Word Sense Disambiguation

Question 1

Where can we get a list of all the different senses of a word

Answer 1

wordnet

Question 2

what is a ‘supersense’

Answer 2

gives a lexicographic category based on word meaning e.g. location

Question 3

what is a ‘sense relation’

Answer 3

describes things like ‘part of’, or subbordinate

Question 4

What is the word sense disambiguation task

Answer 4

given a word in its context and lots of different possible word senses

output the correct word sense given the context

Question 5

describe the hypernym relationships

Answer 5

from concepts to superordinates

breakfast->meal

Question 6

describe the hyponym relationship

Answer 6

from concepts to subtypes

meal->lunch

Question 7

describe the instance hypernym relationship

Answer 7

from instances to their concepts

jane austen->author

Question 8

describe the instance hyponym relationship

Answer 8

from concepts to their instances

composer->bach

Question 9

describe the part meronym relationship

Answer 9

has part
from wholes to parts
table->leg

Question 10

describe the part holonym relationship

Answer 10

part of
from parts to wholes
course->meal

Question 11

describe the antonym relationship

Answer 11

semantic opposition between lemmas

leader->follower

Question 12

describe the derivation relationship

Answer 12

lemmas with the same morphological root

destruction -> destroy

Question 13

what are the two possible tasks to tackle word sense disambiguation

Answer 13

• the lexical simple task- the sense of just one word

- all words task- the sense of every word in the sentence

Question 14

what is the lesk algorithm

Answer 14

a simple but good baseline approach

• choose the word whose dictionary or definition shares the most words with the target words neighbourhood

Question 15

what are the benefits of the lesk algorithm (2)

Answer 15

simple

no training data

Question 16

what are the disadvantages of the lesk algorithm (3)

Answer 16

not all words have word net definitions
ambiguous context words
poor performance compared to other methods

Question 17

what is semcor

Answer 17

the largest corpus with word net labelled senses

Question 18

how can we use supervised learning for all word WSD

Answer 18

tag each word with its word net senses
POS tags
collocation features of words and n-grams
weighted average of word embeddings

^ as features into a svm