Principles of data mining and knowledge discovery : Second European Symposium, PKDD '98, Nantes, France, September 23-26, 1998 : proceedings

This Book Constitutes The Refereed Proceedings Of The Second European Symposium On Principles Of Data Mining And Knowledge Discovery, Pkdd '98, Held In Nantes, France, In September 1998. The Volume Presents 26 Revised Papers Corresponding To The Oral Presentations Given At The Conference; Also Included Are Refereed Papers Corresponding To The 30 Poster Presentations. These Papers Were Selected From A Total Of 73 Full Draft Submissions. The Papers Are Organized In Topical Sections On Rule Evaluation, Visualization, Association Rules And Text Mining, Kdd Process And Software, Tree Construction, Sequential And Spatial Data Mining, And Attribute Selection. On Objective Measures Of Rule Surprisingness -- Discovery Of Surprising Exception Rules Based On Intensity Of Implication -- A Metric For Selection Of The Most Promising Rules -- For Visualization-based Analysis Tools In Knowledge Discovery Process: A Multilayer Perceptron Versus Principal Components Analysis: A Comparative Study -- Trend Graphs: Visualizing The Evolution Of Concept Relationships In Large Document Collections -- Ranked Rules And Data Visualization -- Textvis: An Integrated Visual Environment For Text Mining -- Text Mining At The Term Level -- A New Algorithm For Faster Mining Of Generalized Association Rules -- Knowledge Discovery With Clustering Based On Rules. Interpreting Results -- Efficient Construction Of Comprehensible Hierarchical Clusterings -- Cost Sensitive Discretization Of Numeric Attributes -- Handling Kdd Process Changes By Incremental Replanning --^ Object Mining: A Practical Application Of Data Mining For The Construction And Maintenance Of Software Components -- A Relational Data Mining Tool Based On Genetic Programming -- Inducing Cost-sensitive Trees Via Instance Weighting -- Model Switching For Bayesian Classification Trees With Soft Splits -- Interactive Visualisation For Predictive Modelling With Decision Tree Induction -- Discovery Of Diagnostic Patterns From Protein Sequence Databases -- The Psp Approach For Mining Sequential Patterns -- Knowledge Discovery In Spatial Data By Means Of Ilp -- Querying Inductive Databases: A Case Study On The Mine Rule Operator -- Classes Of Four-fold Table Quantifiers -- Detection Of Interdependences In Attribute Selection -- Postponing The Evaluation Of Attributes With A High Number Of Boundary Points -- A Hybrid Approach To Feature Selection -- Discretization And Grouping: Preprocessing Steps For Data Mining -- Fuzzy Spatial Oql For Fuzzy Knowledge Discovery In Databases --^ Extended Functional Dependencies As A Basis For Linguistic Summaries -- A Comparison Of Batch And Incremental Supervised Learning Algorithms -- Knowledge Discovery With Qualitative Influences And Synergies -- Language Support For Temporal Data Mining -- Resampling In An Indefinite Database To Approximate Functional Dependencies -- Knowledge Discovery From Client-server Databases -- Discovery Of Common Subsequences In Cognitive Evoked Potentials -- Improving The Discovery Of Association Rules With Intensity Of Implication -- Generalization Lattices -- Overcoming Fragmentation In Decision Trees Through Attribute Value Grouping -- Data Mining At A Major Bank: Lessons From A Large Marketing Application -- Polyanalyst Data Analysis Technique And Its Specialization For Processing Data Organized As A Set Of Attribute Values -- Representative Association Rules And Minimum Condition Maximum Consequence Association Rules -- Discovery Of Decision Rules From Databases: An Evolutionary Approach --^ Using Loglinear Clustering For Subcategorization Identification -- Exploratory Attributes Search For Time-series Data: An Experimental System For Agricultural Application -- A Procedure To Compute Prototypes For Data Mining In Non-structured Domains -- From The Data Mine To The Knowledge Mill: Applying The Principles Of Lexical Analysis To The Data Mining And Knowledge Discovery Process -- Preprocessing Of Missing Values Using Robust Association Rules -- Similarity-driven Sampling For Data Mining -- Modeling The Business Process By Mining Multiple Databases -- Data Transformation And Rough Sets -- Conceptual Knowledge Discovery In Databases Using Formal Concept Analysis Methods -- Clasitex+: A Tool For Knowledge Discovery From Texts -- Discovery Of Approximate Medical Knowledge Based On Rough Set Model -- Scalable, High-performance Data Mining With Parallel Processing -- Practical Text Mining -- Industrial Applications Of Data Mining. Jan M. Żytkow, Mohamed Quafafou (eds.). Includes Bibliographical References And Index. On objective measures of rule surprisingness....Pages 1-9 Discovery of surprising exception