Overview
Save Big on Coursera Plus. 7,000+ courses at $160 off. Limited Time Only!
Explore abstract meaning representation (AMR) in natural language processing through this 44-minute conference talk from ODSC Europe 2019. Delve into the history of text meaning representation and learn about AMR graphs, their construction algorithms, and potential applications in question answering systems, text summarization, and simplification. Examine various approaches to representing text meaning, including bag of words, syntactic parsing, and semantic parsing. Understand AMR notation, examples of frames and modality, and parsing techniques such as graph-based and transition-based methods. Discover how AMR is evaluated using smatch and its applications in natural language generation and abstractive text summarization. Gain insights into the pros and cons of AMR and its potential to advance NLP research and applications.
Syllabus
Intro
Option 1: bag of words
syntactic parsing
semantic parsing
Lambda calculus
Semantic role labelling: PropBank
Semantic role labelling: AllenNLP
Abstract meaning representation
What is AMR?
AMR pros and cons
AMR notation
AMR examples: frames
AMR examples: modality
AMR parsing
Graph-based parsing
Transition-based parsing
AMR evaluation: smatch
Application of AMR
Natural Language Generation: example
Abstractive text summarization
Taught by
Open Data Science