ACL RD-TEC 1.0 Summarization of N06-1012
Paper Title:
REDUCING WEIGHT UNDERTRAINING IN STRUCTURED DISCRIMINATIVE LEARNING
REDUCING WEIGHT UNDERTRAINING IN STRUCTURED DISCRIMINATIVE LEARNING
Authors: Charles Sutton and Michael Sindelar and Andrew McCallum
Primarily assigned technology terms:
- 2-fold cross validation
- adaboost
- algorithm
- bagging
- bayes classifier
- boosting
- boosting algorithm
- brill tagger
- chunking
- classifier
- classifiers
- coding
- computing
- conditional random field
- conditional random fields
- crf training
- crfs
- cross validation
- databases
- decision tree
- decision trees
- decoding
- decoding procedure
- discriminative classifier
- discriminative learning
- dynamic programming
- ensemble learning
- entity recognition
- error analysis
- error reduction
- feature bagging
- finite state
- finite state machines
- forward-backward algorithm
- graphical model
- hidden markov
- hidden markov models
- hmms
- language processing
- learning
- logistic regression
- machine learning
- map training
- markov model
- maximum entropy
- maximum-entropy
- meta-learning
- modeling
- naive bayes
- naive bayes classifier
- named entity recognition
- named-entity recognition
- natural language processing
- natural-language processing
- neural network
- neural networks
- nlp
- normalization
- noun-phrase chunking
- np chunking
- parsing
- part-of-speech tagging
- probabilistic inference
- processing
- recognition
- regression
- sequence labeling
- tagger
- tagging
- validation
- viterbi
- voting
Other assigned terms:
- adjective
- analogy
- arithmetic mean
- bias
- binary features
- case
- chunk
- chunks
- concreteness
- conditional distribution
- conditional model
- conditional probability
- conll-2003 data set
- data set
- data sets
- development set
- distribution
- entropy
- fact
- feature
- feature sets
- gaussian prior
- generative models
- geometric mean
- joint probability
- labeling
- lexicon
- lexicon model
- markov models
- method
- model parameters
- named entities
- named entity
- named entity task
- named-entity
- names
- natural language
- natural-language
- ngram
- ngram model
- noise
- normalization factor
- nouns
- opinion
- part of speech
- part-of-speech
- part-of-speech tags
- penn treebank
- penn treebank project
- pos tag
- precision
- prefixes and suffixes
- probabilistic model
- probabilistic models
- probabilities
- probability
- procedure
- processing tasks
- proper noun
- regular expressions
- relation
- semantic
- sentences
- sequence model
- standard decision tree
- stems
- suffixes
- tags
- technique
- terms
- test data
- testing data
- testing set
- training
- training data
- training set
- training time
- transition probabilities
- tree
- treebank
- treebank project
- trees
- understanding
- weight vector
- word
- wordnet
- words
