Description: ParInfoGain - Computes parallel information gain and gain ratio in Matlab using the Matlab Parallel Computing Toolbox or the Distributed Server (if available)
Information gain is defined as:
InfoGain(Class,Attribute) = H(Class) - H(Class | Attribute)
Gain ratio is defined as:
GainRatio(Class, Attribute) = (H(Class) - H(Class | Attribute)) / H(Attribute).
...where H is the entropy, defined as - sum(i=1 to k) pi log2 pi-ParInfoGain - Computes parallel information gain and gain ratio in Matlab using the Matlab Parallel Computing Toolbox or the Distributed Server (if available)
Information gain is defined as:
InfoGain(Class,Attribute) = H(Class) - H(Class | Attribute)
Gain ratio is defined as:
GainRatio(Class, Attribute) = (H(Class) - H(Class | Attribute)) / H(Attribute).
...where H is the entropy, defined as - sum(i=1 to k) pi log2 pi
Platform: |
Size: 3072 |
Author:agkj |
Hits:
Search - MATLAB Distributed Computing Server