Sparse Multinomial Logistic Regression

GPTKB entity

Statements (32)
Predicate Object
gptkbp:instanceOf gptkb:model
statistical analysis
gptkbp:advantage automatic feature selection
sparse solutions
gptkbp:application bioinformatics
image recognition
text classification
gptkbp:author gptkb:John_D._Lafferty
gptkb:Andrew_McCallum
gptkb:Fernando_Pereira
gptkbp:basedOn logistic regression
gptkbp:citation gptkb:Lafferty,_McCallum,_Pereira,_2001
gptkbp:featureSelection yes
https://www.w3.org/2000/01/rdf-schema#label Sparse Multinomial Logistic Regression
gptkbp:implementedIn gptkb:LIBLINEAR
gptkb:scikit-learn
R glmnet package
gptkbp:input feature vectors
gptkbp:limitation computationally intensive for large datasets
requires tuning of regularization parameter
gptkbp:optimizedFor convex optimization
gptkbp:output probabilities for each class
gptkbp:publishedIn gptkb:NIPS_2001
gptkbp:regularization lasso
L1 regularization
gptkbp:relatedTo gptkb:lasso_regression
softmax regression
multinomial logistic regression
gptkbp:usedFor gptkb:dictionary
multiclass classification
gptkbp:bfsParent gptkb:SMLR
gptkbp:bfsLayer 6