Linking Enterprise Data (eBook)

David Wood (Herausgeber)

eBook Download: PDF

2010 | 2010
XXVI, 291 Seiten
Springer US (Verlag)
978-1-4419-7665-9 (ISBN)

Infos

Enterprise data is growing at a much faster rate than traditional technologies allow. New enterprise architectures combining existing technologies are desperately needed. This book suggests a way forward by applying new techniques of the World Wide Web to enterprise information systems.

Linking Enterprise Data is an edited volume contributed by worldwide leaders in Semantic Web and Linked Data research, standards development and adoption. Linking enterprise data is the application of World Wide Web architecture principles to real-world information management issues faced by commercial, not-for-profit and government enterprises. This book is divided into four sections: Benefits of applying Linked Data principles in enterprise settings, enterprise approval and support of Linked Data projects, specific Linked Data techniques and a number of real-world success stories from early enterprise adopters.

Linking Enterprise Data targets professionals working as CTOs, CIOs, enterprise architects, project managers and application developers in commercial, not-for-profit and government organizations concerned with scalability, flexibility and robustness of information management systems. Computer science graduate students and researchers focusing on enterprise information integration will also benefit.

---

Enterprise data is growing at a much faster rate than traditional technologies allow. The World Wide Web is the only information system that scales to the degree that it does and is robust to both changes and failure of components. Most software does not work nearly as well as the Web does. Applying the Web's architectural principles (versus using the public Web directly) for enterprise problems maybe the only way to effectively address the current and future information glut. Linking Enterprise Data is an edited volume contributed by worldwide leaders in Semantic Web research, standards development and early adopters of Semantic Web standards and techniques. Linking enterprise data is the application of World Wide Web architecture principles to real-world information management issues faced by commercial, not-for-profit and government enterprises. The first section of this book will address business, technical and social values of applying Web architecture to enterprise content. An important focus of the first section will be the building of social (human-centered) communities to curate distributed data. The second section, Infrastructure, will address the need and options for persistent identifiers of data, terms and vocabularies for use by enterprise architects, and data storage and retrieval infrastructure appropriate for use. Case studies and examples that demonstrate the applicability of the proposed architectures are presented in the last section of Linking Enterprise Data.About this book:Provides practical approaches to addressing common information management issues by the application of leading edge research.Application approaches are based on international standards.Includes case studies and examples that demonstrate the applicability of the proposed architectures.This book targets professionals working as CTOs, CIOs, enterprise architects, project managers and application developers in commercial, not-for-profit and government organizations concerned with scalability, flexibility and robustness of information management systems and the Semantic Web. Advanced-level students and researchers focusing on computer science will also find this book valuable as a secondary text or reference book.

