|
gptkbp:instanceOf
|
gptkb:algorithm
gptkb:model
|
|
gptkbp:advantage
|
efficient for large datasets
simple to implement
assumes feature independence
not suitable for continuous data
|
|
gptkbp:assumes
|
features are conditionally independent
features follow multinomial distribution
|
|
gptkbp:author
|
gptkb:Andrew_McCallum
gptkb:Kamir_Nigam
|
|
gptkbp:basedOn
|
gptkb:Bayes'_theorem
|
|
gptkbp:category
|
supervised learning
|
|
gptkbp:commonIn
|
gptkb:NLTK
gptkb:scikit-learn
|
|
https://www.w3.org/2000/01/rdf-schema#label
|
Multinomial naive Bayes
|
|
gptkbp:input
|
discrete features
|
|
gptkbp:introducedIn
|
1998
|
|
gptkbp:output
|
class label
|
|
gptkbp:pdf
|
gptkb:A_Comparison_of_Event_Models_for_Naive_Bayes_Text_Classification
|
|
gptkbp:relatedTo
|
gptkb:Bernoulli_naive_Bayes
gptkb:Gaussian_naive_Bayes
|
|
gptkbp:requires
|
feature counts
likelihood probabilities
prior probabilities
|
|
gptkbp:usedFor
|
spam filtering
text classification
document classification
|
|
gptkbp:usedIn
|
sentiment analysis
language detection
news categorization
topic classification
|
|
gptkbp:bfsParent
|
gptkb:naive_Bayes
|
|
gptkbp:bfsLayer
|
6
|