Skip to content

Discretization for Naive Bayes

Warning: Invalid argument supplied for foreach() in /home/giwebb5/public_html/i/wp-content/plugins/papercite/lib/BibTex_osbib.php on line 173

Naive Bayes has distinct requirements of discretization due to its attribute independence assumption.  This work provides theoretical analysis of those requirements and new techniques that improve classification accuracy.

PKIDiscretize is a standard Weka component that implements our proportional k-interval discretization technique.