This paper gives an overview of the current level of techniques used in syntactic parsing , focusing on parsing of human language. Different modes of grammatical representation and grammar types are presented as well as the different approaches to parsing (e.g. robust/shallow vs. integrative/probabilistic).
Table of Contents
1 Introduction
1.1 Introduction
1.2 Definitions
1.3 Why parse?
2 Input / data
2.1 What can be parsed?
2.1.1 Texts
2.1.2 Spoken language
2.2 Tagging
2.2.1 Tagging techniques
2.2.2 Output
3 Grammar
3.1 Representing syntax
3.1.1 Dependency Syntax
3.1.2 Constituent Structure Syntax
3.2 Grammar Types
3.2.1 Context Free Grammars
3.2.2 Probabilistic Context Free Grammars
4 Parsing
4.1 Algorithms
4.1.1 Direction of processing
4.1.2 Direction of analysis
4.1.3 Search strategy
4.1.4 Backtracking vs. Chart parsing
4.2 Approaches to parsing
4.2.1 Robust parsing
4.2.2 Shallow parsing
4.2.3 Integrative vs. sequential architectures
4.2.4 Probabilistic approaches
4.3 Ambiguity
4.3.1 Types of ambiguity
4.3.2 Disambiguation techniques
4.4 Evaluating parsing systems
4.4.1 Coverage
5 Conclusion
Objectives and Core Topics
The paper provides a fundamental overview of contemporary techniques in syntactic parsing, focusing on how language is processed, represented, and analyzed within computational systems. It explores the methodologies for managing grammatical structures, search algorithms, and the challenges of ambiguity in natural language processing.
- Methods for representing formal grammars and syntactic structures.
- The distinction between rule-based, stochastic, and robust parsing approaches.
- Algorithm design including search strategies and the use of charts versus backtracking.
- Strategies for handling lexical, structural, and semantic ambiguity.
- Evaluation metrics for assessing the performance and coverage of parsing systems.
Excerpt from the Book
1.2 Introduction
This paper intends to provide a brief introduction to the current techniques used in syntactic parsing. It will present different techniques used in representing grammars, conducting searches and resolving ambiguities. Different approaches used in robust parsing will be presented, and a brief look will be taken on the evaluation of parser performance.
Although sample projects are mentioned throughout the text, they will not be presented in full due to limitations of time and space.
Summary of Chapters
1 Introduction: Provides definitions for parsing and explains the necessity of syntactical analysis for modern computational applications.
2 Input / data: Examines the types of inputs that can be parsed, specifically differentiating between text processing and the more complex challenges of spoken language.
3 Grammar: Discusses the theoretical frameworks for representing syntax, comparing dependency and constituent structure, and introduces different grammar types like CFGs and PCFGs.
4 Parsing: Details the algorithmic foundations of parsing, including search strategies, approaches to handling ambiguity, and methods for evaluating system performance.
5 Conclusion: Summarizes the current state of parsing, emphasizing the need for integrative systems that combine stochastic and semantic information.
Keywords
Syntactic Parsing, Natural Language Processing, Context Free Grammars, PCFG, Tagging, Robust Parsing, Shallow Parsing, Ambiguity, Disambiguation, Speech Recognition, Treebanks, Earley Algorithm, Computational Linguistics, Syntax Analysis
Frequently Asked Questions
What is the core purpose of this document?
The paper serves as an introductory technical overview of the current methodologies and challenges in the field of syntactic parsing within computational linguistics.
What are the primary themes discussed?
The main themes include input processing, formal grammar representation, parsing algorithms, methods for managing linguistic ambiguity, and system evaluation.
What is the main objective of the author?
The goal is to provide a comprehensive summary of how machines effectively assign syntactical structure to natural language input using various representational and algorithmic models.
Which scientific methods are analyzed?
The author discusses various algorithmic approaches such as top-down and bottom-up analysis, depth-first and breadth-first search strategies, and the use of chart parsing versus backtracking.
What content is covered in the main section of the paper?
The main section covers the technical architecture of parsers, moving from the input level (tagging) through grammar definition to the actual parsing algorithms and strategies for handling data errors and ambiguities.
Which keywords define this work?
Key terms include Syntactic Parsing, Context Free Grammars, Robust Parsing, Ambiguity, and Treebanks.
What is the difference between a parser and a recognizer?
A parser performs a detailed syntactic annotation of a sentence, whereas a recognizer only makes a binary decision regarding whether a sentence conforms to a given grammar.
Why is parsing spoken language considered more difficult than parsing text?
Parsing speech involves handling additional noise, such as ungrammatical input, register variations, self-corrections, and false starts, which are not present in clean digital text.
What is the role of the Penn Treebank in modern parsing?
The Penn Treebank is used to provide large sets of annotated natural language data, allowing developers to calculate probabilities for rules and train stochastic parsing systems.
How does the author propose to improve future parsing systems?
The author suggests that future systems should move toward integrative architectures that incorporate semantic information alongside existing stochastic and contextual data.
- Quote paper
- Jan Niehues (Author), 2005, Current parsing techniques - an overview, Munich, GRIN Verlag, https://www.grin.com/document/52290
