ACL RD-TEC 1.0 Summarization of N04-3008
Paper Title:
SENSECLUSTERS - FINDING CLUSTERS THAT REPRESENT WORD SENSES
SENSECLUSTERS - FINDING CLUSTERS THAT REPRESENT WORD SENSES
Authors: Amruta Purandare and Ted Pedersen
Primarily assigned technology terms:
- agglomerative clustering
- classification
- clustering
- cutoff
- decomposition
- dimensionality reduction
- document classification
- feature representation
- feature selection
- grouping
- identification
- jaccard coefficient
- language processing
- latent semantic analysis
- link clustering
- matching
- maximal matching
- natural language processing
- ontology construction
- open source software
- processing
- semantic analysis
- sense discrimination
- singular value decomposition
- statistical test
- summarization
- text summarization
- tokenization
- unsupervised clustering
- vector representation
- visualization
- word sense discrimination
- word space
Other assigned terms:
- approach
- bigram
- case
- cluster
- clusters
- co-occurrence
- co-occurrences
- coefficient
- concept
- concepts
- confusion matrix
- context vector
- context vectors
- corpora
- dice
- dice coefficient
- dimensionality
- distribution
- document
- evaluation metrics
- feature
- feature space
- feature type
- feature types
- gnu public license
- gold standard
- hypothesis
- implementation
- index
- knowledge
- language processing tasks
- large corpora
- latent semantic
- lexical features
- log-likelihood
- log-likelihood ratio
- mapping
- maps
- meaning
- meanings
- measures
- mechanisms
- methodology
- mutual information
- natural language
- natural language processing tasks
- ngram
- ontology
- pairs of words
- pointwise mutual information
- polysemy
- precision
- process
- processing tasks
- regular expressions
- representations
- second order context
- semantic
- semantic space
- sentences
- similarity matrix
- similarity measures
- statistics
- svdpack
- synonyms
- synonymy
- tags
- target word
- terms
- test data
- text
- toolkit
- training
- training corpus
- training data
- tree
- unigram
- user
- vector space
- word
- word level
- word meaning
- word sense
- word senses
- words