ACL RD-TEC 1.0 Summarization of W03-1306
Paper Title:
BOOSTING PRECISION AND RECALL OF DICTIONARY-BASED PROTEIN NAME RECOGNITION
BOOSTING PRECISION AND RECALL OF DICTIONARY-BASED PROTEIN NAME RECOGNITION
Authors: Yoshimasa Tsuruoka and Jun'ichi Tsujii
Primarily assigned technology terms:
- algorithm
- approximate string matching
- automatic information extraction
- bayes classifier
- binary classification
- boosting
- boyer-moore algorithm
- candidate recognition
- classification
- classifier
- classifiers
- database
- dictionary matching
- dynamic programming
- dynamic programming technique
- elastic matching
- encoding
- entity recognition
- exact matching
- fast matching
- gene\/protein name recognition
- hidden markov
- hidden markov model
- identification
- image recognition
- information extraction
- information processing
- information retrieval
- language processing
- learning
- learning method
- learning technique
- learning techniques
- longest matching
- machine learning
- machine learning techniques
- markov model
- matching
- matching algorithm
- matching technique
- maximum entropy
- naive bayes
- naive bayes classifier
- name recognition
- named entity recognition
- natural language processing
- partial match
- processing
- programming technique
- protein name recognition
- recognition
- recognition system
- recognizer
- search
- searching
- spelling
- string matching
- string searching
- support vector machines
- transcription
Other assigned terms:
- abbreviation
- annotated corpus
- annotation
- bayes model
- binary feature
- candidate term
- candidate terms
- case
- characters
- class probability
- classification task
- conditional independence
- contextual features
- device
- dictionary
- edit distance
- entropy
- entropy models
- estimation
- exact match
- experimental results
- f-measure
- f-score
- fact
- feature
- feature sets
- feature vector
- gene\/protein name
- genia
- genia corpus
- implementation
- information sources
- kappa
- local context
- mapping
- maximum entropy models
- measure
- medline
- mesh
- method
- naive bayes model
- named entity
- names
- natural language
- ontology
- partial match criterion
- precision
- probability
- procedure
- process
- protein names
- protein-protein interaction
- recognition phase
- search results
- semantic
- semantic class
- sentence
- similarity measure
- support vector
- symbols
- technique
- term
- terms
- test data
- text
- training
- training data
- uniform-cost edit distance
- word
- words
