures.- Adaptive Nonlinear Auto-Associative Modeling Through Manifold Learning.- Maximizing Tree Diversity by Building Complete-Random Decision Trees.- SETRED: Self-training with Editing.- Adjusting Mixture Weights of Gaussian Mixture Model via Regularized Probabilistic Latent Semantic Analysis.- Training Support Vector Machines Using Greedy Stagewise Algorithm.- Cl-GBI: A Novel Approach for Extracting Typical Patterns from Graph-Structured Data.- Improved Bayesian Spam Filtering Based on Co-weighted Multi-area Information.- Novel Algorithms.- An Efficient Framework for Mining Flexible Constraints.- Support Oriented Discovery of Generalized Disjunction-Free Representation of Frequent Patterns with Negation.- Feature Selection Algorithm for Data with Both Nominal and Continuous Features.- A Two-Phase Algorithm for Fast Discovery of High Utility Itemsets.- On Multiple Query Optimization in Data Mining.- USAID: Unifying Signature-Based and Anomaly-Based Intrusion Detection.- Spatial Data.- Mining Mobile Group Patterns: A Trajectory-Based Approach.- Can We Apply Projection Based Frequent Pattern Mining Paradigm to Spatial Co-location Mining?.- PatZip: Pattern-Preserved Spatial Data Compression.- Temporal Data.- A Likelihood Ratio Distance Measure for the Similarity Between the Fourier Transform of Time Series.- The TIMERS II Algorithm for the Discovery of Causality.- A Recent-Biased Dimension Reduction Technique for Time Series Data.- Graph Partition Model for Robust Temporal Data Segmentation.- Accurate Symbolization of Time Series.- A Novel Bit Level Time Series Representation with Implication of Similarity Search and Clustering.- Finding Temporal Features of Event-Oriented Patterns.- An Anomaly Detection Method for Spacecraft Using Relevance Vector Learning.- Cyclic Pattern Kernels Revisited.- Text and Web Data Mining.- Subspace Clustering of Text Documents with Feature Weighting K-Means Algorithm.- Using Term Clustering and Supervised Term Affinity Construction to Boost Text Classification.- Technology Trends Analysis from the Internet Resources.- Dynamic Mining Hierarchical Topic from Web News Stream Data Using Divisive-Agglomerative Clustering Method.- Collecting Topic-Related Web Pages for Link Structure Analysis by Using a Potential Hub and Authority First Approach.- A Top Down Algorithm for Mining Web Access Patterns from Web Logs.- Kernel Principal Component Analysis for Content Based Image Retrieval.- Mining Frequent Trees with Node-Inclusion Constraints.