Skip to content

Latest commit

 

History

History
17 lines (11 loc) · 968 Bytes

README.md

File metadata and controls

17 lines (11 loc) · 968 Bytes

Implementation of AdaBoost classifier

Description

This is an implementation of the AdaBoost algorithm for a two-class classification problem. The algorithm sequentially applies a weak classification to modified versions of the data. By increasing the weights of the missclassified observations, each weak learner focuses on the error of the previous one. The predictions are aggregated through a weighted majority vote.

Methods

Adaboost algorithm:

Example

Using the Hastie (10.2) dataset, we can appreciate a significant reduction in the error rate as we increase the number of iterations.

References

  • Trevor Hastie, Robert Tibshirani, Jerome Friedman - The Elements of Statistical Learning