Skip to content

Latest commit

 

History

History
31 lines (27 loc) · 939 Bytes

README.md

File metadata and controls

31 lines (27 loc) · 939 Bytes
Raw

Feature selction

Using Feature selection methods in high dimensionality of feature space are vital. In machine learning, we can use feature selection techniques in order to select proper subset of feature to increase machine learning accuracy. In this Repository we implemented various feature methods.

Feature selection methods

  • Filter
  • Wrapper
  • Embeded
  • Hybrid#

Feature Selection strategies:

  • Forward selection
  • Backward elimination
  • Forward stepwise selection
  • Backward stepwise elimination
  • Random mutation

implemented filter methods

  • Chi-square
  • Cross Entropy
  • Fuzzy Entropy Measure
  • Gini index
  • Information Gain
  • Mutual Information
  • Relative Discrimination Criteria
  • Term Strength

How to use code

In order to use these code you need to send your data and classes in 2d dimention. also you need to pass your dataset type we provide varoius dataset in our github.