ACL RD-TEC 1.0 Summarization of C02-1004
Paper Title:
COMBINING UNSUPERVISED AND SUPERVISED METHODS FOR PP ATTACHMENT DISAMBIGUATION
COMBINING UNSUPERVISED AND SUPERVISED METHODS FOR PP ATTACHMENT DISAMBIGUATION
Primarily assigned technology terms:
- algorithm
- boundary recognition
- cascaded disambiguation
- chunking
- clause boundary recognition
- computer science
- computing
- cross validation
- database
- disambiguation
- disambiguation algorithm
- learner
- learning
- lemmatization
- maximum likelihood
- name recognition
- parsing
- part-of-speech tagging
- pp attachment disambiguation
- pp attachmentdisambiguation
- pp disambiguation
- processing
- proper name recognition
- recognition
- search
- search engines
- sentence recognition
- shallow parsing
- splitting
- statistical methods
- supervised back-off
- supervised learner
- supervised learning
- supervised training
- tagging
- unsupervised method
- unsupervised training
- validation
Other assigned terms:
- analogy
- annotation
- approach
- association score
- back-off model
- bigram
- boundary information
- case
- clause boundary
- collocation
- corpora
- corpus size
- cross validation experiment
- distribution
- fact
- information sources
- large corpora
- lexical association
- likelihood
- measures
- method
- name class
- names
- negra
- negra corpus
- nouns
- occurrence frequency
- part-of-speech
- part-of-speech tags
- penn treebank
- phrase
- pp attachment
- preposition
- prepositional phrase
- prepositions
- probabilities
- probability
- probability estimates
- pronoun
- pronouns
- proper name
- proper names
- raw text corpora
- relative frequency
- sentence
- sentences
- sparse data
- sparse data problem
- tags
- test corpus
- test set
- text
- text corpora
- thesaurus
- tokens
- training
- training corpus
- training data
- training material
- training set
- treebank
- uniform distribution
- unigram
- verb
- verb attachment
- word
- word pair
- wordnet
- wordnet thesaurus
- words
