ACL RD-TEC 1.0 Summarization of W06-1643
Paper Title:
A SKIP-CHAIN CONDITIONAL RANDOM FIELD FOR RANKING MEETING UTTERANCES BY IMPORTANCE
A SKIP-CHAIN CONDITIONAL RANDOM FIELD FOR RANKING MEETING UTTERANCES BY IMPORTANCE
Primarily assigned technology terms:
- 5-fold cross validation
- algorithm
- approximation
- artificial intelligence
- asr transcription
- automated evaluation
- automatic speech recognition
- classifier
- computational linguistics
- computing
- conditional random field
- conditional random fields
- content selection
- crfs
- cross validation
- cross-validating
- cross-validation
- decoding
- decomposition
- discriminative framework
- document understanding
- dynamic bayesian networks
- email summarization
- exact inference
- extractive summarization
- feature selection
- forward-backward algorithm
- gibbs sampling
- graphical model
- greedy search
- hidden markov
- hidden markov models
- information retrieval
- l-bfgs optimization
- language processing
- latent semantic analysis
- linking
- local normalization
- markov model
- matching
- meeting summarization
- modeling
- multi-document summarization
- natural language processing
- networking
- nlp
- non-sequential classifier
- normalization
- optimization
- parameter estimation
- probabilistic inference
- processing
- pyramid evaluation
- ranking
- recognition
- recognition system
- recognizer
- sampling
- scoring
- search
- semantic analysis
- singular value decomposition
- speech recognition
- speech recognition system
- summarization
- summarization evaluation
- summarizer
- tagging
- terminology
- text summarization
- transcription
- utterance selection
- validation
- viterbi
- viterbi decoding
Other assigned terms:
- acoustic models
- adjacency pair
- annotation
- annotators
- approach
- association for computational linguistics
- base noun
- case
- compression ratio
- content evaluation
- conversational speech
- conversational telephone speech
- correlation
- crf model
- decision rule
- dialog
- dialog act
- dialogues
- discourse
- discourse relations
- distribution
- document
- duration
- empirical results
- estimation
- evaluation method
- evaluation metric
- evaluations
- f-measure
- fact
- feature
- feature set
- genre
- human performance
- hypothesis
- icsi meeting corpus
- index
- intelligence
- interpretation
- language processing applications
- latent semantic
- length constraint
- lexical features
- linguistics
- log-likelihood
- log-linear model
- log-linear models
- markov models
- meaning
- meanings
- measure
- measures
- meeting corpus
- method
- model parameters
- mutual information
- n-grams
- natural language
- natural language processing applications
- nlp tasks
- noun phrases
- paraphrases
- posterior
- posterior probability
- precision
- predicates
- prior probability
- probabilistic approach
- probabilities
- probability
- probability distribution
- probability distributions
- procedure
- representations
- semantic
- sentence
- sentence importance
- sentences
- sequence model
- statistics
- style
- summarization task
- tags
- technique
- term
- terms
- test data
- test set
- text
- tf \* idf
- topics
- training
- training data
- transcriptions
- tree
- understanding
- utterance
- viterbi path
- word
- words
