GVOTE		GREL		GDEF		GHEA		GENT		
GVOTE		0.97080		0.00000		0.02919		0.00000		0.00000		
GREL		0.14285		0.65714		0.11428		0.05714		0.02857		
GDEF		0.01923		0.00000		0.98076		0.00000		0.00000		
GHEA		0.08219		0.00000		0.01369		0.90410		0.00000		
GENT		0.06250		0.00000		0.02083		0.02083		0.89583		
Accuracy: 0.924433249370277


Classifier type: NB_MN
File Lists Root Dir: h:/Reuters/data.new/reduced_set
Data files root dir: h:/Reuters
Feature selector type: TF
Feature cutoff: 30000
[GHEA, GVOTE, GDEF, GREL, GENT]
Applying TF feature selection...
Feature Selector:TF, Cutoff:30000
Total word types: 40147
Now testing classifier: Naive Bayes Multinomial, EM: false
		GVOTE		GREL		GDEF		GHEA		GENT		
GVOTE		0.93430		0.00000		0.05109		0.01459		0.00000		
GREL		0.20000		0.51428		0.20000		0.05714		0.02857		
GDEF		0.14423		0.00000		0.82692		0.02884		0.00000		
GHEA		0.21917		0.01369		0.04109		0.72602		0.00000		
GENT		0.18750		0.04166		0.04166		0.06250		0.66666		
Accuracy: 0.7984886649874056

Classifier type: NB_MN
File Lists Root Dir: h:/Reuters/data.new/reduced_set
Data files root dir: h:/Reuters
Feature selector type: TF
Feature cutoff: 3
[GHEA, GVOTE, GDEF, GREL, GENT]
Feature Selector:TF, Cutoff:3
Total word types: 40147
  Instances: 1204488, Features: 0, Output: 19426
Now testing classifier: Naive Bayes Multinomial, EM: false
		GVOTE		GREL		GDEF		GHEA		GENT		
GVOTE		0.95620		0.00000		0.02919		0.00729		0.00729		
GREL		0.05714		0.77142		0.05714		0.08571		0.02857		
GDEF		0.01923		0.00000		0.98076		0.00000		0.00000		
GHEA		0.06849		0.00000		0.01369		0.91780		0.00000		
GENT		0.04166		0.00000		0.02083		0.02083		0.91666		
Accuracy: 0.9345088161209067
LDAP: couldn't connect to LDAP server
nlp/tf-feature-selection-with-naive-bayes-multinomial.txt · Last modified: 2015/04/23 21:43 by ryancha
Back to top
CC Attribution-Share Alike 4.0 International
chimeric.de = chi`s home Valid CSS Driven by DokuWiki do yourself a favour and use a real browser - get firefox!! Recent changes RSS feed Valid XHTML 1.0