Weaving Services and People on the World Wide Web (eBook)

Irwin King is Associate Professor at the Web Intelligence and Social Computing Lab of the Chinese University of Hong Kong, China. He serves as a member of the Engineering Panel with the Research Grants Council (RGC), Hong Kong SAR Government; and is also currently serving as a Vice-President and Governing Board Member of the Asian Pacific Neural Network Assembly (APNNA). In addition, he is also actively involved in the education of students outside the classroom. For example, he has led several ACM Programming Contest Teams to the ACM ICPC World Finals since 2000. His research interests include machine learning, web intelligence & social computing, and multimedia processing. Ricardo Baeza-Yates is VP of Research for Europe and Latin America, leading the Yahoo! Research labs at Barcelona, Spain and Santiago, Chile, and also supervising the lab in Haifa, Israel. Until 2005 he was the director of the Center for Web Research at the Department of Computer Science of the Engineering School of the University of Chile, and ICREA Professor and founder of the Web Research Group at the Dept. of Information and Communication Technologies of Univ. Pompeu Fabra in Barcelona, Spain. He maintains ties with both mentioned universities as a part-time professor for the Ph.D. program. His research interests includes algorithms and data structures, information retrieval, web mining, text and multimedia databases, software and database visualization, and user interfaces.

Preface 5
Contents 8
Contributors 10
Part I Web Services 14
Classification of Automated Search Traffic 15
Introduction 15
Related Work 19
Data Set Description 20
Qualitative Analysis 20
Quantitative Analysis 23
Physical Model Feature Set 24
Behavioral Feature Set 25
Active Learning for High Classifier Accuracy and Rare-Class Discovery 30
Experimental Results 33
Binary Classification Results 35
Multiclass Classification Results 35
Conclusions 37
References 37
Semantic Services for Wikipedia 39
Introduction 40
Wikipedia 41
Knowledge Extraction from Wikipedia 43
Relation Extraction Using Infoboxes 43
Relation Extraction Using Full Text 44
Relation Extraction Using Categories 46
Semantic Editing for Wikipedia 48
A Unified Suggestion Method 48
Applications of the Method 51
Demonstration 53
Enhanced Search for Wikipedia 54
Data and Query Model 54
System Overview 55
Demonstration 58
Conclusion 59
References 59
Context-based Semantic Mediation in Web Service Communities 60
Introduction 60
General Architecture 62
Communities of Web Services 62
Context Representation 63
Domain and Context Ontologies 64
Context Annotation of WSDL 65
Semantic Mediation for Web Service Communities 67
Accessing the Community 67
Details on Context-Based Mediation 67
Interoperability with Semantic Web Services Frameworks: The Case of WSMO 69
Semantic Web Services and WSMO 69
Implementing Context-Based Mediation Through WSMO and IRS-III 70
Related Work 73
Semantics and Mediation for Web Services 73
Communities of Web Services 74
Conclusion 75
References 76
An Effective Aggregation Policy for RSS Services 78
Introduction 78
Related Works 80
Minimum Missing Aggregation Policy 82
Results and Analysis 90
Conclusion and Future Work 94
References 94
Evolution of the Mobile Web 96
Introduction 96
[MobEA 2003] -- Emerging Applications for the Mobile Web 97
[MobEA 2004] -- Multimedia and Collaborative Systems 98
[MobEA 2005] -- Customer Focused Mobile Services 100
[MobEA 2006] -- Empowering the Mobile Web 101
[MobEA 2007] -- Mobile Web in the Developing World 102
[MobEA 2008] -- Personal Rich Social Media 104
The Mobile Web Initiative 105
Conclusion 107
References 107
Personalized Service Creation and Provision for the Mobile Web 109
Introduction 109
Towards the Mobile Web 111
Expanding the World Wide Web to the Mobile Web 111
Opening Up the Mobile Web for Development 112
A Motivating Scenario 114
Design Principles 114
Service Provision Approach 114
Context Information Retrieval 116
Service Creation Approach 117
System Architecture 119
Agent Layer 119
Service and Network Layer 125
Implementation and Evaluation 125
Related Work 127
Conclusion 129
References 130
Selecting the Best Mobile Information Service with Natural Language User Input 132
Introduction 132
Related Work 134
Architecture of the NL3S 136
Multi-Service Ontology 137
Background 137
MO Representation 138
MO Creation 141
Multi-Service Query Language 142
MQL Representation 142
MQL Creation 144
Service Selection Engine 145
Query Identification 145
Service Selection for Accurate Query 146
Service Selection for Similar Query 146
Service Selection for Incomplete Query 147
Adaptive Service Recommendation 148
Implementation and Evaluation 149
