ACL RD-TEC 1.0 Summarization of P98-1081
Paper Title:
IMPROVING DATA DRIVEN WORDCLASS TAGGING BY SYSTEM COMBINATION
IMPROVING DATA DRIVEN WORDCLASS TAGGING BY SYSTEM COMBINATION
Authors: Hans van Halteren and Jakub Zavrel and Walter Daelemans
Primarily assigned technology terms:
- abstracting
- algorithm
- beam search
- capitalization
- categorization
- classifier
- classifiers
- combined classifier
- corpus annotation
- decision tree
- decision trees
- error reduction
- hidden markov
- hidden markov model
- induction
- java
- knowledge representation
- language processing
- learner
- learning
- learning method
- learning system
- machine learning
- markov model
- maximum entropy
- maximum entropy model
- measuring
- memory-based learning
- modelling
- natural language processing
- neural net
- nlp
- pairwise voting
- processing
- pruning
- random selection
- search
- selection method
- tagger
- taggers
- tagging
- tagging system
- viterbi
- viterbi algorithm
- voting
- voting system
- weighting
Other assigned terms:
- ambiguity
- annotated training corpus
- annotation
- approach
- beam
- benchmark
- bias
- case
- characters
- context features
- contextual information
- coordination conjunction
- distribution
- english text
- entropy
- error rate
- estimation
- fact
- feature
- feature information
- implementation
- information gain
- knowledge
- language model
- language models
- lexical information
- lexicon
- linguistic
- lob corpus
- measure
- measures
- method
- model context
- n-gram
- names
- natural language
- nlp task
- nlp tasks
- penn treebank
- penn treebank corpus
- precision
- probabilities
- probability
- probability tag sequence
- relation
- sentence
- similarity metric
- statistics
- suffix
- tag sequence
- tagging task
- tags
- tagset
- term
- text
- tokens
- training
- training corpus
- training data
- training material
- training phase
- transformation
- transformation rules
- tree
- treebank
- treebank corpus
- trees
- utterance
- word
- wordform
- words
