Web Mining and Social Networking

by Xu, Guandong; Zhang, Yanchun; Li, Lin

• ISBN13: 9781441977342
• ISBN10: 1441977341
• Format: Hardcover
• Copyright: 2010-10-28
• Publisher: Springer-Verlag New York Inc

Summary

This book examines the techniques and applications involved in the Web Mining, Web Personalization and Recommendation and Web Community Analysis domains, including a detailed presentation of the principles, developed algorithms, and systems of the research in these areas. The applications of web mining, and the issue of how to incorporate web mining into web personalization and recommendation systems are also reviewed. Additionally, the volume explores web community mining and analysis to find the structural, organizational and temporal developments of web communities and reveal the societal sense of individuals or communities. The volume will benefit both academic and industry communities interested in the techniques and applications of web search, web data management, web mining and web knowledge discovery, as well as web community and social network analysis.

Table of Contents

Foundation
Introduction	p. 3
Background	p. 3
Data Mining and Web Mining	p. 5
Web Community and Social Network Analysis	p. 7
Characteristics of Web Data	p. 7
Web Community	p. 8
Social Networking	p. 9
Summary of Chapters	p. 10
Audience of This Book	p. 11
Theoretical Backgrounds	p. 13
Web Data Model	p. 13
Textual, Linkage and Usage Expressions	p. 14
Similarity Functions	p. 16
Correlation-based Similarity	p. 17
Cosine-Based Similarity	p. 17
Eigenvector, Principal Eigenvector	p. 17
Singular Value Decomposition (SVD) of Matrix	p. 19
Tensor Expression and Decomposition	p. 20
Information Retrieval Performance Evaluation Metrics	p. 22
Performance measures	p. 22
Web Recommendation Evaluation Metrics	p. 24
Basic Concepts in Social Networks	p. 25
Basic Metrics of Social Network	p. 25
Social Network over the Web	p. 26
Algorithms and Techniques	p. 29
Association Rule Mining	p. 29
Association Rule Mining Problem	p. 29
Basic Algorithms for Association Rule Mining	p. 31
Sequential Pattern Mining	p. 36
Supervised Learning	p. 46
Nearest Neighbor Classifiers	p. 46
Decision Tree	p. 46
Bayesian Classifiers	p. 49
Neural Networks Classifier	p. 50
Unsupervised Learning	p. 52
The k-Means Algorithm	p. 52
Hierarchical Clustering	p. 53
Density based Clustering	p. 55
Semi-supervised Learning	p. 56
Self-Training	p. 56
Co-Training	p. 57
Generative Models	p. 58
Graph based Methods	p. 59
Markov Models	p. 59
Regular Markov Models	p. 60
Hidden Markov Models	p. 61
K-Nearest-Neighboring	p. 62
Content-based Recommendation	p. 62
Collaborative Filtering Recommendation	p. 63
Memory-based collaborative recommendation	p. 63
Model-based Recommendation	p. 64
Social Network Analysis	p. 64
Detecting Community Structure in Networks	p. 64
The Evolution of Social Networks	p. 67
Web Mining: Techniques and Applications
Web Content Mining	p. 71
Vector Space Model	p. 71
Web Search	p. 73
Activities on Web archiving	p. 73
Web Crawling	p. 74
Personalized Web Search	p. 76
Feature Enrichment of Short Texts	p. 77
Latent Semantic Indexing	p. 79
Automatic Topic Extraction from Web Documents	p. 80
Topic Models	p. 80
Topic Models for Web Documents	p. 83
Inference and Parameter Estimation	p. 84
Opinion Search and Opinion Spam	p. 84
Opinion Search	p. 85
Opinion Spam	p. 86
Web Linkage Mining	p. 89
Web Search and Hyperlink	p. 89
Co-citation and Bibliographic Coupling	p. 90
Co-citation	p. 90
Bibliographic Coupling	p. 90
PageRank and HITS Algorithms	p. 91
PageRank	p. 91
HITS	p. 93
Web Community Discovery	p. 95
Bipartite Cores as Communities	p. 96
Network Flow/Cut-based Notions of Communities	p. 97
Web Community Chart	p. 97
Web Graph Measurement and Modeling	p. 100
Graph Terminologies	p. 101
Power-law Distribution	p. 101
Power-law Connectivity of the Web Graph	p. 101
Bow-tie Structure of the Web Graph	p. 102
Using Link Information for Web Page Classification	p. 102
Using Web Structure for Classifying and Describing Web Pages	p. 103
Using Implicit and Explicit Links for Web Page Classification	p. 105
Web Usage Mining	p. 109
Modeling Web User Interests using Clustering	p. 109
Measuring Similarity of Interest for Clustering Web Users	p. 109
Clustering Web Users using Latent Semantic Indexing	p. 115
Web Usage Mining using Probabilistic Latent Semantic Analysis	p. 118
Probabilistic Latent Semantic Analysis Model	p. 118
Constructing User Access Pattern and Identifying Latent Factor with PLSA	p. 120
Finding User Access Pattern via Latent Dirichlet Allocation Model	p. 124
Latent Dirichlet Allocation Model	p. 124
Modeling User Navigational Task via LDA	p. 128
Co-Clustering Analysis of weblogs using Bipartite Spectral Projection Approach	p. 130
Problem Formulation	p. 131
An Example of Usage Bipartite Graph	p. 132
Clustering User Sessions and Web Pages	p. 132
Web Usage Mining Applications	p. 133
Mining Web Logs to Improve Website Organization	p. 134
Clustering User Queries from Web logs for Related Query	p. 137
Using Ontology-Based User Preferences to Improve Web Search	p. 141
Social Networking and Web Recommendation: Techniques and Applications
Extracting and Analyzing Web Social Networks	p. 145
Extracting Evolution of Web Community from a Series of Web Archive	p. 145
Types of Changes	p. 146
Evolution Metrics	p. 146
Web Archives and Graphs	p. 148
Evolution of Web Community Charts	p. 148
Temporal Analysis on Semantic Graph using Three-Way Tensor Decomposition	p. 153
Background	p. 153
Algorithms	p. 155
Examples of Formed Community	p. 156
Analysis of Communities and Their Evolutions in Dynamic Networks	p. 157
Motivation	p. 158
Problem Formulation	p. 159
Algorithm	p. 160
Community Discovery Examples	p. 161
Socio-Sense: A System for Analyzing the Societal Behavior from Web Archive	p. 161
System Overview	p. 163
Web Structural Analysis	p. 163
Web Temporal Analysis	p. 165
Consumer Behavior Analysis	p. 166
Web Mining and Recommendation Systems	p. 169
User-based and Item-based Collaborative Filtering Recommender Systems	p. 169
User-based Collaborative Filtering	p. 170
Item-based Collaborative Filtering Algorithm	p. 171
Performance Evaluation	p. 174
A Hybrid User-based and Item-based Web Recommendation System	p. 175
Problem Domain	p. 175
Hybrid User and Item-based Approach	p. 176
Experimental Observations	p. 178
User Profiling for Web Recommendation Based on PLSA and LDA Model	p. 178
Recommendation Algorithm based on PLSA Model	p. 178
Recommendation Algorithm Based on LDA Model	p. 181
Combing Long-Term Web Achieves and Logs for Web Query Recommendation	p. 183
Combinational CF Approach for Personalized Community Recommendation	p. 185
CCF: Combinational Collaborative Filtering	p. 186
C-U and C-D Baseline Models	p. 186
CCF Model	p. 187
Conclusions	p. 189
Summary	p. 189
Future Directions	p. 191
References	p. 195
Table of Contents provided by Ingram. All Rights Reserved.