ACL RD-TEC 1.0 Summarization of N03-1003
Paper Title:
LEARNING TO PARAPHRASE: AN UNSUPERVISED APPROACH USING MULTIPLE-SEQUENCE ALIGNMENT
LEARNING TO PARAPHRASE: AN UNSUPERVISED APPROACH USING MULTIPLE-SEQUENCE ALIGNMENT
Authors: Regina Barzilay and Lillian Lee
Primarily assigned technology terms:
- algorithm
- blind evaluation
- clustering
- complete-link clustering
- computational biology
- extraction application
- extraction system
- induction
- information extraction
- information extraction system
- learning
- learning algorithm
- matching
- measuring
- multiple-sequence alignment
- nlp
- pairwise alignment
- paraphrase acquisition
- paraphrase extraction
- paraphrasing
- parsing
- pattern induction
- pattern-matching
- question answering
- reading
- scoring
- scoring function
- sentence clustering
- sentence-alignment
- summarization
- template-induction
- text-to-text generation
- unsupervised approach
- unsupervised paraphrase acquisition
- word ordering
Other assigned terms:
- approach
- biology
- case
- cluster
- clusters
- comparable corpora
- comparable corpus
- corpora
- data sets
- dependency trees
- distributional similarity
- evaluations
- events
- fact
- generation
- inference rules
- kappa
- kappa statistic
- kappa value
- language model
- large corpus
- lattice
- lattices
- meaning
- measure
- method
- mutual information
- n-gram
- names
- non-parallel corpora
- opinions
- parallel corpus
- parameter values
- paraphrase
- paraphrase corpus
- paraphrases
- pigeonhole principle
- precision
- procedure
- process
- proper name
- proper names
- propositions
- random order
- random sample
- reuters corpus
- sentence
- sentence level
- sentences
- similarity between words
- similarity measure
- similarity metric
- slot
- statistic
- synonym
- synonyms
- terms
- test set
- text
- tokens
- training
- training corpus
- training set
- transformation
- trees
- understanding
- word
- word lattice
- word lattices
- wordnet
- words
