Semantic mining: Unsupervised acquisition of multilingual semantic - - PowerPoint PPT Presentation

Date

Presenter: Zheng ZHANG Supervisors: Pierre ZWEIGENBAUM & Yue MA Research group: ILES Doctoral school: STIC Funding: CSC

Semantic mining: Unsupervised acquisition of multilingual semantic classes from texts

Journée des Doctorants du LIMSI 30/05/2017

Distributional hypothesis

• Distributional hypothesis: Words that occur in the same

contexts tend to have similar meanings (Harris, 1954).

• Take a word and its contexts:
• pomme rouge
• pomme délicieuse
• …
• By looking at a word’s context, one can infer its meaning

1

Distributional similarity

• Distributional hypothesis: Words that occur in the same

contexts tend to have similar meanings (Harris, 1954).

• Take a word and its contexts:
• pomme rouge
• pomme délicieuse
• …
• By looking at a word’s context, one can infer its meaning

1

Multilingual semantic classes

fruits

couleurs pomme poire pêche prunelle

2

• Semantic class: A group of words

clustered by using distributional similarity measures.

Multilingual semantic classes

fruits

couleurs pomme poire pêche prunelle

colors

fruits

apple pear peach FR EN

2

Multilingual semantic classes

fruits

couleurs pomme poire pêche prunelle

colors

fruits

apple pear peach FR EN

2

Applications: Unknown words “translation”

fruits

couleurs pomme poire pêche prunelle

colors

fruits

apple pear peach FR EN roux

3

Applications: Unknown words “translation”

fruits

couleurs pomme poire pêche prunelle

colors

fruits

apple pear peach FR EN roux

3

Applications: Universal classes extraction

fruits

pomme poire pêche prunelle

fruits

apple pear peach FR EN

Universal class

• f “fruits”

4

Merci