Database and Expert Systems Applications: 10th International Conference, DEXA'99, Florence, Italy, August 30 - September 3, 1999, Proceedings (Lecture Notes in Computer Science, 1677)

The Database and Expert Systems Applications (DEXA) conferences bring together researchers and practitioners from all over the world to exchange ideas, experiences and opinions in a friendly and stimulating environment. The papers are at once a record of what has been achieved and the first steps towards shaping the future of information systems. DEXA covers a broad field, and all aspects of database, knowledge base and related technologies and their applications are represented. Once again there were a good number of submissions: 241 papers were submitted and of these the programme committee selected 103 to be presented. DEXA’99 took place in Florence and was the tenth conference in the series, following events in Vienna, Berlin, Valencia, Prague, Athens, London, Zurich, Toulouse and Vienna. The decade has seen many developments in the areas covered by DEXA, developments in which DEXA has played its part. I would like to express thanks to all the institutions which have actively supported and made possible this conference, namely: • University of Florence, Italy • IDG CNR, Italy • FAW – University of Linz, Austria • Austrian Computer Society • DEXA Association In addition, we must thank all the people who have contributed their time and effort to make the conference possible. Special thanks go to Maria Schweikert (Technical University of Vienna), M. Neubauer and G. Wagner (FAW, University of Linz). We must also thank all the members of the programme committee, whose careful reviews are important to the quality of the conference. Database and Expert Systems Applications Preface Program Committee Table of Contents On Tractable Queries and Constraints 1 Introduction 2 Conjunctive Queries 3 Constraints 4 Acyclic Queries and Constraints 5 Decomposition Methods 6 Hypertree Width 7 Comparing the Different Methods 8 An Algorithm for Computing Optimal Hypertree Decompositions References Instances Evolution vs Classes Evolution 1. Introduction 1.1. Object Evolution Vs Artificial Evolution 1.2. Artificial Evolution 1.3. Object Evolution and Artificial Evolution 2. A Genetic Evolution Object Model 2.1. Concepts 2.2. Evolutionary processes 2.3. Operation of the model 2.4. Integration of the emergent entity and notion of semantic distance 3. Conclusion Bibliography Dynamic Relationships in Object Oriented Databases: A Uniform Approach Introduction Relationships: Objects and Meta-objects The Relationship Model Desirable Relationship Features The Relationship Model The Meta-Model Application to Evolution Related Work Summary and Conclusions References IFOOD: An Intelligent Object-Oriented Database Architecture 1 Introduction 2 The Architecture of IFOOD 3 IFOOD Language 3.1 Uncertain Types 3.2 Classes 3.3 Virtual (derived) Classes 3.4 Additional Definitions for Handling Uncertainty 3.5 Object Queries 4 Coupling Properties 4.1 Interaction between the FOODB and the FKB 4.2 Class-Rule Bindings 5 Conclusions References An OODBMS-IRS Integration Based on a Statistical Corpus Extraction Method for Document Management Introduction Fundamental Concepts of Signature File Techniques Background System Features Related Work System Architecture Signature File Scheme and Query Process A Corpus Based Signature File Approach for Document Retrieval Improper Signature Sharing Caused by Hash-function-based Approaches Grouping of Dissimilar Indexed Terms Experimental Results Conclusions References Controlled Hypertextual Navigation in the SgmlQL Language 1 Introduction 2 The SgmlQL language 3 Intuitive approach 4 The navigate operator 4.1 Hypertext model 4.2 Path filters 4.3 The "navigate" operator 4.4 Using index server 5 Guards 6 Dealing with cycles 7 Related works 8 Conclusion References SECONDO/QP: Implementation of a Generic Query Processor 1 Introduction 2 Second-Order Signature 3 Interaction of Query Processor and Algebra Modules 