ACL RD-TEC 1.0 Summarization of J03-1002
Paper Title:
A SYSTEMATIC COMPARISON OF VARIOUS STATISTICAL ALIGNMENT MODELS
A SYSTEMATIC COMPARISON OF VARIOUS STATISTICAL ALIGNMENT MODELS
Authors: Franz Josef Och and Hermann Ney
Primarily assigned technology terms:
- algorithm
- alignment computation
- alignment system
- approximation
- automatic evaluation
- baum-welch algorithm
- bootstrap
- bootstrapping
- categorization
- computational linguistics
- computer science
- computing
- decomposition
- dynamic programming
- dynamic programming approach
- em algorithm
- em training
- expectation maximization
- greedy search
- greedy search algorithm
- heuristic symmetrization
- hidden markov
- hidden markov models
- hill climbing
- hill-climbing
- hmms
- learning
- learning procedure
- linear interpolation
- linking
- machine translation
- machine translation system
- machine translation systems
- markov alignment model
- maximum-entropy
- maximum-likelihood
- measuring
- modeling
- normalization
- parallel training
- parameter estimation
- phrase-based statistical machine translation
- postprocessing
- recognition
- scoring
- search
- search algorithm
- smoothing
- smoothing techniques
- speech recognition
- speech translation
- statistical alignment
- statistical estimation
- statistical learning
- statistical machine translation
- statistical machine translation system
- statistical methods
- statistical modeling
- symmetrization
- training algorithm
- translation process
- translation system
- translation systems
- viterbi
- viterbi alignment
- viterbi training
- word alignment
Other assigned terms:
- aligned sentence
- alignment error rate
- alignment model
- alignment models
- alignment probability
- annotated corpus
- annotation
- annotation scheme
- annotator
- annotators
- appointment scheduling
- approach
- association for computational linguistics
- association score
- bilingual corpus
- bilingual dictionary
- case
- co-occurrence
- co-occurrence count
- coefficient
- concept
- corpora
- derivation
- dialogues
- dice
- dice coefficient
- dictionary
- dictionary entries
- dictionary entry
- distribution
- error rate
- estimation
- evaluation methodology
- fact
- french
- function words
- german-english verbmobil task
- heuristic
- heuristics
- human annotators
- idiomatic expressions
- implementation
- index
- information science
- interpolation
- knowledge
- language pair
- language pairs
- large training
- lexicographer
- lexicography
- lexicon
- lexicon entry
- likelihood
- linear combination
- linguistics
- log-linear combination
- mapping
- markov models
- measure
- measures
- method
- methodology
- model parameters
- multinomial model
- natural language
- natural languages
- parallel text
- parallel training corpus
- parameter values
- phrase
- precision
- prepositions
- probabilities
- probability
- probability distributions
- procedure
- process
- programming approach
- punctuation
- punctuation marks
- relation
- search problem
- sentence
- sentence boundaries
- sentence pair
- sentences
- similarity function
- source language
- source language word
- source sentence
- speech translation task
- statistical alignment models
- statistical approach
- statistical model
- statistical models
- statistics
- symbol
- target language
- target language string
- target string
- target word
- technique
- technology
- test corpus
- text
- theory
- training
- training corpora
- training corpus
- training data
- training time
- transcriptions
- translation model
- translation probability
- translation quality
- translation task
- translations
- uniform distribution
- vocabulary
- word
- word alignments
- word classes
- word order
- word strings
- words
