ACL RD-TEC 1.0 Summarization of J99-3003
Paper Title:
VECTOR-BASED NATURAL LANGUAGE CALL ROUTING
VECTOR-BASED NATURAL LANGUAGE CALL ROUTING
Authors: Jennifer Chu-Carroll and Bob Carpenter
Primarily assigned technology terms:
- algorithm
- automatic speech recognizer
- call classification
- call routing
- categorization
- category disambiguation
- character recognition
- classification
- classifier
- clustering
- computational linguistics
- computing
- corpus analysis
- database
- decomposition
- dialogue evaluation
- dimensionality reduction
- disambiguation
- disambiguation process
- document construction
- error reduction
- indexing
- information retrieval
- information retrieval system
- interactive dialogue
- latent semantic analysis
- latent semantic indexing
- learning
- lexical clustering
- logistic regression
- matching
- maximum likelihood
- morphological processor
- n-gram extraction
- neural network
- noisy speech recognizer
- normalization
- noun phrase disambiguation
- optical character recognition
- phrase disambiguation
- phrasing
- processing
- processor
- query formulation
- querying
- reasoning
- recognition
- recognizer
- regression
- response generation
- retrieval system
- selection process
- semantic analysis
- semantic indexing
- singular value decomposition
- speech recognition
- speech recognizer
- synthesizer
- template filling
- term extraction
- term selection
- text categorization
- text-to-speech
- text-to-speech synthesizer
- training process
- transcription
- vector representation
- vector-based information retrieval
- verb phrases
- weighting
- word filtering
Other assigned terms:
- ambiguity
- approach
- asr output
- association for computational linguistics
- bigram
- call center
- candidate terms
- case
- cluster
- co-occurrence
- community
- confidence score
- confidence scores
- content words
- conversational language
- cosine measure
- data sets
- data sparseness
- dialogue behavior
- dialogue strategies
- dialogues
- dimensionality
- distribution
- document
- document vector
- document vectors
- entropy
- error rate
- error reduction rate
- euclidean distance
- extraction process
- feature
- finite set
- finite-state language model
- frequency counts
- function words
- generation
- generation process
- head noun
- human operator
- human-human dialogue
- hypotheses
- hypothesis
- implementation
- inflected forms
- information retrieval community
- kappa
- kappa statistic
- language model
- latent semantic
- likelihood
- linguistics
- mapping
- mappings
- maps
- meaning
- measure
- measures
- mechanisms
- method
- n-best list
- n-gram
- n-grams
- names
- natural language
- ngram
- noise
- noisy speech
- noun phrase
- noun phrases
- performance comparison
- perplexity
- phrase
- precision
- prior distribution
- probabilities
- probability
- probability distribution
- process
- proper names
- queries
- query
- query vector
- r-dimensional space
- recognition accuracy
- recognition errors
- representations
- run-time
- semantic
- semantic networks
- semantic types
- sentence
- sigmoid function
- similarity measure
- speech data
- speech input
- speech recognition accuracy
- speech recognition errors
- spoken input
- statistic
- statistics
- synonyms
- system design
- system performance
- technique
- technologies
- term
- term-document matrix
- terms
- test data
- test set
- testing corpus
- text
- topics
- training
- training corpus
- training data
- training set
- transcript
- transcriptions
- trigram
- underspecification
- understanding
- unigram
- user
- user utterance
- user utterances
- utterance
- vector space
- verb
- verb forms
- vocabulary
- weighting scheme
- wh-question
- word
- word co-occurrence
- word error rate
- word sequence
- words
