ACL RD-TEC 1.0 Summarization of C04-1167
Paper Title:
STATISTICAL LANGUAGE MODELING WITH PERFORMANCE BENCHMARKS USING VARIOUS LEVELS OF SYNTACTIC-SEMANTIC INFORMATION
STATISTICAL LANGUAGE MODELING WITH PERFORMANCE BENCHMARKS USING VARIOUS LEVELS OF SYNTACTIC-SEMANTIC INFORMATION
Authors: Dharmendra Kanejiya and Arun Kumar and Surendra Prasad
Primarily assigned technology terms:
- algorithm
- analysis method
- cognitive modeling
- decomposition
- dimensionality reduction
- factorization
- geometric interpolation
- grouping
- interpolation method
- knowledge representation
- language modeling
- language understanding
- latent semantic analysis
- length normalization
- likelihood estimation
- matrix factorization
- maximum likelihood
- maximum likelihood estimation
- modeling
- natural language understanding
- normalization
- parsing
- recognition
- recognition system
- rescoring
- secondpass rescoring
- semantic analysis
- singular value decomposition
- smoothing
- smoothing techniques
- speech recognition
- speech recognition system
- statistical language modeling
- syntactic analysis
- tagger
- tagging
- vector representation
- verb phrases
- word prediction
Other assigned terms:
- adjective
- adjunction
- anchor
- approach
- benchmark
- case
- co-occurrence
- co-occurrence statistics
- cognitive
- conditional probability
- content words
- corpora
- correlation
- cosine measure
- dimensionality
- document
- document length
- entropy
- error rate
- estimation
- experimental results
- frequency counts
- function word
- function words
- geometric mean
- grammars
- hypothesis
- information space
- interpolation
- joint probability
- knowledge
- language model
- language model probability
- language models
- latent semantic
- latent semantic space
- lexical item
- lexicalized tree
- likelihood
- likelihood probability
- linguistic
- linguistic structure
- local context
- long distance dependencies
- lsa matrix
- ltags
- measure
- method
- model probability
- modeling problem
- n-best list
- n-gram
- n-gram language model
- n-gram model
- n-gram models
- n-grams
- natural language
- noun phrase
- noun phrases
- nouns
- paragraph
- parse
- parse-tree
- part-of-speech
- part-of-speech tag
- part-of-speech tags
- parts of speech
- perplexity
- perplexity measure
- phrase
- phrase level
- phrase type
- predicate-argument
- predictive power
- prepositions
- probabilities
- probability
- probability distributions
- procedure
- process
- projection
- r-dimensional space
- recognition accuracy
- recognition task
- selsa language model
- semantic
- semantic description
- semantic information
- semantic similarity
- semantic similarity measure
- semantic space
- sentence
- sentences
- similarity measure
- speech recognition accuracy
- speech recognition task
- statistical language model
- statistics
- subject noun phrase
- supertag
- supertags
- syntactic context
- syntactic description
- syntactic information
- syntactic knowledge
- syntactic structure
- syntax
- tag information
- tag sequence
- tagged corpus
- tags
- tagset
- target word
- technique
- terms
- test corpus
- test data
- text
- text corpus
- text documents
- training
- training corpus
- tree
- tree adjoining grammars
- understanding
- verb
- verb phrase
- vocabulary
- vocabulary size
- word
- word error rate
- word frequencies
- word sequence
- word sequences
- word type
- words
- wsj corpus