3.1 Overview 3.2 Implementation of Evaluation Functions 3.3 Implementation of Type Mapping Functions 4 Evaluation of Query Plans 4.1 Structure of the Operator Tree 4.2 Constructing the Operator Tree 4.3 The Evaluation Procedure 5 Related Work 6 Conclusions References A Definition of the Operator Tree B Structure of Procedure annotate C Structure of Procedure subtree D Procedure eval Hybrid Simultaneous Scheduling and Mapping in SQL Multi-query Parallelization 1 Introduction 2 Overview of the PSA/MEG Static Scheduling and Mapping Method 2.1 Models of the application and of the target parallel architecture 2.2 Formalization of the static mapping problem 2.3 Features of the PSA/MEG static method 3 Description of our dynamic parallelization strategy proposal 3.1 Formalization of the dynamic mapping problem 3.2 The PSA/MEG-based parallelization strategy in a dynamic context 4 Presentation of the incremental memory allocation heuristics 4.1 The ModDeg modification heuristic of the intra-operation parallelism degree 4.2 Performance evaluation of the ModDeg heuristic 5 Conclusion and Future Work References Cluster-Based Database Selection Techniques for Routing Bibliographic Queries 1 Introduction 1.1 Objective and Scope 2 Related Work 2.1 Database Selection for Text Collections 2.2 Database Selection for Collections with Multiple Attributes 2.3 Database Selection for Collections Accessible Through Query Interface Only 3 Overview of Cluster-based Database Selection 4 Clustering Techniques for Bibliographic Databases 4.1 Similarity Between a Bibliographic Record and a Cluster 4.2 Proposed Database Clustering Techniques 5 Cluster-Based Database Ranking Formulas 5.1 Cluster-based Database Ranking based on Estimated Result Sizes (ERS) 5.2 Cluster-based Database Ranking based on Estimated Goodness Score (EGS) 6 Experiments 7 Conclusions References Developing Patterns as a Mechanism for Assisting the Management of Knowledge in the Context of Conducting Organisational Change 1. Introduction 2. Background to Patterns 3. Encapsulating Organisational Knowledge in Patterns 4. Patterns in Change Management 4.1 The process and the Product of Change 4.2 Using Patterns to Represent the Elements of Change 5. Developing Organisational Patterns for Change Management 5.1 The Way-of-working 5.2 An Example From the Case Study 6. Conclusion References Knowledge Acquisition for Mobile Robot Environment Mapping 1 Introduction 2 Sensor Based Maps 3 Feature Extraction Using Occupancy Grid 4 Obstacle Detection During Tracking Process 5 Implementation and Experiments 6 Summary and Conclusion References Text Understanding for Knowledge Base Generation in the SynDiKATe System Introduction System Architecture Sentence Level Understanding Text Level Understanding SynDiKATe Applications Evaluation Related Work Conclusions References Knowledge Discovery with the Associative Memory Modell Neunet 1. Introduction 2. The Model Neunet-3 3. Improving Neunet-3 for calculation frequency, support and confidence 4. Examples 5. Conclusion References Tracking Mobile Users Utilizing Their Frequently Visited Locations 1 Introduction 2 The Frequently Visited Locations First (FVLF) Scheme 3 The Cost Model 4 Experiments and Results 4.1 The effect of Ch/Cv 4.2 The effect of n 5 Enhancement - Considering CMR's Effect 6 Conclusions and Future work References VPSF: A Parallel Signature File Technique Using Vertical Partitioning and Extendable Hashing 1 Introduction 2 Conventional Parallel Signature Files 3 Vertically-partitioned Parallel Signature File (VPSF) 3.1 Structure of the VPSF 3.2 Insertion and Deletion of the VPSF 3.3 Retrieval of the VPSF 4 Performance Evalulation 4.1 Retrieval Performance in terms of Data Distribution 4.2 Insertion Time and Storage Overhead 5 Conclusion References A Case for Deltas in Business-to-Business Electronic Commerce 1 Introduction 2 Data transparency and replicated databases 2.1 Data transparency 2.2 Conflict detection between proposed updates 3 Future directions 4 Conclusion References Flexible Workflow