A method for WSD on Unrestricted Text

1
A method for WSD on Unrestricted Text

• Authors Rada Mihalcea and Dan Moldovan
• Presenter Marian Olteanu

2
Introduction

• WSD methods
• Information in MRD (machine readable
  dictionaries)
• Supervised training (info from a disambiguated
  corpus)
• Unsupervised training (info from a raw corpus)
• Hybrid methods

3
Approach

• Unsupervised learning
• Tag all content words (nouns, verbs, adjectives,
  adverbs)
• Use Web as a corpus (Altavista search engine)
• Use semantic density (using WordNet)

4
Algorithm

• Use word pairs (one word in the context of the
  other)
• Verb-noun pairs (syntactically linked)
• I.e. investigate report
• report1, study, report2, news report, story,
  account, write up

5
Algorithm (cont.)

• Search for investigate report and investigate
  study first sense
• Search for investigate report, investigate
  news report, , investigate write up second
  sense
• Order sense by counts

6
Algorithm (cont.)

• Repeat for verbs
• Use both phrases and NEAR operator similar
  results
• Select first 4 senses for N and V, first 2 for J
  and R

7
Algorithm step 2

• Compute conceptual density
• Apply only for N-V pair (because WN doesnt have
  adequate hierarchies for J and R)
• Between senses found at step 1
• Count match between nouns in the sub-glosses of
  the verb and all the hyponyms (noun) for the noun

8
Algorithm step 2 (cont.)

• Formula
• I find it flawed (log part)
• revise law

9
Evaluation

• SemCor
• Step 1
• Step 2

10
Comparison