Natural Language Processing · free preview
Dependency Parsing
Head-dependent relations and Universal Dependencies, projectivity, transition-based parsing with an oracle, graph-based parsing as maximum spanning tree, and attachment scores.
1. Dependency Parsing
Dependency Parsing
2. Two families of parsers
Two families of parsers
3. Arc-standard parsing and UAS in Python
Arc-standard parsing and UAS in Python
4. What makes a valid dependency tree
Which set of constraints must a dependency structure satisfy to be a rooted tree over the words?
5. Naming the relation
In 'the ferry docks at the pier', which Universal Dependencies relations are correct for the arcs from docks to ferry and from at to pier?
6. Choosing an operator
The stack is [ROOT, ferry, docks] and the buffer is [at, the, pier], with ferry's head being docks. What should the arc-standard oracle do?
7. Computing UAS
A dependency parser processes a -word sentence and assigns the correct head to words. Compute the unlabeled attachment score, rounding to three decimals.
8. Two families, two strengths
Which statement correctly contrasts transition-based and graph-based dependency parsers?
Practice this interactively
Free account · instant grading · spaced review that schedules itself.
Start this course for free