ACL RD-TEC 1.0 Summarization of W05-0638
Paper Title:
EXPLOITING FULL PARSING INFORMATION TO LABEL SEMANTIC ROLES USING AN ENSEMBLE OF ME AND SVM VIA INTEGER LINEAR PROGRAMMING
EXPLOITING FULL PARSING INFORMATION TO LABEL SEMANTIC ROLES USING AN ENSEMBLE OF ME AND SVM VIA INTEGER LINEAR PROGRAMMING
Authors: Tzong-Han Tsai and Chia-Wei Wu and Yu-Chun Lin and Wen-Lian Hsu
Primarily assigned technology terms:
- argument classification
- argument identification
- charniak parser
- classification
- classifier
- classifiers
- collins parser
- computational linguistics
- computational natural language learning
- full parsing
- identification
- integer linear programming
- kernel
- language learning
- learning
- learning algorithms
- linear learning
- linear programming
- machine learning
- machine learning algorithms
- maxent
- maximum entropy
- natural language learning
- parser
- parsers
- parsing
- polynomial kernel
- pruning
- scoring
- semantic role labeling
- sentence tagging
- srl system
- svm-light
- tagging
Other assigned terms:
- association for computational linguistics
- brown corpus
- chunk
- chunk type
- chunks
- classification accuracy
- classification model
- context features
- corpora
- development set
- entropy
- evaluation metrics
- experimental results
- f-score
- feature
- head word
- heuristic
- heuristic rules
- heuristics
- labeling
- lemma
- linguistics
- meaning
- method
- named entities
- natural language
- parse
- parsing information
- parsing tree
- penn treebank
- phrase
- phrase structure
- phrase structure rule
- phrase type
- pos tag
- propbank
- propbank corpus
- representations
- role labeling
- semantic
- semantic role
- semantic roles
- sentence
- structure rule
- syntactic constraints
- syntactic information
- system architecture
- tagging problem
- test set
- training
- training corpus
- training data
- training time
- tree
- treebank
- trees
- verb
- word
- words
