CONTENTS -- PREFACE -- Part I. Content Extraction and Web Mining -- CHAPTER 1 CLUSTERING OF WEB DOCUMENTS USING A GRAPH MODEL -- 1. Introduction -- 2. Graphs: Formal Notation -- 3. The Extended k-Means Clustering Algorithm -- 4. Clustering of Web Documents using the Graph Model -- 5. Experimental Results -- Acknowledgments -- References -- CHAPTER 2 APPLICATIONS OF GRAPH PROBING TO WEB DOCUMENT ANALYSIS -- 1. Introduction -- 2. Related Work -- 3. A Formalism for Graph Probing -- 4. Experimental Evaluation -- 4.1. Graph Model -- 4.2. Generating "Random" Collections of Web Pages -- 4.3. Experiment #1: Full Graph Matching -- 4.4. Experiment #2: Subgraph Matching -- 5. Conclusions -- 6. Acknowledgments -- References -- CHAPTER 3 WEB STRUCTURE ANALYSIS FOR INFORMATION MINING -- 1. Introduction -- 2. Object Model Architecture -- 2.1. HTML Parsing Library -- 2.2. Single-Slot HTML Parsing Functions -- 2.3. Multi-Slot/Pattern HTML Parsing Functions -- 3. User Interface -- 4. News Article Extraction -- 5. Link Extraction -- 6. Stock Quote Extraction -- 7. Conclusion -- Acknowledgment -- References -- CHAPTER 4 NATURAL LANGUAGE PROCESSING FOR WEB DOCUMENT ANALYSIS -- 1. Introduction -- 2. Design Principles -- 2.1. Why XML? -- 2.2. User Orientation -- 2.3. Portability -- 3. Document Suite XDOC -- 3.1. Preprocessing Module -- 3.1.1. HTML Cleaner -- 3.1.2. Structure Tagger -- 3.1.3. POS Tagger -- 3.2. Syntactic Module -- 3.2.1. Syntactic Parser -- 3.2.2. Phrase Detector -- 3.3. Corpus Based Module -- 3.4. Semantic Module -- 3.4.1. Semantic Tagger -- 3.4.2. Case Frame Analysis -- 3.4.3. Semantic Interpretation of Syntactic Structure -- 4. Related Work -- 5. Conclusion -- References -- Part II. Document Analysis for Adaptive Content Delivery -- CHAPTER 5 REFLOWABLE DOCUMENT IMAGES -- 1. Introduction -- 2. Image and Layout Analysis -- 2.1. Text/Image Segmentation.

2.2. Preprocessing -- 2.3. Layout Analysis -- 3. HTML-Based Representations -- 4. Reader Applications -- 5. New Document Formats -- 6. Summary and Conclusions -- Acknowledgments -- References -- CHAPTER 6 EXTRACTION AND MANAGEMENT OF CONTENT FROM HTML DOCUMENTS -- 1. Introduction -- 2. Research Direction -- 3. Current State of the Art -- 3.1 Handcrafting -- 3.2 Transcoding -- 3.3 Adaptive Re-authoring -- 4. Proposed Approach -- 4.1. Web Page Segmentation -- 4.2. Contextual Analysis and Segment Labeling -- 4.3. Web-Page Summarization -- 4.4. Post-processing -- 4.5. Overall Summary of the Content Extraction and Display Process -- 5. Results -- 6. Discussion -- 6.1 Web Page Segmentation -- 6.2 Contextual Analysis and Segment Labeling -- 6.3 Web-Page Summarization -- 6.4. Display Capabilities -- 6.5. Language Independence -- 6.6. Current State of Research -- 6.7. Supported Devices -- 7. Concluding Remarks -- References -- CHAPTER 7 HTML PAGE ANALYSIS BASED ON VISUAL CUES -- 1. Introduction -- 1.1. Document Analysis for Search Engines -- 1.2. Document Analysis for Adaptive Content Delivery -- 2. Visual Similarity of HTML Objects -- 2.1. Visual Similarity of Simple Objects -- 2.2. Visual Similarity of Container Objects -- 3. Pattern Detection and Construction of Structured Documents -- 3.1. Quantization -- 3.2. Frequency Counting -- 3.3. Selection and Confirmation -- 3.4. Construction of Structured Document -- 3.5. Special Consideration of HTML Tables -- 4. Experimental Results and Analysis -- 5. Application in an Adaptive Content Delivery System -- 6. Conclusions -- Acknowledgements -- References -- Part III. Table Understanding on the Web -- CHAPTER 8 AUTOMATIC TABLE DETECTION IN HTML DOCUMENTS -- 1. Introduction -- 2. Features for Web Table Detection -- 2.1. Layout Features -- 2.2. Content Type Features -- 2.3. Word Group Feature.