Inhalt

Preface 8
Contents 12
List of Contributors 20
Acronyms 24
Part I Why Link Enterprise Data? 28
Semantic Web and the Linked Data Enterprise 30
1 Social Data in the Enterprise 30
1.1 Causes 32
1.2 Technology Solutions 33
1.2.1 Data Warehousing 33
1.2.2 Master Data Management 34
1.2.3 Metadata Repositories 34
1.2.4 Controlled Vocabularies 35
1.2.5 Natural Language Processing 35
1.3 Localization and Globalization 36
2 The Linked Data Enterprise 37
2.1 Controlled Vocabularies 38
2.1.1 Tagging-style vocabularies 39
2.1.2 Schema-style vocabularies 41
2.1.3 Linked Enterprise Vocabularies 43
2.2 Prerequisites for Linked Data Vocabularies 45
3 Examples 47
3.1 Publishing 47
3.2 Government 48
4 Conclusions 49
References 49
The Role of Community-Driven Data Curation for Enterprises 52
1 Introduction 52
2 The Business Need for Curated Data 53
3 Data Curation 55
3.1 How to Curate Data 55
3.1.1 Setting up a Curation Process 56
4 Community-based Curated Enterprise Data 57
4.1 Internal Corporate Community 57
4.2 External Pre-competitive Communities 58
5 Case Study: Wikipedia - The World Largest Open Digital Curation Community 59
5.1 Social Organization 60
5.2 Artifacts, Tools and Processes 61
5.3 DBPedia - Community Curated Linked Open Data 62
6 Case Study: The New York Times - 100 Years of Expert Data Curation 63
6.1 Data Curation 63
6.2 Publishing Curated Linked Data 64
7 Case Study: Thomson Reuters - Data Curation, a Core Business Competency 65
7.1 Data Curation 66
8 Case Study: ChemSpider - Open Data Curation in the Global Chemistry Community 67
8.1 Community Objectives 68
8.2 Curation Approach & Types
9 Case Study: Protein Data Bank, Pre-competitive Bioinformatics 69
9.1 Serving the Community 69
9.2 Curation Approaches & Types
9.3 Observations 70
10 Case Study Learnings 71
10.1 Social Best Practices 71
10.2 Technical Best Practices 72
11 Conclusion 73
References 73
Part II Approval and Support of Linked Data Projects 76
Preparing for a Linked Data Enterprise 78
1 Introduction 79
2 The Cost of Linked Data 79
2.1 The Cost of Services and Support 80
2.2 Education and Training 80
2.3 Infrastructure 81
3 Is your Organization Ready for Linked Data? 81
4 The Linked Data Initiative 84
5 A Decentralized Approach to Data Management 85
6 Being On theWeb vs. In the Web 86
7 Leverage Vocabularies 87
8 A Simple Approach to Linked Data 88
9 Conclusions 89
9.1 Prepare for a Linked Data Enterprise 89
References 90
Selling and Building Linked Data: Drive Value and Gain Momentum 92
1 The Data Burden 93
2 Driving Value Principles 94
3 Building a Team 97
4 Committing to Something Bigger 99
5 Putting it together 100
6 Conclusions 101
References 103
Part III Techniques for Linking Enterprise Data 104
Enhancing Enterprise 2.0 Ecosystems Using Semantic Web and Linked Data Technologies:The Sem SLATES Approach 106
1 Introduction 107
2 Issues with Current Enterprise 2.0 Ecosystems 108
2.1 Information Fragmentation and Heterogeneity of Data Formats 109
2.2 Knowledge Capture and Re-use 110
2.3 Tagging and Information Retrieval 110
3 SemSLATES: A Social and Semantic Middleware Approach for Enterprise 2.0 111
3.1 The SemSLATES Architecture 112
3.2 Ontologies for Enterprise 2.0 115
3.3 Generating Semantic Annotations Through Software Add-ons 116
3.4 Deploying Additional Services 117
4 Case-study: Enabling SemSLATES at EDF R& D
4.1 Background 118
4.2 Extending Popular Ontologies 119
4.3 Automated SIOC-based Annotations 120
4.4 Knowledge Capture Using UfoWiki 120
4.5 Semantic Tagging Add-ons 122
4.6 Additional Features of the Platform 123
4.6.1 Enhancing the Wiki Features 123
4.6.2 Semantic Search 124
4.6.3 Semantic Mashups 125
5 Conclusion 126
References 127
Linking XBRL Financial Data 130
1 Introduction 130
1.1 XBRL 133
1.1.1 Instances 133
1.1.2 Taxonomies 134
1.2 Related Work 135
2 Approach 136
2.1 XSD2OWL Mapping 137
2.2 XML2RDF Mapping 139
2.3 Algorithm 140
3 Results 140
3.1 Links to External Data 142
3.2 Semantic Integration 145
4 Evaluation 146
4.1 Use Case 149
5 Conclusions and Future Work 149
References 151
Scalable Reasoning Techniques for Semantic Enterprise Data 154
1 Introduction 154
2 Survey of Reasoning Techniques 155
2.1 Traditional Rule Engines 157
2.2 Forward Chaining and the RETE algorithm 158
2.3 Backward Chaining 159
3 Bayesian Networks 160
3.1 Representing Probabilities within the Ontological Model 162
4 Unsupervised Reasoning 163
5 Semantic Reasoning 164
5.1 Performance and Reasoning 166
5.2 Applying Best-First Search (A* Search) to Semantic Reasoning 167
5.3 High-level View of Distributed Reasoning 167
5.4 Map-Reduce and Similar Techniques 168
5.5 Performance and Ontology Engineering 170
6 Semantic Reasoning vs. Business Intelligence 170
7 Best Practices for Application Developers and System Integrators 171
8 Summary 173
References 173
Reliable and Persistent Identification of Linked Data Elements 176
1 Introduction 177
2 Metadata Before the World Wide Web 177
3 Metadata on the World Wide Web 181
4 Persistent URLs 186
