ACL RD-TEC 1.0 Summarization of W05-0637
Paper Title:
APPLYING SPELLING ERROR CORRECTION TECHNIQUES FOR IMPROVING SEMANTIC ROLE LABELLINGAuthors: Erik Tjong Kim Sang and Sander Canisius and Antal
APPLYING SPELLING ERROR CORRECTION TECHNIQUES FOR IMPROVING SEMANTIC ROLE LABELLING
Authors: Erik Tjong Kim Sang and Sander Canisius and Antalvan den Bosch and Toine Bogers
Primarily assigned technology terms:
- algorithm
- automatic feature selection
- classification
- classifier
- classifiers
- computational linguistics
- computational natural language learning
- error correction
- feature selection
- feature selection and post-processing
- feature selection process
- feature weighting
- hill-climbing
- k-nn
- language learning
- learner
- learning
- learning algorithms
- learning approach
- learning technique
- learning techniques
- levenshtein
- machine learning
- machine learning algorithms
- machine learning approach
- maximum entropy
- memory-based learner
- memory-based learning
- natural language learning
- optimisation
- post-processing
- pruning
- repair
- search
- searching
- selection process
- semantic role labeling
- spelling
- spelling correction
- support vector machines
- uniform feature weighting
- voting
- weighting
Other assigned terms:
- approach
- association for computational linguistics
- case
- classification tasks
- constraint satisfaction
- data set
- data sets
- development set
- edit distance
- entropy
- entropy models
- feature
- feature set
- labeling
- levenshtein distance
- lexicon
- linguistics
- maximum entropy models
- method
- natural language
- phrase
- procedure
- process
- propbank
- relation
- role labeling
- semantic
- semantic role
- sentence
- spelling error
- support vector
- syntactic trees
- technique
- toolkit
- training
- training corpus
- training material
- trees
- verb
- word
- word data
- word level
- words
