ACL RD-TEC 1.0 Summarization of P97-1030
Paper Title:
MISTAKE-DRIVEN MIXTURE OF HIERARCHICAL TAG CONTEXT TREES
MISTAKE-DRIVEN MIXTURE OF HIERARCHICAL TAG CONTEXT TREES
Authors: Masahiko Haruno and Yuji Matsumoto
Primarily assigned technology terms:
- adaboost
- algorithm
- automaton
- data compression
- decision trees
- decision-tree
- dependency parser
- estimator
- japanese dependency parser
- language processing
- learning
- learning algorithm
- learning methods
- likelihood estimator
- machine learning
- maximum likelihood
- natural language processing
- nlp
- parser
- part-of-speech tagger
- part-of-speech tagging
- pos tagging
- pre-processing
- predictor
- processing
- random guess
- re-sampling
- search
- statistical methods
- statistical natural language processing
- stochastic tagger
- tagger
- taggers
- tagging
- tile
- tree construction
- weak learning
- weak learning algorithm
- weighting
Other assigned terms:
- approach
- binary tree
- break
- case
- context model
- corpora
- data sparseness
- dictionary
- distance measure
- distribution
- experimental results
- finite automaton
- generation
- generation process
- hierarchical structure
- input string
- japanese dependency
- japanese sentences
- kl divergence
- leaf
- likelihood
- linguistics
- linguists
- measure
- method
- methodology
- natural language
- part-of-speech
- particles
- probabilities
- probability
- probability distribution
- probability distributions
- procedure
- process
- proper noun
- root node
- search space
- sentence
- sentences
- sparse data
- statistical natural language
- statistics
- subtrees
- suffixes
- symbol
- symbols
- tag model
- tag set
- tagging accuracy
- tagging problem
- tagging scheme
- tags
- target language
- term
- terms
- test corpus
- theory
- training
- training examples
- training phase
- transfer rule
- tree
- tree model
- trees
- weight vector
- word
- word level
- word model
- words
