Feature selection algorithms in Scikit-learn

Question

I am using Scikit-learn for a binary classification task.. and I am trying to perform Feature selection using one of these methods : Correlation-based Feature Selection (CFS) Subset Evaluator and Best First search algorithm... I couldn't find any of them in the Feature selection module ! are they listed with another name in Scikit learn or is there any equivalent ?! because I want to repeat an experiment and they have used these algorithms in reducing there data.

ASU’s scikit feature is available as a Python package (Ayan Das has already mentioned it, but only referring to the Matlab implementation). It includes a CFS algorithm and many other similarity based and statistical based functions (ReliefF, Chi-square, …). — Sebastian Berns
– Sebastian Berns, Commented Dec 31, 2016 at 16:56

Ami Tavory · Accepted Answer · 2016-01-30 18:43:57Z

2

In sklearn.feature_selection: SelectKBest and SelectPercentile assess subset performance, and RFE does recursive feature elimination.

As for Best First Search, see Smart Feature Selection with scikit-learn and BigML’s API.

I am very skeptical you'll be able to exactly reproduce some experimental results based on the names of the algorithms used, FWIW.

answered Jan 30, 2016 at 18:43

Ami Tavory

76.7k13 gold badges152 silver badges196 bronze badges

Sign up to request clarification or add additional context in comments.

Collectives™ on Stack Overflow

Feature selection algorithms in Scikit-learn

1 Answer 1

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

Comments

Your Answer

Sign up or log in

Post as a guest

Related