ACL RD-TEC 1.0 Summarization of P06-1048
Paper Title:
MODELS FOR SENTENCE COMPRESSION: A COMPARISON ACROSS DOMAINS, TRAINING REQUIREMENTS AND EVALUATION MEASURES
MODELS FOR SENTENCE COMPRESSION: A COMPARISON ACROSS DOMAINS, TRAINING REQUIREMENTS AND EVALUATION MEASURES
Authors: James Clarke and Mirella Lapata
Primarily assigned technology terms:
- abstracting
- algorithm
- automatic evaluation
- automatic evaluation method
- automatic sentence compression
- automatic speech recognition
- classi cation
- classi er
- compression process
- computational linguistics
- corpus creation
- decision-tree
- dynamic programming
- information retrieval
- internet
- judgement elicitation
- language generation
- language modeling
- learning
- leave-one-out resampling
- machine translation
- matching
- modeling
- morphology
- mt systems
- natural language generation
- noisy-channel model
- optimisation
- parameter estimation
- parameter tuning
- paraphrasing
- parser
- parsers
- parsing
- processing
- processing tools
- recognition
- recognition system
- resampling
- scoring
- scoring function
- search
- sentence compression
- shift-reduce parsing
- speech recognition
- speech recognition system
- statistical parsing
- summarisation
- taggers
- term weighting
- text processing
- text rewriting
- tuning
- weighting
- word deletion
- word reordering
- word-based compression
Other assigned terms:
- annotated corpora
- annotator
- annotators
- approach
- appropriate compression model
- association for computational linguistics
- bleu
- break
- british national corpus
- broadcast news
- broadcast news corpus
- case
- comparative study
- compression corpus
- context-free grammar
- corpora
- corpus frequency
- correlation
- data set
- data sets
- debugging
- decision-based model
- decision-tree model
- derivations
- development cycle
- distribution
- document
- document frequency
- edit distance
- estimation
- evaluation data
- evaluation measures
- evaluation method
- evaluation metric
- evaluations
- f-score
- fact
- feature
- feature space
- function words
- generation
- genre
- gold standard
- grammar
- grammatical relations
- grammatical-functional information
- grammaticality
- head word
- human compression performance
- human performance
- implementation
- information content
- knowledge
- lambda
- language model
- language modeling toolkit
- language models
- lexical items
- likelihood
- linguistic
- linguistic knowledge
- linguistics
- measure
- measures
- mechanisms
- method
- model parameters
- modeling toolkit
- n-gram
- natural language
- news corpus
- non-parallel corpus
- nouns
- parallel corpora
- parallel corpus
- parse
- parse tree
- parsing paradigm
- part-of-speech
- part-of-speech tags
- penn treebank
- pos tag
- precision
- probability
- process
- relation
- reordering
- scalability
- semantic
- sentence
- sentence level
- sentences
- significance score
- speech corpora
- speech data
- spoken language
- string edit distance
- style
- syntactic constituent
- syntactic constituents
- syntactic trees
- system development
- tags
- technique
- television
- term
- terms
- text
- text corpus
- tokens
- toolkit
- training
- training data
- training set
- transcript
- transcripts
- translation quality
- tree
- treebank
- trees
- trigram
- trigram language model
- verb
- vocabulary
- vocabulary size
- wilcoxon test
- word
- word level
- word order
- word-based model
- words
- ziff-davis corpus