Management Systems: An Approach Based on Generic Process Models 1 Introduction 2 Generic process models 2.1 Concepts 2.2 Routing diagrams 2.3 Inheritance diagrams 3 Open problems References Object-Based Ordered Delivery of Messages in Object-Based Systems Introduction Object-Based Ordered Delivery Object-based systems Significant precedence Object-based ordered delivery Object-Based Group Protocol Object vector Message transmission and receipt Message ordering Message delivery Evaluation Concluding Remarks References Storage and Retrieval of XML Documents Using Object-Relational Databases 1 Introduction 1.1 Related Work 2 Storage Method of XML Documents 2.1 A Tree Structure Representing an XML Document 2.2 Design Strategies for Storing XML Documents 2.3 Addition of Abstract Data Type 2.4 Relational Database Schemas for Storing XML Documents 2.5 Storage of XML Documents to Relational Tables 3 Retrieval Method of XML Documents 4 Implementation 4.1 The Result of Experiments 5 Conclusions References SQL/LPP: a Time Series Extension of SQL based on Limited Patience Patterns Introduction Motivation Previous Work Limited Patience Patterns Extensional Data Model and Data Definition Language SQL/LPP The Language Pattern Functions Pattern Expressions Pattern Formulae and Pattern Sentence Searching Directives Defining SQL LPP Patterns Applying Patterns in Queries Conclusion References A Knowledge Based Approach for Modeling and Querying Multidimensional Databases 1 Introduction 2 Example 3 Preliminaries 4 The Languages 4.1 The Schema Language (SL) 4.2 The Query Language (QL) 5 Conclusion References An Incremental Hypercube Approach for Finding Best Matches for Vague Queries 1 Introduction 2 Related Research 3 Querying with the VQS 3.1 Representing Semantic Information 3.2 Mapping Semantics to Attributes 3.3 Processing the Similarity 3.4 Drawbacks 4 Multi Space Nearest Neighbor Search 4.1 Incremental Hypercube Search 4.2 Applying the Pyramid Technique for Handling the Hypercubes 5 Conclusion References Formalising Ontologies and Their Relations 1 Introduction 2 Ontological Specifications 3 Relating Ontologies 3.1 Combining and Sharing Ontologies 3.2 Compatibility of Ontologies 4 Conclusions References Is This Argument Justified, Overruled or Defensible? 1 Introduction 2 Rules 3 Arguments and defeasibility chains 4 Conclusions Addressing Efficiency Issues During the Process of Integrity Maintenance 1. Introduction 2. Deductive Databases 3. The Augmented Database [Oli91, UO92] 4. Making Explicit an Order for Integrity Maintenance 5. Translating a Repair Corresponding to a View Update 5.1 Analysis Step 5.2 Translation Step 6. A Process for Integrity Maintenance 7. Relation with Previous Work 7.1 Previous Work on Efficient Integrity Maintenance 7.2 Previous Work on View Updating 8 Conclusions. References Quality and Recommendation of Multi-source Data for Assisting Technological Intelligence Applications 1 Introduction 2 Data Quality : the multi-source perspective 3 Methodology for multi-source data quality expertise 4 Multicriteria recommendation based on relative data quality 4.1 Multi-source objects and quality metadata 4.2 Data quality scoring and ranking for recommendation 4.3 Multicriteria recommendation algorithm 5 Application of sQuaL in Technological Intelligence 6 Conclusion and perspectives References On Benchmarking Attribute Cardinality Maps for Database Systems Using the TPC-D Speci cations 1 Introduction 1.1 Query Optimization in Database Systems 1.2 Comparison of the ACMs to the Current State-of-the-Art 2 TPC-D Benchmark Specification 2.1 TPC-D Benchmark Queries 2.2 TPC-D Benchmark Database 3 Experiments with the TPC-D Benchmark Database 4 Analysis of the Experimental Results 5 Conclusion References Using Self-Organizing Maps to Organize Document Archives and to Characterize Subject Matter: How to Make a Map Tell the News of the World 1 Introduction 2 Self-Organizing Maps 3 Experimental Document Collection 4 A Map of Time Magazine Articles 5 Interpretation of Self-Organizing Maps 6 The