Ontologies and Adaptivity in Dialogue for Question Answering

Question answering (QA) has become one of the fastest growing topics in computational linguistics and information access. To advance research in the area of dialogue-based question answering, we propose a combination of methods from different scientific fields (i.e., information retrieval, dialogue systems, semantic web and machine learning). This book sheds light on adaptable dialogue-based question answering. We demonstrate the technical and computational feasibility of the proposed ideas, the introspective methods in particular, by beginning with an extensive introduction to the dialogical problem domain which motivates the technical implementation. The ideas have been carried out in a mature natural language processing (NLP) system, the SmartWeb dialogue system, which was developed between 2004 and 2007 by partners from academia and industry. We have attempted to make this book a self-containing text and provide an extra section on the interdisciplinary scientific background. The target audience for this book comprises of researchers and students interested in the application potential of semantic technologies for difficult AI tasks such as working dialogue and QA systems.

Chapter 1. Introduction (PDF)
Chapter 2. Scientific Background

Chapter 3. Ontology-Based Dialogue Processing
Chapter 4. Dialogue Management

Chapter 5. Introspective Mechanism
Chapter 6. Adaptable Dialogue Management
Chapter 7. Evaluation
Chapter 8. Further Applications

Ontologies and Adaptivity in Dialogue for Question Answering contains over 70 illustrations, 20 of them are in full-colour.

• Information Retrieval experts who plan to integrate semantics into databases and work on interfaces for semantic multimedia retrieval.

• Machine Learning experts who seek practical solutions to practical problems of model generation, such as the lack of supervised training material and the integration of learned models into a (dialogue) manager application. The learning aspect of a complex AI system is shown by the exploitation of the Semantic Web knowledge structure (i.e., ontological queries/answers and dialogue abox structures).

• Dialogue System experts interested in new forms of dialogue adaptivity and the formulation of meta dialogue; we also provide dialogue integration examples showing (1) how to predict answer times, (2) how to provide question feedback based on (ontological) metadata, and (3) how to learn to present incremental results from different (semantic) answer streams.

• Semantic Web experts interested in the potentials of answering engines that combine information from knowledge servers, Web Services, and open-domain QA. Meta dialogue allows us to mitigate the negative effect of different quality characteristics. We also address answer merging and result provenance aspects from a dialogue engineer’s point of view. The learning aspect of a complex AI system, i.e., an intelligent, dialogue-based user interface, is achieved by exploiting the Semantic Web knowledge structure with a combination of the abovementioned fields. In order to better understand the relationships among the scientific fields, the basics of natural language processing, ontologies, and machine learning are introduced as far as needed.

