machine learning - Mutual information for continuous/numeric features -


I want to calculate mutual information for continuous / numerical features. I want to apply the feature selection based on this. Feature set details are given below

Feature can assume any value between 1: 1 - 10000 feature 2: Measures are spent at some time - Thus any value can be assumed but Integer (big) .... let me have these features.

I am confused in implementing mutual information threads for this, Wikipedia says that integration requires continuous variables.

Do I have to exclude facilities before implementing MI?

I think you have to separate the features before applying MI

When the information benefit applies for feature selection, variable speed continues, a split point is chosen to split the value position for the variable. In different parts, one needs to evaluate all the possible split points to get the best one for the convenience. I think this feature is similar to mutual information for selection, so you can separate the different deviation values ​​from that place, instead I think that would be the same, if the price position is smooth

Comments

Popular posts from this blog

Pass DB Connection parameters to a Kettle a.k.a PDI table Input step dynamically from Excel -

multithreading - PhantomJS-Node in a for Loop -

c++ - MATLAB .m file to .mex file using Matlab Compiler -