ACL RD-TEC 1.0 Summarization of W02-2011
Paper Title:
COMBINING LABELLED AND UNLABELLED DATA: A CASE STUDY ON FISHER KERNELS AND TRANSDUCTIVE INFERENCE FOR BIOLOGICAL ENTITY RECOGNITION
COMBINING LABELLED AND UNLABELLED DATA: A CASE STUDY ON FISHER KERNELS AND TRANSDUCTIVE INFERENCE FOR BIOLOGICAL ENTITY RECOGNITION
Authors: Cyril Goutte and Herv� D�jean and Eric Gaussier and Nicola Cancedda and Jean-Michel Renders
Primarily assigned technology terms:
- algorithm
- automatic annotation
- binary classi cation
- biological entity recognition
- bootstrap
- categorisation
- classi cation
- classi er
- clustering
- clustering technique
- database
- databases
- entity extraction
- entity recognition
- expectation-maximisation
- fisher kernel
- inductive inference
- inductive learning
- information extraction
- inner product
- kernel
- kernels
- knowledge processing
- latent semantic analysis
- learning
- learning algorithms
- learning process
- learning techniques
- linear kernel
- machine learning
- machine learning techniques
- nearest neighbors
- optimisation
- parameterization
- polynomial kernel
- pre-processing
- processing
- processing tools
- querying
- radial basis function
- recognition
- retrieving
- search
- semantic analysis
- spelling
- supervised classi cation
- supervised learning
- support vector machines
- tagger
- text categorisation
- transductive inference
- transductive learning
- unsupervised learning
Other assigned terms:
- annotated dataset
- annotation
- approach
- benchmark
- candidate term
- case
- clusters
- co-occurrences
- concepts
- contextual features
- contextual information
- development set
- distribution
- document
- document collection
- drosophila
- english lexicon
- experimental results
- fact
- feature
- feature space
- french
- gene names
- generalisation
- generation
- generative model
- generative models
- heuristics
- interannotator agreement
- interpolation
- kappa
- knowledge
- labeling
- latent semantic
- learning problem
- lexical resources
- lexicon
- likelihood
- linguistic
- log-likelihood
- measure
- medline
- method
- names
- noise
- nouns
- optimisation problem
- part-of-speech
- precision
- prepositions
- probabilistic model
- probabilistic models
- probabilities
- process
- protein names
- search strategy
- semantic
- sentence
- similarity measure
- supervised learning problem
- support vector
- svms
- technique
- term
- terms
- test data
- test set
- text
- tokens
- training
- training data
- training set
- understanding
- user
- user interaction
- word
- words
