Deep learning approaches to text production / Shashi Narayan, Claire Gardent.Material type: TextSeries: Synthesis digital library of engineering and computer science | Synthesis lectures on human language technologies ; #44.Publisher: [San Rafael, California] : Morgan & Claypool, Description: 1 PDF (xxiv, 175 pages) : illustrations (some color)Content type: text Media type: electronic Carrier type: online resourceISBN: 9781681737591; 9781681738215Subject(s): Text processing (Computer science) | Neural networks (Computer science) | Machine learning | Artificial intelligence | copy | coverage | AMR generation | RDF generation | verbalise | simplification | compression | paraphrasing | dialogue generation | summarisation | content selection | adequacy | input understanding | sentence representation | document representation | communication goals | deep generators | reinforcement learning | evaluation | grammatical | fluent | meaning-preserving | BLEU | ROUGE | relevant | coherent | text production | text generation | deep learning | neural networks | meaning-to-text | data-to-text | text-to-text | recurrent neural networks | sequence-to-sequence models | attentionGenre/Form: Electronic books.Additional physical formats: Print version:: No titleDDC classification: 006.3/5 LOC classification: QA76.9.T48 | N374 2020ebOnline resources: Abstract with links to resource | Abstract with links to full text Also available in print.
|Item type||Current library||Call number||Status||Date due||Barcode|
|Ebooks||Indian Institute of Technology Delhi - Central Library||Available|
Mode of access: World Wide Web.
System requirements: Adobe Acrobat Reader.
Part of: Synthesis digital library of engineering and computer science.
Includes bibliographical references (pages 139-173).
1. Introduction -- 1.1. What is text production? -- 1.2. Roadmap -- 1.3. What's not covered? -- 1.4. Our notations
part I. Basics. 2. Pre-neural approaches -- 2.1. Data-to-text generation -- 2.2. Meaning representations-to-text generation -- 2.3. Text-to-text generation -- 2.4. Summary
3. Deep learning frameworks -- 3.1. Basics -- 3.2. The encoder-decoder framework -- 3.3. Differences with pre-neural text-production approaches -- 3.4. Summary
part II. Neural improvements. 4. Generating better text -- 4.1. Attention -- 4.2. Copy -- 4.3. Coverage -- 4.4. Summary
5. Building better input representations -- 5.1. Pitfalls of modelling input as a sequence of tokens -- 5.2. Modelling text structures -- 5.3. Modelling graph structure -- 5.4. Summary
6. Modelling task-specific communication goals -- 6.1. Task-specific knowledge for content selection -- 6.2. Optimising task-specific evaluation metric with reinforcement learning -- 6.3. User modelling in neural conversational model -- 6.4. Summary
part III. Data sets and conclusion. 7. Data sets and challenges -- 7.1. Data sets for data-to-text generation -- 7.2. Data sets for meaning representations to text generation -- 7.3. Data sets for text-to-text generation
8. Conclusion -- 8.1. Summarising -- 8.2. Overview of covered neural generators -- 8.3. Two key issues with neural NLG -- 8.4. Challenges -- 8.5. Recent trends in neural NLG.
Abstract freely available; full-text restricted to subscribers or individual document purchasers.
Google book search
Text production has many applications. It is used, for instance, to generate dialogue turns from dialogue moves, verbalise the content of knowledge bases, or generate English sentences from rich linguistic representations, such as dependency trees or abstract meaning representations. Text production is also at work in text-to-text transformations such as sentence compression, sentence fusion, paraphrasing, sentence (or text) simplification, and text summarisation. This book offers an overview of the fundamentals of neural models for text production. In particular, we elaborate on three main aspects of neural approaches to text production: how sequential decoders learn to generate adequate text, how encoders learn to produce better input representations, and how neural generators account for task-specific objectives. Indeed, each text-production task raises a slightly different challenge (e.g, how to take the dialogue context into account when producing a dialogue turn, how to detect and merge relevant information when summarising a text, or how to produce a well-formed text that correctly captures the information contained in some input data in the case of data-to-text generation). We outline the constraints specific to some of these tasks and examine how existing neural models account for them. More generally, this book considers text-to-text, meaning-to-text, and data-to-text transformations. It aims to provide the audience with a basic knowledge of neural approaches to text production and a roadmap to get them started with the related work. The book is mainly targeted at researchers, graduate students, and industrials interested in text production from different forms of inputs.
Also available in print.
Title from PDF title page (viewed on April 6, 2020).