Analyzing and Influencing Search Engine Results (eBook)

Dr. Sönke Lieberam-Schmidt completed this doctoral thesis at the University of Siegen at the Institute of Management Information Sciences of Prof. Dr. Erwin Pesch using his several years of work experience on international consulting projects.

Foreword 6
Acknowledgements 7
Contents 8
List of Figures 11
List of Tables 14
List of Abbreviations 15
Introduction 17
1 Web Search 20
1.1 Information Demand 22
1.1.1 User’s Information Need 22
1.1.2 User Behavior 25
1.2 Search Tools 28
1.3 How Search EnginesWork 34
1.3.1 Crawler 35
1.3.2 Indexer 48
1.3.3 Analyzer 50
1.3.4 Searcher 51
1.3.5 Presenter 52
1.4 A New Approach to Relevancy 53
1.5 Evolution ofWeb Search Engines 61
2 Web Structure 64
2.1 Evolution of Information Supply 64
2.2 Web Content 65
2.3 TheWeb Graph 68
2.3.1 Reasons for Regarding the Web Graph 69
2.3.2 Properties of the Web Graph 70
2.3.3 Representation and Storage of the Web Graph 76
2.4 Link Analysis 82
2.4.1 Ranking 82
2.4.2 Authorities and Hubs 86
2.4.3 Other Link-based Methods 92
2.5 Data Structures 98
2.5.1 Feasible Data Structures 99
2.5.2 Evaluation 104
2.5.3 Selection 108
2.6 HITS in Practice 110
2.6.1 Implementation of the Algorithm 110
2.6.2 Experimental Results and Evaluation 115
3 Result Clustering 119
3.1 Clustering Methods 120
3.2 Web Clustering 122
3.3 Text Based Approaches 123
3.4 Link Based Approaches 126
3.4.1 Robust Clustering Algorithm 127
3.4.2 Inclusion of Link Direction 133
3.4.3 Architecture 135
3.4.4 Experimental Results 137
3.5 Combined Approaches 139
3.5.1 Architecture 139
3.5.2 Preprocessing 140
3.5.3 Document Representation 144
3.5.4 Clustering Method Combination 147
3.5.5 Implementation 165
3.5.6 Experimental Results 167
4 Search Engine Optimization 176
4.1 Objectives 176
4.1.1 Website Visitors 179
4.1.2 Result Page Position 181
4.2 Possibilities and Means 184
4.2.1 Visibility 186
4.2.2 Static Relevance 187
4.2.3 Dynamic Relevance 188
4.3 Keyword Selection 189
4.3.1 Keyword Selection Position Model (KSPM) 191
4.3.2 Enhanced Keyword Selection Position Model (EKSPM) 195
4.3.3 Keyword Selection Top Position Model (KSTPM) 197
4.4 Limits 198
4.4.1 Search Engine Spam 199
4.4.2 Spamming Methods 200
4.5 Implementation 201
4.5.1 Tasks 202
4.5.2 Website Creation Process 204
4.6 Practical Results 208
4.6.1 Project and Test Websites 208
4.6.2 Reporting and Results 210
5 Conclusion 217
Bibliography 219
Index 231