Ensemble classifiers have been widely used and have shown a great performance improvement on classification accuracy. Now Ensemble learning have gained big attention by the Machine Learning researchers for the dimensionality reduction. But the question is which parameters for ensembles should be taken into consideration when designing and comparing ensemble learning?
Ensemble known for its diversity (that is creates in the training data for training a classifier model). It uses various techniques for selecting the optimal set of data or features (like RF, random forest uses the entropy, gain, gini index etc.). Based on these features the diversity is achieved.
Therefore, during the dimension reduction procedure above said techniques can be used with different features parameters.
hope it will help.
I suggest the following link, it may be helpful for you .
https://arxiv.org/ftp/arxiv/papers/1710/1710.04484.pdf
Check this out https://ieeexplore.ieee.org/abstract/document/6247498
- Badges
- Science topic