ACL RD-TEC 1.0 Summarization of W05-0610
Paper Title:
USING UNEVEN MARGINS SVM AND PERCEPTRON FOR INFORMATION EXTRACTION
USING UNEVEN MARGINS SVM AND PERCEPTRON FOR INFORMATION EXTRACTION
Authors: Yaoyong Li and Kalina Bontcheva and Hamish Cunningham
Primarily assigned technology terms:
- algorithm
- annie system
- automatic extraction
- binary classi cation
- capitalization
- classi cation
- classi er
- computational linguistics
- computational natural language learning
- cubic kernel
- encoding
- entity recognition
- hmms
- information extraction
- information gathering
- kernel
- knowledge management
- language learning
- language processing
- learning
- learning algorithm
- learning algorithms
- learning methods
- learning system
- learning techniques
- linear kernel
- machine learning
- machine learning algorithm
- machine learning algorithms
- machine learning methods
- machine learning techniques
- matching
- named entity recognition
- natural language learning
- natural language processing
- ne recognition
- nlp
- on-line learning
- optimisation
- perceptron
- perceptron algorithm
- perceptron learning
- perceptron system
- post-processing
- pre-processing
- preprocessing
- processing
- recogniser
- recognition
- rule learning
- semantic web
- statistical learning
- statistical system
- supervised machine learning
- support vector machines
- svm problem
- svm-based system
- template filling
- thresholding
- tuning
- voted perceptron
- weighting
Other assigned terms:
- annotation
- approach
- association for computational linguistics
- benchmark
- capitalization information
- context window
- corpora
- data sets
- development set
- document
- document structure
- english corpus
- entity type
- entity types
- entropy
- events
- f-measure
- feature
- feature vectors
- gazetteer
- generalisation
- ie task
- implementation
- knowledge
- large training
- lattice
- lemma
- linguistic
- linguistic features
- linguistic information
- linguistics
- manual annotation
- mechanisms
- method
- named entities
- named entity
- names
- natural language
- optimisation problem
- parameter settings
- part-of-speech
- part-ofspeech
- positive and negative examples
- probabilities
- probability
- procedure
- process
- punctuation
- quadratic kernel
- semantic
- semantic classes
- semantic information
- sigmoid function
- slot
- statistical model
- support vector
- svm model
- svms
- symbols
- tags
- test set
- text
- tokens
- training
- training data
- training dataset
- training documents
- training example
- training examples
- training set
- training time
- transition probabilities
- user
- web pages
- weighting scheme
- window size
- word