Ontologies and Adaptivity in Dialogue for Question Answering Daniel Sonntag January 2010, xii, 410 pp., Hardcover, ISBN 978-3-89838-623-4 (AKA) ISBN 978-1-60750-054-4 (IOS Press) Printed in Germany. (bibtex)	The appeal of being able to ask a question to a mobile internet terminal and receive an answer immediately has been renewed by the broad availability of information on the Web. Ideally, a spoken dialogue system that uses the Web as its knowledge base would be able to answer a broad range of questions. A new generation of natural language dialogue systems is emerging that transforms traditional keyword search engines into semantic answering machines by providing exact and concise answers formulated in natural language instead of today’s long lists of document references, which the user has to check by himself for relevant answers. This book presents the anatomy of the fully operational SmartWeb system (funded by the German Federal Ministry of Education and Research with grants totaling 14 million euros) that provides not only an open-domain question answering machine but a multimodal web service interface for coherent dialogue, where questions and commands are interpreted according to the context of the previous conversation. One of the key innovations described in this book is the ability of the system to learn how to predict the probability that it can answer a complex user query in a given time interval. Daniel Sonntag’s work within the SmartWeb project has laid important foundations for Theseus’s efforts towards semantic web technologies for the Web 3.0. Theseus is the German flagship project on the Internet of Services, where the user can delegate complex tasks to dynamically composed semantic web services by utilizing multimodal interaction combining speech and multi-touch input on advanced smartphones. Wolfgang Wahlster and Randy Goebel
Additional Material The document SmartWeb Handheld Interaction (V1.1) (updated December 2009), General Interactions and Result Display for User-System Multmodal Dialogue, provides the interaction storyboard (the multimodal design of SmartWeb handheld's interaction), and a description of the actual implementation. A demo of SmartWeb's mobile interaction design and evaluation questionnaires is also available. The industrial prototype system Radspeech, which builds upon this dialogue technology, won a German High Tech Award in 2011.	The book may be purchased here: Publisher's Book Website The book on amazon.com Studies on the Semantic Web Question answering (QA) has become one of the fastest growing topics in computational linguistics and information access. To advance research in the area of dialogue-based question answering, we propose a combination of methods from different scientific fields (i.e., information retrieval, dialogue systems, semantic web and machine learning). This book sheds light on adaptable dialogue-based question answering. We demonstrate the technical and computational feasibility of the proposed ideas, the introspective methods in particular, by beginning with an extensive introduction to the dialogical problem domain which motivates the technical implementation. The ideas have been carried out in a mature natural language processing (NLP) system, the SmartWeb dialogue system, which was developed between 2004 and 2007 by partners from academia and industry. We have attempted to make this book a self-containing text and provide an extra section on the interdisciplinary scientific background. The target audience for this book comprises of researchers and students interested in the application potential of semantic technologies for difficult AI tasks such as working dialogue and QA systems.
Table of Contents Preface by Wolfgang Wahlster and Randy Goebel Prologue (PDF) Part I. Introduction and Scientific Background Chapter 1. Introduction (PDF) Chapter 2. Scientific Background Part II. Ontologies and Dialogue-Based QA Chapter 3. Ontology-Based Dialogue Processing Chapter 4. Dialogue Management Part III. Introspection and Dialogue Adaptation Chapter 5. Introspective Mechanism Chapter 6. Adaptable Dialogue Management Chapter 7. Evaluation Chapter 8. Further Applications Epilogue Bibliography Detailed table of contents (PDF)	Ontologies and Adaptivity in Dialogue for Question Answering contains over 70 illustrations, 20 of them are in full-colour.
	The bibliography is publicly available here for research and education purposes. .
Who should read this book? Those interested in interdisciplinary AI research and the application potential of semantic technologies for difficult AI tasks such as working dialogue and QA systems should read this book. This book should interest researchers and professionals, i.e., • Information Retrieval experts who plan to integrate semantics into databases and work on interfaces for semantic multimedia retrieval. • Machine Learning experts who seek practical solutions to practical problems of model generation, such as the lack of supervised training material and the integration of learned models into a (dialogue) manager application. The learning aspect of a complex AI system is shown by the exploitation of the Semantic Web knowledge structure (i.e., ontological queries/answers and dialogue abox structures). • Dialogue System experts interested in new forms of dialogue adaptivity and the formulation of meta dialogue; we also provide dialogue integration examples showing (1) how to predict answer times, (2) how to provide question feedback based on (ontological) metadata, and (3) how to learn to present incremental results from different (semantic) answer streams. • Semantic Web experts interested in the potentials of answering engines that combine information from knowledge servers, Web Services, and open-domain QA. Meta dialogue allows us to mitigate the negative effect of different quality characteristics. We also address answer merging and result provenance aspects from a dialogue engineer’s point of view. The learning aspect of a complex AI system, i.e., an intelligent, dialogue-based user interface, is achieved by exploiting the Semantic Web knowledge structure with a combination of the abovementioned fields. In order to better understand the relationships among the scientific fields, the basics of natural language processing, ontologies, and machine learning are introduced as far as needed. The reader can learn a great deal about the fields by just reading through and/or consulting the references. An extensive bibliography is included to allow for further study especially in the case when the provided ideas are too detailed for the uninformed reader. Therefore, this book can be used by research scientists, as well as by students and practitioners who are particularly interested in the interdisciplinary nature of the subject matter, the application potentials of semantic technologies, and introspective mechanisms. The material is also suitable for a two-semester graduate course.
Web Counter