5 Extending Persistent URLs for Web Resource Curation 187
6 Redirection of URL Fragments 191
7 Using Persistent URLs and Retrieved Metadata 191
8 Federations of PURL Servers 193
9 Conclusions and Further Work 197
References 198
Part IV Success Stories 202
Linked Data for Fighting Global Hunger: Experiences in setting standards for Agricultural Information Management 204
1 Agricultural information and Semantic Web 204
Promoting trusted URIs for use in Linked Data 206
2 Integrating access using Dublin Core metadata 207
Feedback from application profile users 209
Metadata enrichment and conversion to Linked Data 211
Accepting “whatever you can get” 212
3 AGROVOC and specialized domain ontologies 213
User experience of AGROVOC and AIMS ontologies 217
AGROVOC as a “quarry” of terms 218
Correcting the model for less precision 220
4 Networking, capacity development, and outreach 224
Fishing in a Sea of Agrovoc? 225
The global “coherence” of information about food 227
References 228
Enterprise Linked Data as Core Business Infrastructure 230
1 Introduction 230
2 Motivations 231
3 Garlik’s System Architectures 233
3.1 DataPatrol 234
3.2 QDOS 238
3.2.1 foaf.qdos.com 240
4 Schema Driven Software Deployment 242
5 Technology and the Need to Scale 243
5.1 4store 243
5.2 5store 244
5.2.1 Platform 244
5.2.2 Performance 245
6 Conclusions 245
7 Future Work 246
References 246
Standardizing Legal Content with OWL and RDF 248
1 Introduction 248
1.1 The problem domain 248
1.2 Application of Semantic Web technologies 249
2 Toward a Common Legal Content Format 250
3 OWL Ontology 251
3.1 Creating the ontology 251
3.1.1 Metadata properties 251
3.1.2 Document model 252
3.1.3 Knowledge structures 253
3.2 Domain Ontology Mapping 253
4 Content Architecture 254
4.1 Modularized XHTML + RDFa for Textual Content 254
4.2 RDF for Metadata, Relations and Classifications 255
5 Working with RDF in a Content Supply Chain 256
5.1 The Open World Enigma 257
5.2 Ensuring RDF Data Integrity 257
5.3 Managing Fragmented Ontologies 259
5.4 Managing Performance 259
5.5 Using RDF with XSLT 260
6 Enabling Large-Scale Triple Production 261
6.1 Experimental XSD generation 262
6.2 RDFBeans 263
7 Conclusions 265
References 266
A Role for Semantic Web Technologies in Patient Record Data Collection 268
1 Introduction 269
2 Architectural Styles 270
2.1 REST Architectural Style 270
2.2 Service Oriented Architecture 271
2.2.1 Workflow Systems 272
2.2.2 Process Representation 272
3 Semantic Web Technologies 273
4 SemanticDB Concurrent Data CollectionWorkflow 274
4.1 Requirements 275
4.2 XML and RDF Content Management 276
4.3 RESTful XSLT Services 277
4.4 Declarative AJAX Framework 277
4.5 Implementation 278
5 General Architectural Observations 284
6 Review of Service-oriented Metrics 284
7 Conclusions 287
References 287
Use of Semantic Web technologies on the BBC Web Sites 290
1 Introduction 290
1.1 Linking microsites for cross-domain navigation 291
1.2 Making data available to developers 291
1.3 Making use of the wider Web 292
2 Programme support on theWeb 292
2.1 BBC Programmes 293
2.2 The Programmes Ontology 293
2.2.1 Main terms 293
2.2.2 Tagging programmes 294
2.2.3 A flexible segmentation model 295
2.3 Web identifiers for broadcast radio and television sites 296
3 BBC Music 297
3.1 BBC Music as Linked Data 298
3.2 Web identifiers for BBC Music 298
3.3 The Web as a content management system 299
3.4 Using the BBC Programmes and the BBC Music Linked Data 299
3.4.1 Programmes and locations 299
3.4.2 Artist recommendations 299
4 BBC Wildlife Finder 301
4.1 The Wildlife Ontology 301
4.1.1 Main terms 302
4.1.2 Species as Classes vs Species as Instances 302
4.1.3 Web identifiers - using DBpedia as a controlled vocabulary 303
4.2 Web identifiers 303
4.3 The Web as a Content Management System 305
4.4 The importance of curation 305
5 Journalism 306
5.1 Populating and using the ontology 307
5.2 Future developments 308
6 Conclusion 309
References 310
Glossary 312
Index 316

Produkt-Details

• Erscheint lt. Verlag: 10.11.2010
• Zusatzinfo: XXVI, 291 p.
• Verlagsort: New York
• Sprache: englisch
• Themenwelt: Mathematik / Informatik ► Informatik ► Datenbanken
• Themenwelt: Mathematik / Informatik ► Informatik ► Netzwerke
• Themenwelt: Informatik ► Software Entwicklung ► User Interfaces (HCI)
• Themenwelt: Mathematik / Informatik ► Informatik ► Web / Internet
• Themenwelt: Mathematik / Informatik ► Mathematik ► Finanz- / Wirtschaftsmathematik
• Themenwelt: Wirtschaft ► Betriebswirtschaft / Management ► Wirtschaftsinformatik
• Schlagworte: currentjm • data architecture • inform • information architecture • Information organization • linked data • Management • open standards • organization • Persistent • Persistent Uniform Resource Locators (PURLs) • Resource Description Framework (RDF) • semantic web • Simple Knowledge Organisation System (SKOS) • Text • Uniform Resource Identifiers (URIs) • World Wide Web
• ISBN-10: 1-4419-7665-5 / 1441976655
• ISBN-13: 978-1-4419-7665-9 / 9781441976659