Cover
Contents
Preface
1.1 Motivation
1.2 Description of the Task
1.3 Previous Work in NLP for Question Answering
1.4 Question Answering at the Text REtrieval Conference
1.4.1 Test Collection
1.4.3 Scoring Metrics
1.5 Book Overview
2.2.1 A Generic Document Retrieval Architecture
2.2.2 A Generic Question Answering Architecture
2.3.1 Background on External Resources
2.3.2 Architecture Description
2.4 Answering Natural Language Questions: An Example
2.5.1 Review of Information Retrieval (IR) for QA
2.5.2 Review of Information Extraction (IE) for QA
2.5.3 Review of Text-Based Inference (TI) for QA
2.5.4 Impact on QA Subproblems
2.6 Summary
3.1 Introduction
3.2.1 Layer 1: Lexical Terms
3.2.2 Layer 2: Inter-Term Relations
3.2.3 Layer 3: Question Stems and Expected Answer Types
3.2.4 Layer 4: Semantic Constraints
3.3.1 Model Description
3.3.3 Application to Answer Extraction
3.4 Construction of Dependency Representations
3.5 Summary
4.1 Introduction
4.2.1 Overview of the Hierarchy
4.2.2 Connecting the Answer Types with WordNet Hierarchies
4.2.3 Correlation between Answer Types and Named Entities
4.3.1 Part of Speech Coverage
4.3.3 Refinement of the Hierarchy Nodes
4.4 Derivation of the Expected Answer Type of a Question
4.4.1 Derivation of the Question Stem and Answer Type Term
4.4.2 Hierarchy Filtering Based on the Question Stem
4.4.3 Hierarchy Search Guided by the Answer Type Term
4.4.4 Extraction of the Expected Answer Type
4.5.1 Refining the Hierarchy of Answer Types
4.5.2 Dynamic Answer Type Categories
4.5.3 Pattern-Based Answer Type Recognition
4.6 Evaluation
4.7 Summary
5.1 Introduction
5.2.1 Factors in the Selection of Question Terms as Keywords
5.2.2 High-Relevance Terms
5.2.5 Assembling Ordered Sequences of Keywords
5.3.1 Query Definition
5.3.3 Control of Passage Granularity
5.4 Summary
6.1 Introduction
6.2.1 Matching the Question on a Passage
6.2.2 Lexical-Matching Relevance Features for Passage Ranking
6.2.3 Passage Ranking Scheme
6.3.1 Named-Entity Based Identification of Candidate Answers
6.3.2 Pattern-Based Identification of Candidate Answers
6.4 Extraction of Answer Strings
6.5 Empirical Ranking of Candidate Answers
6.5.1 Semantic-Matching Relevance Features for Answer Ranking
6.5.2 An Empirical Answer Scoring Formula
6.5.3 Evaluation
6.6.1 Perceptron-Based Learning for Answer Ranking
6.6.2 Evaluation
6.7 Summary
7.1 Introduction
7.2.1 Architecture Overview
7.2.2 Retrieval of Text Passages from Web Search Engines
7.3 Evaluation
7.3.1 Results in Terms of Precision/MRR
7.3.2 Results in Terms of Time Saving
7.4 Summary
8. Related Work
8.1 Question Processing
8.2 Passage Retrieval
8.3 Answer Extraction
9. Conclusion
References
Name Index
Subject Index
Back Cover