Natural Language Processing in Artificial Intelligence―NLPinAI 2020 (Studies in Computational Intelligence, 939)

✍ Scribed by Roussanka Loukanova (editor)

• Publisher: Springer
• Year: 2021
• Tongue: English
• Leaves: 250
• Category: Library

✦ Synopsis

This book covers theoretical work, applications, approaches, and techniques for computational models of information and its presentation by language (artificial, human, or natural in other ways). Computational and technological developments that incorporate natural language are proliferating. Adequate coverage encounters difficult problems related to ambiguities and dependency on context and agents (humans or computational systems). The goal is to promote computational systems of intelligent natural language processing and related models of computation, language, thought, mental states, reasoning, and other cognitive processes.

✦ Table of Contents

Preface
Contents
Lambek Calculus with Classical Logic
1 Introduction
1.1 Overview
1.2 Categorial Grammars
2 Algebras and Frames
3 H-Systems and Modal Logics
3.1 Unary Modalities
3.2 Binary Modalities
3.3 Other Modal Axioms
3.4 Filtration
4 Decidability and Complexity
5 Conclusion
References
Partial Orders, Residuation, and First-Order Linear Logic
1 Introduction
2 Categorial Grammars and Residuation
2.1 Residuation in Extended Lambek Calculi
2.2 Residuation and First-Order Linear Logic
3 First-Order Linear Logic
3.1 First-Order Linear Logic and Natural Language Grammars
3.2 Proof Nets
4 Residuation and Partial Orders
4.1 Residuation for the Lambek Calculus Revisited
4.2 Partial Orders
5 The General Case
5.1 How Many Residuated Connectives Are There for Concatenation-Like Operations?
5.2 Well-Nestedness
5.3 Partial Order Constraints in Practice
6 The Empty String
7 Discussion
8 Conclusions
References
A Hyperintensional Theory of Intelligent Question Answering in TIL
1 Introduction
2 Foundations of TIL
3 Natural Deduction in TIL
4 Semantic Rules
4.1 Factive Attitudes and Presuppositions
4.2 Property Modifiers
4.3 Anaphoric References and Substitution Method
5 Wh-Questions
6 Two Case Studies
6.1 Reasoning with Property Modifiers
6.2 Reasoning with Factive Propositional Attitudes
7 Conclusion
References
Learning Domain-Specific Grammars from a Small Number of Examples
1 Introduction
1.1 Use Case: Language Learning
1.2 Use Case: Interlingual Communication
2 Background
2.1 Previous Work on Grammar Learning
2.2 Abstract Grammars
2.3 Wide-Coverage and Resource Grammars
2.4 Constraint Satisfaction Problems
3 Learning a Subgrammar
3.1 Subgrammar Extraction by Tree Selection
3.2 Modeling Subgrammar Extraction as a Constraint Problem
4 Bilingual Grammar Learning
5 Implementation
6 Evaluation
6.1 Rebuilding a Known Grammar
6.2 Comparing Against a Treebank
6.3 Comparing Against a Bilingual Treebank
7 Results
7.1 Results: Rebuilding a Known Grammar
7.2 Results: Comparing Against a Treebank
7.3 Results: Using Bilingual Treebanks
8 Extension 1: Negative Examples
8.1 Examples
8.2 Iterative Grammar Learning Process
9 Extension 2: Extracting Subtrees as Basic Units
9.1 Handling Combinatorial Explosion
9.2 Examples
10 Discussion
11 Conclusion
References
The Semantic Level of Shannon Information: Are Highly Informative Words Good Keywords? A Study on German
1 Introduction
1.1 General Application of Shannon's Information Theory in Language
1.2 Semantic Aspects in Shannon's Information Theory
1.3 Application of Shannon's Information Theory to Our Study
2 Related Work
3 Dataset
3.1 What Are Keywords?
4 Method
4.1 Baseline
4.2 Information Theory Based Methods
4.3 Neural Network
4.4 Evaluation Method
5 Results
6 Conclusion and Discussion
References
Towards Aspect Extraction and Classification for Opinion Mining with Deep Sequence Networks
1 Introduction
1.1 Contribution
1.2 Domain Focus
2 State of Research
3 New Datasets for Aspect Phrases in Physician Reviews
3.1 Data Collection and Overview
3.2 Rating Classes
3.3 Annotation Process
4 Neural Networks for Aspect Phrase Extraction and Classification
5 Evaluation and Discussion
6 Conclusion
References
Dialogical Argumentation and Textual Entailment
1 Presentation: Argumentation, Inference, Semantics
2 Dialogical Logic
2.1 Language, Formulas, Subformulas, Trees
2.2 Argumentation Forms
2.3 Prejustified Sequences
2.4 Games
2.5 Strategies
2.6 Validity
3 Dialogical Validity Is Equivalent to Classical Validity
3.1 From Strategies to Proofs in GKs
3.2 From Proofs in GKs to Strategies
4 Categorical Grammars and Automated Theorem Proving
5 Textual Entailment
5.1 First Example
5.2 Second Example
5.3 Third Example
5.4 Fourth Example
6 Conclusion
References
A Novel Approach to Determining the Quality of News Headlines
1 Introduction
2 Related Work
3 Labelling Data
3.1 Data
3.2 Quality Indicators
4 Predict Headline Quality
4.1 Problem Definition
4.2 Proposed Model
5 Evaluation
5.1 Baselines
5.2 Evaluation Metrics
5.3 Experimental Results
6 Conclusions
References