LabelSOM Method 7 An Index of the Time Magazine Articles 8 Conclusions References Combining C-signature with Path Dictionary for Query Processing of Nested Objects in OODBS 1 Introduction 2 Definitions and Concepts 3 C-signatures 3.1 C-signatures on s-expression 3.2 Query Processing with C-signature 4 Performance Analysis 4.1 Experiment design 4.2 Experiment results 5 Conclusion A CORBA Object-based Caching with Consistency 1 Motivation 2 An Overview of the COCC Model 3 Cache Design 3.1 Logical Architecture 3.2 Example IDL Structure 3.3 Scenarios and Design Tradeoffs 4 Implementation and Performance 5 Conclusion References From Object-Oriented Conceptual Modeling to Component-Based Development 1 Introduction 2 Towards CBD 3 Solution Space 3.1 An abstract Execution Model 3.2 A tiered component architecture for the execution model 3.3 Generating OO code from the formal specification. 3.4 From problem domain classes to business components 4 Conclusions References Query Processing in Relationlog 1 Introduction 2 Relationlog Overview 3 System Architecture 4 Extended Semi-Naive Technique 5 Extended Magic-Set Strategy 6 Conclusion References Rewriting Queries Using Views 1 Introduction 2 Preliminaries 2.1 Program Containment 2.2 Classical negation 3 Rewriting queries 4 Rewriting queries with negation 5 Deriving information by merging rules 6 Functional Dependencies 7 Conclusions References A Query Subsumption Technique 1 Introduction 1.1 Related work 1.2 Language facility 2 The Canonical Form 2.1 Predicate Classification 2.2 Predicate Transformation 2.3 Query transformation 3 Subsumption (Containment) between Predicates 3.1 Contain1(alpha1, alpha2) 3.2 Check the Containment Between Two pi's. Contain2(pii, pij) 3.3 Check the Containment Between Two omega ́s: Contain3(omega i, omega j) 4 Check Subsumption Between Two Queries 4.1 Subsumption algorithm 4.2 Building the filter query 4.3 Examples 5 Conclusion and Further Work. References A Conference Key Multicasting Scheme Using Knapsack and Secret Sharing Introduction Related Works Our Protocols Cryptanalysis Conclusions References Verify Updating Trigger Correctness 1 Introduction 2 Basic Setting 2.1 The database and trigger rules 2.2 Rule selection method 2.3 Trigger execution order 2.4 Constraint 3 State Transform Predicate 3.1 Verification of non-indefinitely-recursive trigger rules 3.2 Verification of purely indefinitely-recursive trigger rules 3.3 Verification of general trigger rules 3.4 Correctness of the entire system 4 Conclusion References A FlexibleWeighting Scheme for Multimedia Documents 1 Introduction 2 Images versus Text: New Weighting Parameters 3 An Outline of Weighting Rich Formalisms 4 Weighting a Conceptual Graph 5 Related Work and Discussion 6 Implementation of the Approach 7 Application to an Image Test Collection 8 Conclusion References A Fast Method for Ensuring the Consistency of Integrity Constraints 1. Introduction 2. Constraint Definition 3. Constraint Maintenance 4. An Alternative Approach to Checking Rule Consistency 5 Conclusions Bibliography A Mechanism for Deriving Specifications of Security Functions in the CC Framework 1 Introduction 2 Specification of Information Security Requirements 3 Overview and Concepts of Harmonization 4 Implementation of Harmonization 4.1 Expression of Information Security Requirements 4.2 Specification and Enforcement of Harmonization Functions 4.3 Merging of Requirements 5 Conclusions and Future Work References Efficient Retrieval of Structured Documents from Object-Relational Databases 1. Introduction 2. TOODOR Document Model 2.1. Some Implementation Issues 3. Document Retrieval Language 3.1 Path Expressions 3.2 Structural Conditions 4. Representing Documents with the Object-Relational Data Model 4.1 Document Representation 4.2 Codifying Document Structures 5. Query Processing in TOODOR 5.1 Evaluation of Path Expressions 5.2 Evaluation of Structural Conditions 5.3 Combining Structural and Contents Conditions 6. Conclusions References Analysis of Active Database Rules Behaviour Using