2.3.1. Vector Space Approach -- 2.3.2. Naive Bayes Approach -- 2.3.3. Weighted kNN Approach -- 3. Classification Schemes -- 3.1. Decision Tree -- 3.2. SVM -- 4. Data Collection and Ground Truthing -- 4.1. Data Collection -- 4.2. Ground Truthing -- 4.3. Database Description -- 5. Experiments -- 6. Conclusion and Future Work -- 7. Acknowledgment -- References -- CHAPTER 9 A WRAPPER INDUCTION SYSTEM fOR COMPLEX DOCUMENTS, AND ITS APPLICATION TO TABULAR DATA ON THE WEB -- 1. Introduction -- 2. Issues in Wrapper Learning -- 3. An Extensible Wrapper Learning System -- 3.1. Architecture of the Learning System -- 3.2. A Generic Representation for Structured Documents -- 3.3. A Generic Representation for Extractors -- 3.4. Representing Training Data -- 3.5. Designing a Bias -- 3.6. The Master Learning Algorithm -- 4. Additional Builders -- 4.1. Composite Builders -- 4.2. Format-based Extraction -- 5. Table-based Extraction -- 5.1. Representing Tables on the World Wide Web -- 5.2. Classes of Table Presentation in HTML -- 5.3. An Abstract Geometric Table Model -- 5.4. Table Location -- 5.4.1. Application of Machine Learning -- 5.4.2. Features -- 5.4.3. Experimental Results -- 5.5. Exploiting Table Context -- 5.6. Exploiting the Table Models -- 6. Experiments -- 7. Conclusions -- Acknowledgments -- References -- CHAPTER 10 EXTRACTING ATTRIBUTES AND THEIR VALUES FROM WEB PAGES -- 1. Introduction -- 2. Ontology Extraction from HTML Tables -- 2.1. Table Structure Recognition -- 2.2. Table Integration -- 3. List Analysis -- 3.1. Term Definition -- 3.2. State Sequence Estimation Module -- 3.2.1. Estimation of P(bls) and P(s) -- 3.2.2. Estimation of P(s'ls) and P(cls,s') -- 4. Experiments -- 5. Conclusion and Future Work -- Acknowledgments -- References -- Part IV. Web Image Analysis and Retrieval.

CHAPTER 11 A FUZZY APPROACH TO TEXT SEGMENTATION IN WEB IMAGES BASED ON HUMAN COLOUR PERCEPTION -- 1. Introduction -- 2. Colour Segmentation Method -- 2.1. Colour Distance -- 2.2. Colour Connected-Component Identification -- 2.3. Propinquity Features -- 2.4. Fuzzy Inference -- 2.5. Colour Component Aggregation -- 3. Results and Discussion -- Acknowledgement -- References -- CHAPTER 12 SEARCHING FOR IMAGES ON THE WEB USING TEXTUAL METADATA -- 1. Introduction -- 2. Background -- 3. Image Search Architecture -- 4. Image Search Experiment -- 4.1. Results -- 5. Conclusion and Future Work -- Acknowledgments -- References -- CHAPTER 13 AN ANATOMY OF A LARGE-SCALE IMAGE SEARCH ENGINE -- 1. Introduction -- 1.1. An Illustrative Example -- 2. System Architecture -- 2.1. Feature Extractor -- 2.2. High-dimensional Indexer -- 2.3. Perception-based Search Engine -- 2.4. Content-based Search Engine -- 3. Active Learning Algorithms -- 3.1. MEGA -- 3.2. SVMActive -- 3.3. Hybrid Algorithms -- 4. Experiments -- 5. Results and Discussion -- 5.1. MEGA, SVMActive, and Pipelining MEGA with SVMActive -- 5.2. Observations -- 6. Comparison with Related Work -- 7. Conclusion -- References -- Part V. New Opportunities -- CHAPTER 14 WEB SECURITY AND DOCUMENT IMAGE ANALYSIS -- 1. Introduction -- 1.1. An Influential Precursor: Turing Tests -- 1.2. Robot Exclusion Conventions -- 1.3. Primitive Means -- 1.4. First Use: The Add-URL Problem -- 1.5. The ChatRoom Problem -- 1.6. Screening Financial Accounts -- 1.7. PessimalPrint -- 2. The First International HIP Workshop -- 3. Implications for DIA Research -- 4. Discussion -- 5. Acknowledgments -- References -- CHAPTER 15 EXPLOITING WWW RESOURCES IN EXPERIMENTAL DOCUMENT ANALYSIS RESEARCH -- 1. Introduction -- 2. Traditional Approaches -- 3. Exploiting WWW Resources -- 4. Proof of Concept: Analysis of a Digital Library.

4.1. Optical Character Recognition -- 4.2. Table Detection -- 5. Examples of Other WWW Resources -- 6. Conclusions -- 7. Acknowledgments -- References -- CHAPTER 16 STRUCTURED MEDIA FOR AUTHORING MULTIMEDIA DOCUMENTS -- 1. Motivation -- 2. Multimedia Authoring -- 3. Multimedia Modelling -- 3.1. Video Content Modelling -- 3.1.1. General Model -- 3.1.2. Description of the Video Structure -- 3.1.3. Extensions of MPEG-7 for Model Definition -- 3.2. Document Modelling with Structured Media -- 3.2.1. Multimedia Document Model -- 3.2.2. Model Extensions -- 4. Multimedia Document Authoring System -- 4.1. Video Content Description Editing Environment -- 4.2. Authoring Multimedia Documents -- 5. Conclusion -- References -- CHAPTER 17 DOCUMENT ANALYSIS REVISITED FOR WEB DOCUMENTS -- 1. Introduction -- 2. Document Model Evolution: An Analysis Perspective -- 3. Web Document Analysis -- 3.1. Goals of Web Document Analysis -- 3.2. Specificities of Web Document Analysis -- 3.2.1. Dealing with Heterogeneous Formats -- 3.2.2. Dealing with Links -- 3.2.3 Dealing with Images and Graphics -- 3.2.4. Dealing with Interactive Aspects of Web Documents -- 4. Some Relevant Applications -- 4.1. Extracting Rich Structures from a Large Collection of Documents -- 4.2. Extracting Structure from Interconnected Documents -- 4.3. Dynamic Aspects of Web Documents -- 4.4. Generation of Metadata -- 5. Methodological Issues -- 5.1. Techniques and Methods -- 5.2. A Detailed Example -- 6 Conclusion and Perspectives -- References -- AUTHOR INDEX.