Systematically presents, describes and discusses representative algorithms for Web community construction and analysisHighlights various important applications of Web communitySummarizes the main work in this area, and identifies several open research directions that readers can pursue in the futureIncludes supplementary material: sn.pub/extras

InhaltsangabeChapter 1: Introduction (10 pages) -- Web Search, -- Information Filtering -- Web Community Chapter 2: Preliminaries (30 pages) -- Statistics -- Similarity -- Markov Model -- Matrix Expression of Hyperlinks -- Eigenvector, Principle Engenvector, Secondary Engenvector -- Singular Value Decomposition (SVD) of Matrix -- Graph Theory Basis (Random walk) Chapter 3: HITS and Related Algorithms (50 pages) -- The Original HITS -- The Stability issues -- The Randomized HITS -- The Subspace HITS -- Weighted HITS -- Vector Space Model (VSM) -- Cover Density Ranking (CDR) -- The In-depth Analysis of the HITS -- HITS Improvement (a significant improvement to clever algorithm) -- Noise Page Elimination Algorithm Based on SVD -- The PHITS algorithm (probabilistic HITS) -- SALSA (Stochastic algorithm) -- Random Walks and the Kleinberg Algorithm Chapter 4: PageRank Related Algorithms (50 pages) -- The Original PageRank -- Probability Combination of Link and Content Information in PageRank -- Topic-Sensitve PageRank -- Search-Order: Breadth-First, Backlink, Random -- Quadratic Extrapolation -- Exporing the Block Structure of the Web for Computing PageRank -- Second Eignevalue of the Google Matrix -- A Latent Linkage Information (LLI) Algorithm -- WebPage Scoring Systems (WPSS) -- Rank Aggregation -- Random Suffer Method -- Voting Model -- SimRank (graph-based) -- When Experts Agree: Using Non-Affliated Experts to Rank Popular Topics -- PageRank, HITS and a Unified Framework for Link Analysis Chapter 5: Web Classification and Clustering (50 pages) -- Web Document Similarity Measurement -- Web Document Classification Based on Hyperlinks and Document Semantics -- Clustering Hypertext with Applications to Web Search -- Link-based Clustering to Improve Web Search Results -- Measure Similarity of Interest for Clustering Web-Users -- Clustering of Web Users Using Session-based Similarity Measures -- Scalable Techniques for Clustering the Web -- Clustering web surfers with mixtures of hidden Markov Models -- Clustering User Queries of a Search Engine -- Using Web Structure for Classifying and Describing Web Pages -- Matrix-Based Hierarchical Clustering Algorithms Chapter 6: Web Log/Content Mining for Web Community (50 pages) -- Cut-and-Pick Transactions for Proxy Log Mining -- Mining Web Logs to Improve Website Organization -- Extracting Large-Scale Knowledge Bases from the Web -- Mining the Space of Graph Properties -- Discovering Test Set Regularities in Relational Domains (classification) -- Enhanced Hypertext Categorization Using Hyperlinks -- The Structure of Broad Topics on the Web -- Discovering Unexpected Information from Your Competitors' Web Sites -- On Integrating Catalogs -- Web Community Mining and Web Log Mi

Hersteller:
Springer Verlag GmbH
juergen.hartmann@springer.com
Tiergartenstr. 17
DE 69121 Heidelberg