Graph-theoretic techniques for web content mining /

Saved in:

Bibliographic Details
Imprint:	[Hackensack], N.J. ; London : World Scientific, 2005.
Description:	1 online resource (xi, 235 pages) : illustrations.
Language:	English
Series:	Series in machine perception and artificial intelligence ; v. 62 Series in machine perception and artificial intelligence ; v. 62.
Subject:	Data mining. Graph theory -- Data processing. Algorithms. Multidimensional scaling. Algorithms. Data mining. Graph theory -- Data processing. Multidimensional scaling. Electronic books.
Format:	E-Resource Book
URL for this record:	http://pi.lib.uchicago.edu/1001/cat/bib/11201035

Hidden Bibliographic Details
Other authors / contributors:	Schenker, Adam.
ISBN:	9789812563392 9812563393 9812569456 9789812569455 9812563393
Digital file characteristics:	data file
Notes:	Includes bibliographical references and index. Print version record.
Summary:	This book describes exciting new opportunities for utilizing robust graph representations of data with common machine learning algorithms. Graphs can model additional information which is often not present in commonly used data representations, such as vectors. Through the use of graph distance - a relatively new approach for determining graph similarity - the authors show how well-known algorithms, such as k-means clustering and k-nearest neighbors classification, can be easily extended to work with graphs instead of vectors. This allows for the utilization of additional information found in graph representations, while at the same time employing well-known, proven algorithms. To demonstrate and investigate these novel techniques, the authors have selected the domain of web content mining, which involves the clustering and classification of web documents based on their textual substance. Several methods of representing web document content by graphs are introduced; an interesting feature of these representations is that they allow for a polynomial time distance computation, something which is typically an NP-complete problem when using graphs. Experimental results are reported for both clustering and classification in three web document collections, using a variety of graph representations, distance measures, and algorithm parameters. In addition, this book describes several other related topics, many of which provide excellent starting points for researchers and students interested in exploring this new area of machine learning further. These topics include creating graph-based multiple classifier ensembles through random node selection and visualization of graph-based data using multidimensional scaling.
Other form:	Print version: Graph-theoretic techniques for web content mining. [Hackensack], N.J. ; London : World Scientific, ©2005 9812563393

Table of Contents:

Cover
Preface
Acknowledgements
Contents
Chapter 1 Introduction to Web Mining
1.1 Overview of Web Mining Methodologies
1.2 Traditional Information Retrieval Techniques
1.2.1 Vector-based distance measures
1.2.2 Special considerations for web documents
1.3 Overview of Remaining Chapters
Chapter 2 Graph Similarity Techniques
2.1 Graph and Subgraph Isomorphism
2.2 Graph Edit Distance
2.3 Maximum Common Subgraph / Minimum Common Supergraph Approach
2.4 State Space Search Approach
2.5 Probabilistic Approach
2.6 Distance Preservation Approach
2.7 Relaxation Approaches
2.8 Mean and Median of Graphs
2.9 Summary
Chapter 3 Graph Models for Web Documents
3.1 Pre-Processing
3.2 Graph Representations of Web Documents
3.3 Complexity Analysis
3.4 Web Document Data Sets
Chapter 4 Graph-Based Clustering
4.1 The Graph-Based k-Means Clustering Algorithm
4.2 Clustering Performance Measures
4.3 Comparison with Previously Published Results
4.4 Comparison of Different Graph-Theoretical Distance Measures and Graph Representations for Graph-Based Clustering
4.4.1 Comparison of distance measures
4.4.2 Comparison of graph representations
4.5 Comparison of Clustering Algorithms
4.6 Visualization of Graph Clustering
4.7 The Graph-Based Global k-Means Algorithm
4.7.1 Global k-means vs. random initialization
4.7.2 Optimum number of clusters
Chapter 5 Graph-Based Classification
5.1 The k-Nearest Neighbors Algorithm
5.1.1 Traditional method
5.1.2 Graph-based approach
5.1.3 Experimental results
5.2 Graph-Based Multiple Classifier Ensembles
5.2.1 Basic algorithm
5.2.2 Experimental results
Chapter 6 The Graph Hierarchy Construction Algorithm for Web Search Clustering
6.1 Cluster Hierarchy Construction Algorithm (CHCA)
6.1.1 A review of inheritance
6.1.2 Brief overview of CHCA
6.1.3 CHCA in detail
6.1.4 CHCA: An example
6.1.5 Examination of CHCA as a clustering method
6.2 Application of CHCA to Search Results Processing
6.2.1 Asynchronous search
6.2.2 Implementation, input preparation and pre-processing
6.2.3 Selection of parameters for web search
6.3 Examples of Results
6.3.1 Comparison with Grouper
6.3.2 Comparison with Vivisimo
6.4 Graph Hierarchy Construction Algorithm (GHCA)
6.4.1 Parameters
6.4.2 Graph creation and pre-processing
6.4.3 Graph Hierarchy Construction Algorithm (GHCA)
6.4.4 GHCA examples
6.5 Comments
Chapter 7 Conclusions and Future Work
Appendix A Graph Examples
College of Engineering
Professional Engineering
Professional Registration
Preparation for Engineering
Student Computer Policy
Departments and Programs
Chemical Engineering
Civil and Environmental Engineering
Computer Science and Engineering
Electrical Engineering
Industrial and Management Systems Engineering
Mechanical Engineering
Computer Service (SC) Courses
College Computing Facilities
Cooperative Education and Internship Programs
Army, Air Force & Navy R.O.T.C. For Engineering Students
Appendix B List of Stop Words
Bibliography
Index.

Graph-theoretic techniques for web content mining /

Similar Items