ACL RD-TEC 1.0 Summarization of W03-1203
Paper Title:
COMBINING OPTIMAL CLUSTERING AND HIDDEN MARKOV MODELS FOR EXTRACTIVE SUMMARIZATION
COMBINING OPTIMAL CLUSTERING AND HIDDEN MARKOV MODELS FOR EXTRACTIVE SUMMARIZATION
Authors: Pascale Fung and Grace Ngai and Chi-Shun Cheung
Primarily assigned technology terms:
- algorithm
- approximation
- backtracking
- centroid computation
- classification
- classification system
- clustering
- clustering algorithm
- comparative evaluation
- computing
- content-based evaluation
- decoder
- decoding
- discourse parsing
- discourse tagging
- document clustering
- document summarization
- domain-independent summarization
- extraction-based summarization
- extraction-based system
- extractive summarization
- hidden markov
- hidden markov model
- hidden markov models
- identification
- information retrieval
- information retrieval tasks
- iterative process
- iterative training
- jaccard coefficient
- k-means
- k-means clustering
- k-means training
- language processing
- machine translation
- markov model
- modeling
- multi-document summarization
- nlp
- parameter training
- parsing
- processing
- question-answering
- re-estimation
- scoring
- segmental clustering
- segmentation
- selection algorithm
- sentence clustering
- sentence extraction
- sentence segmentation
- sentence selection
- single document summarization
- speech processing
- spread activation
- stochastic process
- story segmentation
- summarization
- summarization system
- summarizer
- supervised training
- synthesis
- tagging
- task-oriented evaluation
- text generation
- text segmentation
- theme classification
- thresholding
- topic detection
- topic identification
- training algorithm
- training method
- training process
- unsupervised clustering
- unsupervised training
- viterbi
- viterbi decoder
- viterbi decoding
- viterbi training
Other assigned terms:
- annotation
- approach
- bigram
- case
- cluster
- cluster number
- clustering model
- clusters
- coefficient
- cohesion
- compression ratio
- concept
- concepts
- convergence
- cosine measure
- cosine similarity
- cosine similarity measure
- data sets
- density function
- dice
- dice coefficient
- discourse
- discourse structures
- distribution
- document
- document vectors
- estimation
- euclidean distance
- evaluation method
- experimental results
- feature
- feature vector
- feature vectors
- feature weights
- frame
- frequency counts
- generation
- generation process
- heuristics
- index
- index terms
- knowledge
- labeling
- lexical items
- likelihood
- linguistic
- linguistic information
- linguistic knowledge
- manual annotation
- mapping
- markov models
- measure
- measures
- method
- model parameter
- model parameters
- multi-document summarization task
- negative binomial
- noisy channel
- paragraph
- paragraphs
- part-of-speech
- part-of-speech tags
- poisson distribution
- probabilistic approach
- probabilistic framework
- probabilistic model
- probabilistic models
- probabilities
- probability
- probability density
- probability density function
- probability distribution
- probability distributions
- process
- query
- relative frequency
- relative frequency count
- schema
- segments
- sentence
- sentence boundaries
- sentences
- similarity measure
- similarity measures
- similarity score
- similarity scores
- summarization task
- synonyms
- system performance
- tags
- term
- term distribution
- terms
- testing data
- text
- text cohesion
- text segment
- topics
- training
- training corpus
- training data
- training documents
- training set
- transition probabilities
- unigram
- user
- user query
- vector space
- vocabulary
- words
