ACL RD-TEC 1.0 Summarization of P99-1004
Paper Title:
MEASURES OF DISTRIBUTIONAL SIMILARITY
MEASURES OF DISTRIBUTIONAL SIMILARITY
Primarily assigned technology terms:
- approximation
- backoff smoothing
- classification
- clustering
- disambiguation
- distance-weighted averaging
- error rate reduction
- frequency-controlled pseudoword disambiguation
- information retrieval
- interpolation method
- jaccard coefficient
- language modeling
- language processing
- modeling
- natural language processing
- nearest neighbors
- nlp
- predictor
- probability estimation
- processing
- processing tools
- ranking
- rate reduction
- smoothing
- smoothing method
- statistical methods
- supervised disambiguation
- weighting
Other assigned terms:
- approach
- average error rate
- backoff
- case
- coefficient
- community
- concreteness
- conditional probabilities
- confusion probability
- correlation
- decision rule
- dice
- disambiguation task
- distribution
- distributional similarity
- empirical results
- error rate
- estimation
- euclidean distance
- evaluation methodology
- events
- fact
- feature
- finite set
- formalization
- head noun
- information sources
- interpolation
- jensen-shannon divergence
- joint distribution
- kl divergence
- language model
- measure
- measures
- method
- methodology
- mutual information
- natural language
- norm
- notational simplicity
- nouns
- precision
- priori
- probabilities
- probability
- probability distributions
- probability estimate
- retrieval performance
- schema
- semantic
- semantic similarity
- similarity function
- similarity measure
- similarity measures
- similarity metric
- similarity metrics
- skew divergence
- sparse data
- statistic
- substitutability
- synonyms
- technologies
- term
- terms
- tokens
- training
- training corpus
- training data
- training set
- transformation
- transitive verbs
- translations
- unigram
- user
- verb
- word
- word pair
- words