Advanced Search Tips

Calculate Normalized Information Measures

Submitted by on Jul 15 2014 Suggest Revision
By: Baogang Hu
From:
Paper:
Resource Type:
Code
License:
Bsd
Language:
Scilab
Data Format:
None

Description

The toolbox is to calculate normalized information measures from a given m by (m+1) confusion matrix for objective evaluations of an abstaining classifier. It includes total 24 normalized information measures based on three groups of definitions, that is, mutual information, information divergence, and cross entropy.
Post comment
Cancel