FEATURE SUBSET GENERATION FOR ENSEMBLE LEARNING USING FEATURE CLUSTERING AND MUTUAL INFORMATION
Keywords:
Ensemble Learning (EL), Feature Clustering, Feature Subset Generation (VG), Minimum Redundancy-Maximum Relevance Algorithm, Support Vector Machine (SVM)Abstract
Ensemble learning is a powerful technique for constructing accurate predictive models. Feature subset generation is an important step for ensemble learning. This paper proposes feature clustering and mutual information as a new feature subset generation method for ensemble learning. The proposed feature subset generation technique clusters the features using a hierarchical clustering algorithm. Mutual information is used to compute the similarity between the features within each cluster. Feature subset generation is then performed by selecting the most informative features from each cluster. Experiments are conducted on a real-world dataset to compare the proposed feature subset generation technique to other existing feature subset generation techniques. The experimental results show that the proposed technique outperforms existing feature subset generation techniques.
Downloads
Published
How to Cite
Issue
Section
Copyright (c) 2023 Journal of Applied Artificial Intelligence
This work is licensed under a Creative Commons Attribution 4.0 International License.