Implementation 149
Experiment 1: Accuracy Evaluation 151
Experiment 2: Satisfaction Evaluation 153
Experiment 3: Performance Evaluation 153
Conclusion 154
References 154
Location Concepts for the Web 156
Introduction 156
Standardization Bodies 157
Application Areas 158
Location-Based Services 158
Web of Things 159
Resource-Object Associations 159
Social Networking 160
Scenario 161
Network-Based Positioning Techniques 161
Location on the Web 162
Coordinates 162
Countries 163
Places 164
Location Web Architecture 166
Location Resources 166
HTTP Location Metadata 167
Resource Location Metadata 167
Locative Hyperlinks 168
Scripting Interfaces 168
Geo-Enabled Browser 170
LocProxy 171
Place Markup Language (PlaceML) 171
HTTP Proxy 173
Future Work 174
Conclusions 175
References 175
Ad Hoc Determination of Geographic Regionsfor Concept@Location Queries 178
Introduction 178
Our Approach 179
Aims 180
System Architecture 180
Region-Engine 181
Application Scenarios 186
Arbitrary Terms as Regions 186
Where-is-like Queries 187
Evaluation 188
Performance 188
Quality of Region Representations 189
Geographizing Arbitrary Concepts 197
Related Work 200
Conclusion and Future Work 201
References 202
Acquisition of Vernacular Place Names from Web Sources 204
Introduction 204
Representation of Place 206
Extracting Vernacular Knowledge from Web Sources 208
Mining the Web 208
Geo-References from Business Directories 210
Geo-References from a Social Website 210
Geo-References from Google User Created Geographic Information Sources 212
Determination of the Spatial Extent of Vernacular Place Names 214
Outlier Identification 214
Kernel Density Estimation 214
Current Results 215
Conclusions and Future Work 220
References 221
Part II Social Computing 224
Social Web and Knowledge Management 225
Introduction 225
The Advantage of the Web 226
The Direction of the Workshop 227
The Fallacy of Knowledge Management 2.0 228
Research Directions in Social Web and Knowledge Management 229
Social Web Aware Enterprises 229
Enterprises Not Aware of Social Web Solutions 232
Directions in Active and KIWI 232
KiWi 233
ACTIVE 234
Conclusions 234
References 235
Setting Access Permission through Transitive Relationship in Web-based Social Networks 236
Introduction 236
Related Work 239
From Social Network to Trust network 240
Trust Network and Obfuscation 242
Context Pruning 243
Transitive Trust Network Initialization 244
Permission Value Computation 244
Data Obfuscation 245
Improvement of the Trust Network 246
Improvement 1: By Control Factors 246
Improvement 2: By User Importance 247
Improvement 3: By Query Property 249
Framework Overview 251
Experiment 253
General Characteristics of the Trust Network 253
Control Factors 254
User Importance 255
Discussions 256
Trust Priority 256
Standardized Private Data Levels 257
Other Applications 258
Conclusions 258
References 259
Multiple Interests of Users in Collaborative Tagging Systems 261
Introduction 261
Folksonomies and Personomies 262
Analysis of Personomies 263
Number of Tags and Documents of a User 264
Measuring Diversity of Interests 266
Generating User Profiles of Multiple Interests 268
Community Discovery Algorithms 268
User Profile Generation 269
Evaluation and Discussions 270
Representation of User Profiles 271
Usefulness of the Generated User Profiles 274
Potential Applications 276
Related Work 276
Conclusions 277
References 278
On the Effect of Group Structures on Ranking Strategies in Folksonomies 281
Introduction 281
Folksonomy Systems 283
GroupMe! Folksonomy System 283
GroupMe! Tagging & Grouping Characteristics
Folksonomies 286
Ranking Strategies 287
FolkRank 288
Group-Sensitive FolkRank Algorithms 289
GRank 290
SocialPageRank 291
SocialSimRank 292
Synopsis 293
Evaluations 293
Search Evaluations 293
Tag Recommendation Evaluations 297
Related Work 304
Conclusions 305
References 306
Resolving Person Names in Web People Search 307
Introduction 307
Related Work 309
Data Modeling and Clustering Approaches 311
Single Pass Clustering 312
K-Means Clustering 313
Agglomerative Clustering 314
Probabilistic Latent Semantic Analysis 314
Evaluation Platform 315
Data Set 315
Performance Measures 317
Document Representation 318
PLSA Estimation 319
Experiments and Results 319
Single Pass Clustering 319
K-Means Clustering 321
Agglomerative Clustering 322
PLSA 322
Comparing Methods 323
Group-Level Analysis 325
Discussion and Conclusion 327
References 328
Studies on Editing Patterns in Large-Scale Wikis 330
Introduction 330
Macro-Scale experiment 331
Dataset 332
Procedure 332
Tools Developed 336
Results 337
Summary 343
Micro-Scale Experiment 343
Procedure 344
Results 345
Summary 349
Conclusion 351
Current Strengths 351
Current Weaknesses 352
Toward Solutions 353
References 353
Index 355