Abstract
Ensemble algorithms have been a very popular research topic because of their high performances. In this work, performance based ensemble pruning and decision weighting methods are investigated on 3 ensemble algorithms (Bagging, Random Subspaces, Random Forest) over 26 classification datasets. According to our experiments; the algorithm including most diversity among its base learners is Random Subspaces. The best performed ensemble algorithm is Random Subspaces with decision weighting.
Keywords
Random forest
Weighting
Linear subspace
Pruning
Ensemble learning
Random subspace method
Computer science
Artificial intelligence
Voting
Majority rule
Machine learning
Pattern recognition (psychology)
Base (topology)
Decision tree
Random permutation
Data mining
Algorithm
Mathematics
Subspace topology
Subject Areas
Advanced Statistical Methods and Models
·Statistics and Probability ·Physical Sciences
Advanced Statistical Process Monitoring
·Statistics, Probability and Uncertainty ·Social Sciences
Face and Expression Recognition
·Computer Vision and Pattern Recognition ·Physical Sciences
OpenAlex SDG Match
SDGs auto-classified by OpenAlex (score ≥ 0.4 shown).
Life in Land 63%
Loading…