ACL RD-TEC 1.0 Summarization of P06-1024
Paper Title:
LEARNING MORE EFFECTIVE DIALOGUE STRATEGIES USING LIMITED DIALOGUE MOVE FEATURES
LEARNING MORE EFFECTIVE DIALOGUE STRATEGIES USING LIMITED DIALOGUE MOVE FEATURES
Authors: Matthew Frampton and Oliver Lemon
Primarily assigned technology terms:
- approximation
- cfs subset evaluation
- communicator
- computational linguistics
- database
- dialogue management
- dialogue manager
- dialogue system
- dialogue systems
- feature selection
- information seeking
- learner
- learning
- learning program
- learning technique
- linear function approximation
- modelling
- paradise evaluation
- reinforcement learning
- scoring
- scoring function
- spoken dialogue
- spoken dialogue system
- spoken dialogue systems
- supervised\/reinforcement learning
Other assigned terms:
- association for computational linguistics
- case
- communicator dataset
- context features
- context information
- contextual features
- contextual information
- data set
- database query
- dialogue behavior
- dialogue context
- dialogue history
- dialogue length
- dialogue move
- dialogue strategies
- dialogues
- experimental results
- fact
- feature
- feature sets
- generalisation
- human operator
- hypothesis
- information state
- information states
- knowledge
- linguistics
- maps
- method
- methodology
- n-gram
- n-gram models
- probabilities
- procedure
- qualitative analysis
- queries
- query
- representations
- restricted dialogue
- semantic
- slot
- speech acts
- technique
- term
- terms
- training
- training data
- udm strategy
- user
- user behavior
- user satisfaction
- user utterance
- user utterances
- utterance
- word
- word sequences