rules based on intensity of implication....Pages 10-18 A metric for selection of the most promising rules....Pages 19-27 For visualization-based analysis tools in knowledge discovery process: A multilayer perceptron versus principal components analysis: A comparative study....Pages 28-37 Trend graphs: Visualizing the evolution of concept relationships in large document collections....Pages 38-46 Ranked rules and data visualization....Pages 47-55 TextVis: An integrated visual environment for text mining....Pages 56-64 Text mining at the term level....Pages 65-73 A new algorithm for faster mining of generalized association rules....Pages 74-82 Knowledge discovery with clustering based on rules. Interpreting results....Pages 83-92 Efficient construction of comprehensible hierarchical clusterings....Pages 93-101 Cost sensitive discretization of numeric attributes....Pages 102-110 Handling KDD process changes by incremental replanning....Pages 111-120 Object mining: A practical application of data mining for the construction and maintenance of software components....Pages 121-129 A relational data mining tool based on genetic programming....Pages 130-138 Inducing cost-sensitive trees via instance weighting....Pages 139-147 Model switching for bayesian classification trees with soft splits....Pages 148-157 Interactive visualisation for predictive modelling with decision tree induction....Pages 158-166 Discovery of diagnostic patterns from protein sequence databases....Pages 167-175 The PSP approach for mining sequential patterns....Pages 176-184 Knowledge discovery in spatial data by means of ILP....Pages 185-193 Querying inductive databases: A case study on the MINE RULE operator....Pages 194-202 Classes of four-fold table quantifiers....Pages 203-211 Detection of interdependences in attribute selection....Pages 212-220 Postponing the evaluation of attributes with a high number of boundary points....Pages 221-229 A hybrid approach to feature selection....Pages 230-238 Discretization and grouping: Preprocessing steps for data mining....Pages 239-245 Fuzzy spatial OQL for fuzzy knowledge discovery in databases....Pages 246-254 Extended functional dependencies as a basis for linguistic summaries....Pages 255-263 A comparison of batch and incremental supervised learning algorithms....Pages 264-272 Knowledge Discovery with qualitative influences and synergies....Pages 273-281 Language support for temporal data mining....Pages 282-290 Resampling in an indefinite database to approximate functional dependencies....Pages 291-299 Knowledge discovery from client-server databases....Pages 300-308 Discovery of common subsequences in cognitive evoked potentials....Pages 309-317 Improving the discovery of association rules with intensity of implication....Pages 318-327 Generalization lattices....Pages 328-336 Overcoming fragmentation in decision trees through attribute value grouping....Pages 337-344 Data mining at a major bank: Lessons from a large marketing application....Pages 345-351 PolyAnalyst data analysis technique and its specialization for processing data organized as a set of attribute values....Pages 352-360 Representative association rules and minimum condition maximum consequence association rules....Pages 361-369 Discovery of decision rules from databases: An evolutionary approach....Pages 370-378 Using loglinear clustering for subcategorization identification....Pages 379-387 Exploratory attributes search for time-series data: An experimental system for agricultural application....Pages 388-395 A procedure to compute prototypes for data mining in non-structured domains....Pages 396-404 From the data mine to the knowledge mill: Applying the principles of lexical analysis to the data mining and knowledge discovery process....Pages 405-413 Preprocessing of missing values using robust association rules....Pages 414-422 Similarity-driven sampling for data mining....Pages 423-431 Modeling the business process by mining multiple databases....Pages 432-440 Data transformation and rough sets....Pages 441-449 Conceptual Knowledge Discovery in Databases using formal concept analysis methods....Pages 450-458 Clasitex + : A tool for knowledge discovery from texts....Pages 459-467 Discovery of approximate medical knowledge based on rough set model....Pages 468-476 Scalable, high-performance data mining with parallel processing....Pages 477-477 Practical text mining....Pages 478-478 Industrial applications of data mining....Pages 479-480 A crucial aspect of data mining is that the discovered knowledge (usually expressed in the form of "if-then" rules) should be somehow interesting, where the term interestingness is arguably related to the properties of surprisingness (unexpectedness), usefulness and novelty of the rule [Fayyad et al. 96].