Synergetic KDRF Classifier: A Random Forest-Based Hybrid Approach for Multi-Class Classification Task

Abstract

Multi-class classification presents a significant challenge in supervised machine learning. We introduce a novel methodology that integrates the K-Nearest Neighbor (KNN) classifier and Decision Trees into the Random Forest framework to enhance performance. The proposed method trains KNN and Decision Tree classifiers on extracted features; output probabilities form meta-features fed into a Random Forest as the meta-learner. Through extensive experimentation on 5 datasets, the proposed approach demonstrates superior performance in accuracy and efficiency compared to traditional methods.

Contributions

1. 01

   Novel KDRF methodology integrating KNN and Decision Trees into a Random Forest meta-learner framework.

2. 02

   Probability-based meta-feature extraction that transforms base classifier outputs into richer signals.

3. 03

   Superior accuracy and efficiency demonstrated across 5 diverse benchmark datasets.

Headline results