Unique Classifier Selection Approach for Bagging Algorithm

A.B.M.S. Ali (Australia)


Bagging, Classification, Rule Based Method.


Bagging is a popular method that could improve the classification accuracy for any unstable learning algorithm. A trial and error classifier feeding with the Bagging algorithm is a regular practice for classification tasks in the machine learning community. In this research we propose a rule based method using well established meta learning approach for unique classifier selection. The generated rules are verified using 113 classification problems with 10 fold cross validation processes. That makes Bagging is a computationally faster algorithm and provides a unique solution for classifier selection.

Important Links:

Go Back