|
gptkbp:instanceOf
|
statistical analysis
probabilistic graphical model
|
|
gptkbp:advantage
|
relaxes independence assumptions
supports arbitrary, non-independent features
|
|
gptkbp:application
|
speech recognition
bioinformatics
optical character recognition
information extraction
|
|
gptkbp:featureFunction
|
potential function
|
|
gptkbp:field
|
gptkb:machine_learning
computer vision
natural language processing
|
|
gptkbp:form
|
undirected graphical model
|
|
gptkbp:generalizes
|
gptkb:Hidden_Markov_Model
gptkb:Maximum_Entropy_Markov_Model
|
|
gptkbp:hasModel
|
conditional probability distribution
|
|
gptkbp:hasVariant
|
gptkb:higher-order_CRF
gptkb:linear-chain_CRF
gptkb:semi-Markov_CRF
gptkb:skip-chain_CRF
|
|
https://www.w3.org/2000/01/rdf-schema#label
|
Conditional Random Field
|
|
gptkbp:inferenceMethod
|
gptkb:Viterbi_algorithm
belief propagation
|
|
gptkbp:input
|
observation sequence
|
|
gptkbp:introduced
|
gptkb:Andrew_McCallum
gptkb:Fernando_Pereira
gptkb:John_Lafferty
|
|
gptkbp:introducedIn
|
2001
|
|
gptkbp:limitation
|
computationally expensive inference
parameter estimation complexity
|
|
gptkbp:objective
|
log-likelihood
|
|
gptkbp:openSource
|
gptkb:MALLET
gptkb:CRF++
gptkb:CRFsuite
gptkb:sklearn-crfsuite
|
|
gptkbp:optimizedFor
|
gptkb:L-BFGS
gradient descent
stochastic gradient descent
|
|
gptkbp:output
|
label sequence
|
|
gptkbp:parameter
|
maximum likelihood estimation
|
|
gptkbp:publishedIn
|
gptkb:Proceedings_of_the_Eighteenth_International_Conference_on_Machine_Learning_(ICML_2001)
|
|
gptkbp:relatedTo
|
gptkb:Hidden_Markov_Model
gptkb:Markov_Random_Field
gptkb:Maximum_Entropy_Markov_Model
|
|
gptkbp:type
|
discriminative model
|
|
gptkbp:usedFor
|
image segmentation
part-of-speech tagging
named entity recognition
sequence labeling
|
|
gptkbp:bfsParent
|
gptkb:Markov_Random_Field
|
|
gptkbp:bfsLayer
|
6
|