Rewriting Logic 1 Introduction 2 Influence of the Dimensions of Active Rule Execution Model on the Termination Property 3 Specification of an Active Object-Oriented System with the Rewriting Logic 4 The Three Basic Strategies 5 Taking into Account the Active Rule Dimensions 6 Stratification Algorithms 7 Elaboration of the Local Strategy for a Stratum 8 Overall Process for Analysing the Termination and Confluence Properties 9 Application Example 10 Conclusion Bibliography Parallel Object Server Architecture and Performance 1 Introduction 2 Related work 3 Data structures and parallel processing 4 Hardware and software architecture 5 Implemented prototype OPAS 6 Results, conclusions, future work References Using Contextual Fuzzy Views to Query Imprecise Data 1 Introduction 2 Preliminary 3 Contextual fuzzy view processing 3.1 The user point of view 3.2 Knowledge base structure 3.3 Main steps of the processing 3.4 Similarity measure between fuzzy values 3.5 Software architecture 4 An example of application 4.1 Distance relations 4.2 Categories 4.3 Example of querying 4.4 Result of the querying 5 Conclusion Acknowledgments References Combining Pat-Trees and Signature Files for Query Evaluation in Document Databases 1 Introduction 2 Related work 3 Queries in document databases 4 Representative word hierarchies 5 Pat trees and signature files 5.1 Pat-tree built over representative words 5.2 Signature File Hierarchy 5.3 Combining pat trees and signature files 6 Retrieval 7 Conclusion References Answering Queries by Semantic Caches 1 Introduction 2 Semantic Query Caching 2.1 Overview 2.2 Applications 3 Evaluating Queries by Caches 3.1 Logical Notation 3.2 Determining when Answers are in Cache 3.3 Finding the Answers in Cache 3.4 Semantic Overlap and Semantic Independence 3.5 Semantic Remainder 4 Related Work 5 Conclusions References Simulating the Interaction of Database Agents 1. Introduction 2. Database Narratives 3. Modelling the behaviour of agents 3.1. Goal Relationships 3.2. Interactive Planning 3.3. Progressive Goal Generation 4. The Interactive Plot Generator 4.1. A short description of IPG 4.2. Running Example 5. Concluding Remarks References Automatic and Semantic Techniques for Scheme Integration and Scheme Abstraction Introduction Preliminaries Interscheme Property Derivation The set of mappings and the set of views Structures and contexts of objects Metascheme Integration Algorithm Scheme abstraction References Dialogue Management In a Virtual College 1 Introduction 2 The `Virtual College' Scenario 3 System Architecture 4 Some Specific Instances of Conversation Classes 4.1 Conditions Derived from the Conversation Classes 4.2 Conditions Derived From Policies and Heuristics 5 Discussion 6 Conclusions References Meta Data Based EDI for Small Enterprises 1 Introduction 2. Problems with EDI in Small-scale Enterprises 3. The Meta Data Based Data Interchange Processor 4. Conclusions References Database Challenges for Genome Information in the Post Sequencing Phase 1 Introduction 2 Modelling Genomic Data 3 Managing Expression Data 4 The Genome Information Management System 4.1 Architecture 4.2 Queries and Analyses 5 Conclusions References Supporting Teams In Virtual Organizations 1 Introduction 2 Toolkit Requirements for Virtual Teams 2.1 Workspace requirements 3 A Prototype Toolkit 3.1 Toolkit terms 3.2 The Toolkit Interface to define the collaborative context 4 General Tools for Collaboration and Knowledge Sharing 4.1 Extending To Agents 5 Summary 6 References If We Refuse the Inheritance ... 