How to assign a weight for classifiers when using weighted majority voting?

I am trying to apply weighted majority voting on an ensemble as a combiner method. I read different papers and articles, however, I am still a bit lost on:

  1. How the weighted majority voting works

  2. How to assign a weight for every ensemble base classifiers when using weighted majority voting.

Topic ensemble-learning ensemble ensemble-modeling

Category Data Science

1
answered at

You can test each base classifier on a hold-out dataset and come up with a performance metric say accuracy for each model. You can then use each model's accuracy as a weight when combining predictions in the ensemble.

Here is a useful article.

About

Geeks Mental is a community that publishes articles and tutorials about Web, Android, Data Science, new techniques and Linux security.