ACL RD-TEC 1.0 Summarization of W04-2414
Paper Title:
MEMORY-BASED SEMANTIC ROLE LABELING: OPTIMIZING FEATURES, ALGORITHM, AND OUTPUT
MEMORY-BASED SEMANTIC ROLE LABELING: OPTIMIZING FEATURES, ALGORITHM, AND OUTPUT
Authors: Antal van den Bosch and Sander Canisius and Walter Daelemans and Iris Hendrickx and Erik Tjong Kim Sang
Primarily assigned technology terms:
- algorithm
- classification
- classifier
- classifier stacking
- classifiers
- clustering
- feature relevance weighting
- feature representation
- feature selection
- feature selection process
- feature weighting
- hill-climbing
- k-nn
- labeler
- learner
- learning
- learning method
- machine learner
- majority voting
- matching
- memory-based learner
- memory-based learning
- memory-based learning method
- nearest neighbors
- optimization
- parameter optimization
- post-processing
- predictor
- relevance weighting
- sampling
- search
- searching
- selection process
- semantic role labeler
- semantic role labeling
- simple majority voting
- voting
- voting mechanism
- weighting
- weighting method
Other assigned terms:
- approach
- chunk
- chunks
- classification accuracy
- classification task
- classification tasks
- contextual information
- data set
- development set
- fact
- feature
- grammatical relation
- heuristic
- labeling
- main verb
- mappings
- method
- n-gram
- n-grams
- named entity
- parameter settings
- part-ofspeech
- phrase
- process
- relation
- representations
- role labeling
- semantic
- semantic role
- sentence
- sentences
- tags
- technique
- test data
- test material
- test set
- tokens
- training
- training data
- training material
- training set
- trigram
- unigram
- verb
- words
- wrapper
