声明
摘要
ABSTRACT
TABLE OF CONTENTS
LIST OF ABBREVIATIONS
CHAPTER 1 INTRODUCTION
1.1 Ontologies
1.1.1 Types of Ontologies
1.1.2 Roles of Ontologies
1.2 Ontology Utilization in Web-based Applications
1.2.1 Overview of Web Video Classification
1.2.2 Overview of Disaster Management
1.3 Motivation
1.4 Main Contributions
1.5 Organization of Dissertation
CHAPTER 2 BACKGROUND AND RELATED WORK
2.1 Web Video Classification
2.1.1 Textual Features based Web Video Classification
2.1.2 Low-level Features based Web Video Classification
2.1.3 Multi-Modality based Web Video Classification
2.2 High-level Concept Detectors and Semantic Relatedness Measures
2.2.1 High-level Concept Detectors
2.2.2 Semantic Relatedness Measures
2.2.3 Concept Detector Selection through Semantic Relatedness Measures
2.3 Formal Ontologies for Extracting Disaster Related Information from Web
CHAPTER 3 WEB VIDEO CATEGORIZATION USING CATEGORY-PREDICTIVE CLASSIFIERS AND CATEGORY-SPECIFIC CONCEPT CLASSIFIERS
3.1 Introduction
3.2 Category-Specific Concept Selection for Web Video Classification
3.2.1 FECS Learning
3.2.2 FECS Construction
3.3 Building CNC and CXC Classifiers
3.3.1 CNC Classifiers
3.3.2 CXC Classifiers
3.4 Integrating CNC Classifiers with CSC Classifiers
3.5 Combining refined CNC classifiers with CXC classifiers
3.6 The Algorithm
3.7 Experimental Results
3.7.1 Dataset and Performance Metrics
3.7.2 Feature Extraction and Pre-Processing
3.7.3 Classifier Selection and Parameters Adjustment
3.7.4 Performance of CNC Classifiers
3.7.5 Refining CNC Classifiers using CSC Classifiers
3.7.6 Fusing Refined CNC Classifiers with CXC Classifiers
3.7.7 Comparison with State-of-the-art Techniques
3.8 Summary
CHAPTER 4 WEB VIDEO CLASSIFICATION WITH VISUAL AND TEXTUAL SEMANTIC TECHNIQUES
4.1 Introduction
4.2 Rendering Semantic Support for Web Video Classification
4.2.1 Fetching Category Discriminative Terms through Open Directory Project and Large Scale Web Videos
4.2.2 Semantic Relevance Computation
4.3 Web Video Classification using Content Features
4.3.2 VSR based Web Video Classification
4.3.3 Content Fusion
4.4 Web Video Classification using Contextual Features
4.4.1 VSM based Web Video Classification
4.4.2 TSR based Web Video Classification
4.4.3 Contextual Fusion
4.5 Integrating Content and Context for Web Video Classification
4.6 The Algorithm
4.7 Experimental Results
4.7.1 Dataset and Evaluation Metrics
4.7.2 Feature Extraction and Pre-Processing
4.7.3 Analysis of Category Classifiers and VSR based Video Classification
4.7.4 Analysis of VSM and TSR based Video Classification
4.7.5 Analysis of Content and Context Fusion
4.8 Summary
CHAPTER 5 DESIGNING ONTOLOGY FOR CONCEPTUALIZATION AND EXTRACTION OF DISASTER INFORMATION FROM WEB
5.1 Introduction
5.2 Proposed Ontology
5.2.1 Defining Domain and Scope of Ontology
5.2.2 Reusing Existing Ontologies
5.2.3 Enumerating Important Terms
5.2.4 Defining Classes and Class Hierarchy
5.2.5 Defining Properties
5.2.6 Defining Property Restrictions
5.3 Ontology Evaluation
5.3.1 Ontology based Semantic Web Crawling
5.4 Experimental Setup and Results
5.4.1 Conventional Crawling,Ontology-driven Crawling,and their Comparison on Relevant Disaster Documents
5.4.2 Conventional Crawling, Ontology-driven Crawling,and their Comparison on Semi-relevant Disaster Documents
5.4.3 Conventional Crawling, Ontology-driven Crawling,and their Comparison on Irrelevant Disaster Documents
5.5 Summary
CHAPTER 6 CONCLUSION AND FUTURE WORK
6.1 Conclusion
6.2 Future Work
ACKNOWLEDGEMENTS
REFERENCES
LIST OF PUBLICATIONS