Question Answering for the Curated Web: Tasks and Methods in QA over Knowledge Bases and Text Collections
- Length: 194 pages
- Edition: 1
- Language: English
- Publisher: Morgan & Claypool
- Publication Date: 2021-10-28
- ISBN-10: 1636392385
- ISBN-13: 9781636392387
- Sales Rank: #0 (See Top 100 Books)
Question answering (QA) systems on the Web try to provide crisp answers to information needs posed in natural language, replacing the traditional ranked list of documents. QA, posing a multitude of research challenges, has emerged as one of the most actively investigated topics in information retrieval, natural language processing, and the artificial intelligence communities today. The flip side of such diverse and active interest is that publications are highly fragmented across several venues in the above communities, making it very difficult for new entrants to the field to get a good overview of the topic.
Through this book, we make an attempt towards mitigating the above problem by providing an overview of the state-of-the-art in question answering. We cover the twin paradigms of curated Web sources used in QA tasks ‒ trusted text collections like Wikipedia, and objective information distilled into large-scale knowledge bases. We discuss distinct methodologies that have been applied to solve the QA problem in both these paradigms, using instantiations of recent systems for illustration. We begin with an overview of the problem setup and evaluation, cover notable sub-topics like open-domain, multi-hop, and conversational QA in depth, and conclude with key insights and emerging topics. We believe that this resource is a valuable contribution towards a unified view on QA, helping graduate students and researchers planning to work on this topic in the near future.
Preface Acknowledgments Introduction Motivation Perspectives Relevance to the Community Intended Audience Scope of QA in this Book QA over Knowledge Bases Setup Introducing Knowledge Bases Basic Concepts Storing Contextual Information Wikidata as an Active KB QA over Knowledge Bases Basic Concepts Generating Answers Evaluating KB-QA Systems Organization Getting Started with Simple Questions Introducing Simple QA Answering with Templates Answering with KG Embeddings Wrapping up Simple QA Complex Question Answering Introducing Complex QA Answering with Staged Graphs Answering with Compact Subgraphs Answering with Belief Propagation Wrapping up Complex QA Answering over Heterogeneous Sources Introducing Heterogeneous QA Answering with Early Fusion Answering with Unified Representations Unifying with SPO Triples Unifying with Verbalizations Wrapping up Heterogeneous QA Conversational Question Answering Introducing Conversational QA Answering with Graph Expansion Answering with Seq2seq Models Answering with Reformulations Wrapping up Conversational QA Part I: Summary and Insights Mapping Systems to Methodologies Structured Queries: With or Without? Quad Chart for White Space Analysis Deploying a KB-QA System Quickly QA over Text Collections Setup Question Type Answer Type Context Type Evaluation Benchmarks Organization Reading Comprehension Introducing Reading Comprehension A Deep Learning Approach The Reading Comprehension Pipeline Token Representation Input Representation Interaction Modeling Attention Mechanism Answer Modeling The DrQA Reader The Transformer Architecture BERT and Transfer Learning The BERT Model Pre-Training BERT Fine-Tuning BERT for QA Wrapping up Reading Comprehension Open-Domain Question Answering Introducing Open-Domain QA The Retriever-Reader Framework Challenges for Retrieve-and-Read Frameworks Learning with Distant Supervision Relaxed vs. Restrictive Distant Supervision Answer Generation Parameterized Retriever Dual-Encoder Models Retrieval with Parameterized Retrievers Nearest Neighbor Retrieval Training Dual-Encoder Models Pre-Training for QA Wrapping up Open-Domain QA Multi-Hop Question Answering Introducing Multi-Hop QA Multi-Turn Retrieval Reasoning Paths Treating Text as a Knowledge Base Wrapping up Multi-Hop QA Conversational Question Answering Introducing Conversational QA Challenges in Conversational QA Datasets in Conversational QA The Pointer-Generator Model Answer Extraction Generative Answers Using PGNet Flow Models Wrapping up Conversational QA Part II: Summary and Insights Unsupervised Embeddings and Transfer Learning Utility of a Parametric Retriever Question and Passage Understanding Simple Design Choices Deploying a Text-QA System Quickly Requirements for Commercial QA Systems Open Directions References Authors' Biographies
Donate to keep this site alive
1. Disable the AdBlock plugin. Otherwise, you may not get any links.
2. Solve the CAPTCHA.
3. Click download link.
4. Lead to download server to download.