Information Gain Attribute Evaluation
Contact: Stefan Kramer
Categories: Feature selection
Exposed methods:
Feature selection |
|
---|---|
Input: | Instances, feature vectors, class values |
Output: | Instances, feature vectors, class values |
Input format: | Weka's ARFF format |
Output format: | Weka's ARFF format |
User-specified parameters: | Number of features to select (non-mandatory) Information Gain Threshold (non-mandatory) |
Reporting information: | Attributes ranked by Information Gain |
Description:
InfoGainAttributeEval evaluates the worth of an attribute by measuring the information gain with respect to the
class.
InfoGain(Class,Attribute) = H(Class) – H(Class | Attribute), where H is the information entropy.
Background (publication date, popularity/level of familiarity, rationale of approach, further comments)
Widely used standard feature selection method, disadvantage: does not take into
account feature interaction
Class-blind/class-sensitive feature selection
Class-sensitive feature selection
Type (optimal, greedy, randomized)
Optimal
Filter/wrapper/hybrid approach
Filter
Type of Descriptor:
Interfaces:
Priority: High
Development status:
Homepage:
Dependencies:
External components: WEKA
Technical details
Data: No
Software: Yes
Programming language(s): Java
Operating system(s): Linux, Win, Mac OS
Input format: Weka's ARFF format
Output format: Weka's ARFF format
License: GPL