ACL RD-TEC 1.0 Summarization of P06-1027
Paper Title:
SEMI-SUPERVISED CONDITIONAL RANDOM FIELDS FOR IMPROVED SEQUENCE SEGMENTATION AND LABELING
SEMI-SUPERVISED CONDITIONAL RANDOM FIELDS FOR IMPROVED SEQUENCE SEGMENTATION AND LABELING
Authors: Feng Jiao and Shaojun Wang and Chi-Hoon Lee and Russell Greiner and Dale Schuurmans
Primarily assigned technology terms:
- algorithm
- bootstrapping
- classification
- combined training
- computational linguistics
- computing
- conditional likelihood
- conditional random fields
- crf training
- crfs
- decoder
- disambiguation
- discriminative training
- document classification
- dynamic programming
- dynamic programming approach
- entity recognition
- entropy regularization
- expectation-maximization
- expectation-maximization approach
- global optimization
- gradient optimization
- hidden markov
- hidden markov models
- information extraction
- language processing
- learning
- learning algorithm
- learning approach
- machine learning
- modeling
- named entity recognition
- natural language processing
- optimization
- optimization algorithm
- optimization method
- optimization procedure
- parsing
- part-of-speech tagging
- pattern recognition
- predictor
- processing
- quasi-newton optimization
- recognition
- regularization
- regularization framework
- segmentation
- self-training
- semi-supervised learning
- semi-supervised training
- semi-supervised training procedure
- sense disambiguation
- sequence labeling
- sequence modeling
- sequence segmentation
- stochastic gradient optimization
- structured classification
- structured prediction
- supervised training
- taggers
- tagging
- training algorithm
- training method
- training procedure
- training process
- viterbi
- viterbi decoder
- word segmentation
- word sense disambiguation
Other assigned terms:
- alphabet
- approach
- association for computational linguistics
- biomedical text
- case
- classification accuracy
- classification tasks
- clusters
- composition
- conditional model
- conditional probability
- crf model
- data set
- data sets
- derivation
- distribution
- document
- entropy
- f-measure
- fact
- feature
- feature set
- gene name
- generative models
- kullback-leibler divergence
- labeled training data
- labeling
- language processing tasks
- large corpus
- likelihood
- linguistics
- local maxima
- markov chain
- markov models
- measure
- mechanisms
- method
- model parameters
- named entity
- names
- natural language
- parameter space
- part-of-speech
- part-of-speech tags
- precision
- probability
- probability distribution
- procedure
- process
- processing tasks
- programming approach
- protein names
- recursion
- sentences
- supervised model
- tagging task
- tags
- technique
- term
- terms
- testing data
- text
- training
- training criterion
- training data
- training set
- training time
- unlabeled examples
- word
- word boundaries
- word sense
- words
