ACL RD-TEC 1.0 Summarization of W00-0704
Paper Title:
THE ROLE OF ALGORITHM BIAS VS INFORMATION SOURCE IN LEARNING ALGORITHMS FOR MORPHOSYNTACTIC DISAMBIGUATION
THE ROLE OF ALGORITHM BIAS VS INFORMATION SOURCE IN LEARNING ALGORITHMS FOR MORPHOSYNTACTIC DISAMBIGUATION
Authors: Guy De Pauw and Walter Daelemans
Primarily assigned technology terms:
- algorithm
- approximation
- beam search
- classification
- cross-validation
- decision tree
- disambiguation
- disambiguation problem
- entropy learning
- entropy-based learning
- error analysis
- feature generation
- feature selection
- hardware
- inductive learning
- instantiation
- iterative scaling
- language processing
- lazy learning
- learning
- learning algorithm
- learning algorithms
- learning approaches
- learning methods
- maximum entropy
- memory-based learning
- modeling
- natural language processing
- nearest neighbors
- nlp
- objective evaluation
- preprocessing
- processing
- search
- speech tagging
- tagger
- tagger generator
- taggers
- tagging
- tuning
- validation
- variant selection
Other assigned terms:
- adjective
- approach
- beam
- benchmark
- bias
- binary feature
- binary features
- break
- case
- contextual feature
- contextual features
- corpora
- data set
- distribution
- entropy
- feature
- feature set
- feature space
- feature value
- feature vectors
- generation
- heuristic
- implementation
- information gain
- information source
- information sources
- knowledge
- knowledge structures
- lexicon
- lexicon entry
- likelihood
- linguistic
- linguistic context
- local context
- meaning
- method
- morphological features
- natural language
- part-of-speech
- part-of-speech tag
- past participle
- probabilistic model
- probability
- probability tag sequence
- right-hand side
- sentence
- sentence level
- sources of information
- statistical model
- suffix
- tag sequence
- tagging accuracy
- tagging problem
- tags
- target word
- terms
- test phase
- test set
- training
- training data
- training set
- tree
- verb
- word
- words