1 Introduction 2 Hologram Approach: Implementing Specialization 3 Hologram Approach: Manipulating Objects 3.1 Primitive Methods 3.2 Attribute and Method Redefinition 4 Schema Evolution and Specialization 5 Related Work 6 Conclusion References Viewpoints Handling in an Object Model with Criterium-Based Classes 1 Introduction 2 The model 2.1 Classes and objects 2.2 Criteria 3 Use of CBCs 3.1 Viewpoints merging 3.2 Property sharing 3.3 Viewpoint specialization and inheritance 3.4 Access rights 3.5 Inheritance between classes 4 Related Work 4.1 Fibonacci 4.2 IQL(2) 5 Concluding remarks and future work References The Constructive Method for Query Containment Checking 1 Introduction 2 Base Concepts 3 The Constructive Method for QC Checking 3.1 Example 3.2 Variable Instantiation Patterns 4 Soundness and Completeness of the Constructive Query Containment Method 5 Decidability issues 6 Related Work 7 Conclusions and Further Work References On the Specification of Representation-Based Conditions in a Context of Incomplete Databases 1 Introduction 2 Representing Imperfect Values: a Brief Survey 2.1 Null Values 2.2 Disjunctive Sets 2.3 Weighted Disjunctive Sets 3 A New Type of Conditions: Representation-Based Conditions 3.1 Principles of Representation-Based Queryin 3.2 Motivations 4 A Language for Representation-Based Querying 4.1 Conditions Involving One Representation 4.2 Some Examples in the Context of a Database Describing Images 4.3 Conditions Involving Two Representations 5 Conclusion References Methods and Interpretation of Database Summarisation 1 Introduction 2 Data Summarisation Techniques - Discussion and Scope 2.1 Augmenting the Summary Relation 3 Concepts as Domain Values 3.1 Semantics 4 A Formal Categorisation of Summarisation Strategies 4.1 Information Capacity 4.2 Vertical Reduction by Attribute Projection 4.3 Horizontal Reduction by Tuple Selection 4.4 Concept Ascension and Ranges 4.5 Existential Domain Values 4.6 Source Cardinality, Statistical Summarisation and other Summarisation Techniques 5 Further Discussion and Future Research References An Efficient Scalable Parallel View Maintenance Algorithm for Shared Nothing Multi-processor Machines 1 Introduction 2 PDBMS, join operations and the BSP cost model 2.1 The BSP cost model 2.2 Load balancing in parallel join operations 3 Duplicating join algorithm 4 Incremental view maintenance : A new approach 5 Parallel incremental evaluation 5.1 Case of simple join 5.2 Case of SPJ views 6 Parallelizing materialized view updates 7 Conclusion References Using Back Propagation Algorithm and Genetic Algorithm to Train and Re ne Neural Networks for Object Detection 1 Introduction 1.1 Goals 2 Back Propagation Algorithm and Genetic Algorithm for Network Training 2.1 Back Propagation Algorithm for Network training 2.2 Genetic Algorithm for Network Training 3 Genetic Algorithm for Network Refinement 3.1 Characteristics of GA-refine 3.2 Fitness Function 4 The Image Databases 5 Results 5.1 Easy Pictures 5.2 Coin Pictures 5.3 Retina Pictures 6 Conclusions 7 Future Work References An Adaptive, Maintainable, Extensible Process Agent 1. Introduction 2. Active Component 2.1 Deliberative reasoning 3. Adaptivity 3.1 Blind adaptivity 3.2 Informed adaptivity 4. Maintenance Assistant 5. Extension Assistant 6. Conclusion 7. References Metadata, a ‘Semantic’ Approach 1 Introduction 2 The architecture of NKRL 2.1 The lower layer 2.2 The upper layer 3 A brief survey of some important NKRL features 3.1 Descriptive and factual components 3.2 Definitional and enumerative components 4 Implementation notes : NKRL and RDF 5 Conclusion References HEROSfw: a Framework for Heterogeneous Database Systems Integration 1 Introduction 2 An Overview of the HEROS Project 3 HEROS^fw Framework 4 Hot spots of HEROS^fw Framework 4.1 Data Model Hot Spot 4.2 Communication Hot Spot 4.3 Transaction Hot Spot 4.4 Final Considerations 5 Implementation of HEROS^fw Framework 6 Related Work 7 Final Comments References Resolving Ontological Heterogeneity in the KRAFT Project 1 Introduction 2 The KRAFT architecture in the Router-Configuration domain 3 Ontology Heterogeneity 4 The Construction of the Shared Ontology 5 Ontology Mappings 6 Conclusions References Generation of Conceptual Wrappers for Legacy Databases 1 Introduction 2 Architecture 2.1 Hierarchy architecture 2.2 Component architecture 3 Schema Transformations 3.1 Definition 3.2 Semantics-preserving transformations 3.3 Semantics-preserving transformation sequence 4 Methodology 4.1 Physical schema recovery 4.2. Conceptual schema recovery 4.3 Building the Mappings 4.4 Conceptual wrapper generation 5 Computer-aided design of wrappers 5.1 The DB-MAIN Case environment 5.2 Physical extraction support 5.3 Conceptual schema recovery 5.4 Mapping building 5.5 Conceptual Wrapper generation 6 Conclusions References Design and Implementation of Linear Hash Algorithm in a Nested Transaction Environment 1 Introduction 2 Linear Hash Structures in Nested Transaction Environment 3 Layered System Architecture 4 Object-Oriented Design 5 Nested Transactions as Multithreads 5.1 User Interface Operations 6 Conclusions References Supporting Partial Isolation in Flat Transactions 1 Introduction 2 Motivation 3 Related work 4 Partial isolation and isolation units 5 An adaptable correctness criterion 6 Two-dimensional locking 6.1 Isolation levels and lock components 6.2 Lock stages 6.3 Lock switching and unlocking 7 Recovery 8 Conclusions References Using a Similarity Measurement to Partition a Vocabulary of Medical Concepts 1 Introduction 2 Partitioning Approach 2.1 Similarity based on Property Introduction 2.2 Similarity based on Relationship Overriding 2.3 Partitioning the InterMED 3 Structural Partitioning vs. Expert Partitioning 4 Related Literature 5 Conclusions and Future Work References A Study on Musical Features for Melody Databases 1 Introduction 2 Features 3 Music collection 4 Feature "alphabets" and "words" - how many? 5 Does Zipf's law hold for musical features? 6 Are there any musical "stopwords"? 7 What should n be for n-gram indices? 8 Summary and future work References Using Multimedia in Aeronautical Technical Documentation Introduction 1 Study Background 1.1 Technical Documentation Framework 1.2 Multimedia Documents 2 Needs Analysis in the Environment of Technical Documentation Production 2.1 Extension of Existing Problems Data Evolution Process. Information reuse is built on documentary unit management rules and on models of final documents. It allows separating the respective changes of the document structure (i.e. models) and of the document content (i.e. data). Cha 2.2 New Issues On-line Access and Interaction Management. Multimedia and hypermedia documents give the possibility to navigate within the different documentary units of which they are comprised of. This capability is inherited from hypertext system and allows accessing Control of Document Quality. Multimedia state-of-the-art on multimedia and presentation systems emphasises the following needs: 2.3 Summary 3 Multimedia and Hypermedia: Analysis and Conclusions 3.1 Multimedia Synchronisation 3.2 Hypermedia 3.3 Database Approaches 3.4 Analysis Result 4 Definition of a Multimedia and Hypermedia Repository 4.1 Solving the Basic Issue: Multimedia Document Synchronisation Modelling 4.2 Integration of Structuring Capability 4.3 Synchronisation Unit Conclusion References Data Warehouse Design and Maintenance through View Normalization 1 Introduction 2 Data Warehouse Design by View Normalization 3 Data Warehouse Maintenance by View Normalization 4 Challenges and Open Issues References Cleansing Data for Mining and Warehousing 1 Introduction 2 Motivation 3 Proposed Cleansing Methodology 3.1 Scrubbing Dirty Data Fields 3.2 Tokenizing and Sorting Data Fields 3.3 Comparing Records 4 Data Cleansing System - Implementation and Performance 5 Conclusion References DROLAP A Dense-Region Based Approach to On-Line Analytical Processing 1 Introduction 2 Problem Definition and Related Works 2.1 Problem Statement 2.2 Related Works 3 The EDEM Algorithm for Mining Dense Regions 3.1 Building a k-d Tree to Store the Valid Cells 3.2 Growing Dense Region Covers along Boundaries 3.3 Searching Dense Regions among the Covers 3.4 Complexity of EDEM 3.5 Environments with Limited Memory 4 Performance Studies 4.1 System Implementation and Data Generation 4.2 Performance Results of DROLAP 4.3 Performance of EDEM 5 Discussion and Conclusion References Availability and Reliability Issues in Distributed Databases Using Optimal Horizontal Fragmentation I-Introduction II-Previous Related Work II-a. Fragmentation II-b. Replication III- A Proposed Horizontal Fragmentation Algorithm IV- A Proposed Replication Algorithm V-Conclusion and Future Work References Object Clustering Methods and a Query Decomposition Strategy for Distributed Object-Based Information Systems 1 Introduction 2 Fragtique: An OO distribution design methodology 2.1 An overview of the steps of the Fragtique OO distribution design 2.2 OO distribution design: modeling methods and queries 3 Query decomposition based on tree-fragments 4 Allocation 4.1 Modeling method activity 4.2 Quantifying method activity 4.3 Virtual-blocks 5 Conclusion References Dynamic Adjustment of Localized Constraints 1 Introduction 2 Related Work 3 Dynamic Localization 3.1 A Geometric Method for Localization 3.2 Initial Algorithm for Incremental Update 4 Implementing the Algorithm 4.1 Handling Simultaneity 4.2 Categorizing User Transactions 5 The Experiment 6 Conclusion References Solent - a Platform for Distributed Open Hypermedia Applications 1 Requirements 2 Description of the Architecture 2.1 Dynamic Service Discovery and Execution 2.2 A Storage Engine for XML 3 Sample Applications: Generic Media Player and Car Stereo 4 Summary and Conclusions References Concurrency Control for Global Transaction Management in MDBSs 1 Introduction 2 Research Backgrounds 2.1 Previous Works 2.2 Research Motivation 3 Site-Locking Protocol for Global Concurrency Control 3.1 Integrity Constraints and Transaction Model 3.2 Ensuring the Global Serializability 3.3 Lock Operation for Accessing the Site 3.4 Protocol of Site-Locking 4 Conclusion References Transactional Computation Introduction Two Modes of Computation Transactions Redefined Specification of Transactions Transactions and Processes Conclusions and Research Issues Database Versions to Represent Bitemporal Databases Introduction 1 Temporal and bitemporal databases 2 The Database Version approach 3 Bitemporal database using DBVs 3.1 Associating time dimensions to DBVs 3.2 Managing branching evolution 4 Conclusion References A Case Study in Information Delivery to Mass Retail Markets 1 Introduction 2 System overview 3 Software Architecture 4 Data management 4.1 Operational data 4.2 Analysis data 4.3 Personal device data 4.4 Main data processing steps 5 Initial experience and performance 6 Conclusions References Towards a Scalable System Architecture in Digital Libraries 1 Introduction 2 VD17: An example of a Digital Library Project 2.1 System Architecture 3 Performance Issues 3.1 User and log Studies 3.2 Discussion of Possible Solutions 4 A Suitable Solution 4.1 Description 4.2 Scalable System Architecture 5 Conclusion References MOSAIC: A Multi-Feature Access Method for Large Image Databases 1 Introduction 2 Multi-Feature Cluster-based Image Retrieval 2.1 Color Representation 2.2 Representation of Cluster Information 2.3 Extraction of Multi-feature Clusters 2.4 Similarity Function 2.5 Image Retrieval Process 3 MOSAIC: A Multi-feature Index Structure 3.1 The MOSAIC structure 3.2 Matching and Searching a MOSAIC structure 3.3 Insertion of Clusters 4 Performance Evaluation 5 Conclusion References Enhancing Data Warehousing with Fuzzy Technology 1 Introduction 2 Brief Review of Membership Functions and Fuzzy Sets 3 Integrating Fuzzy Technology Into Data Warehousing Sys

This book constitutes the refereed proceedings of the 10th International Conference on Database and Expert Systems Applications, DEXA'99, held in Florence, Italy in August/September 1999. The 105 revised papers presented together with three invited papers were carefully reviewed and selected from a total of 241 submissions. The volume addresses all current aspects of advanced information systems applications. The papers are organized in topical sections on object-oriented databases; fundamentals; queries; advanced information systems; workflow management systems; heterogeneous, distributed, and federated databases; transactions; data warehousing and data mining; spatial databases; internet applications; and temporal databases.