A Boost Voting Strategy for Knowledge Integration and Decision Making

Authors:
Haibo He;Yuan Cao;Jinyu Wen;Shijie Cheng
Affiliations:
Department of Electrical and Computer Engineering, Stevens Institute of Technology, Hoboken, USA NJ 07030;Department of Electrical and Computer Engineering, Stevens Institute of Technology, Hoboken, USA NJ 07030;College of Electrical and Electronics Engineering, Huazhong University of Science and Technology, Wuhan, China 430074;College of Electrical and Electronics Engineering, Huazhong University of Science and Technology, Wuhan, China 430074
Venue:
ISNN '08 Proceedings of the 5th international symposium on Neural Networks: Advances in Neural Networks
Year:
2008

Citing 10
Cited 0

The Strength of Weak Learnability

Machine Learning
Original Contribution: Stacked generalization

Neural Networks
Bagging predictors

Machine Learning
On Combining Classifiers

IEEE Transactions on Pattern Analysis and Machine Intelligence
The Random Subspace Method for Constructing Decision Forests

IEEE Transactions on Pattern Analysis and Machine Intelligence
Random Forests

Machine Learning
An Adaptive Version of the Boost by Majority Algorithm

Machine Learning
Rotation Forest: A New Classifier Ensemble Method

IEEE Transactions on Pattern Analysis and Machine Intelligence
Pattern Recognition, Third Edition

Pattern Recognition, Third Edition
Adaptive mixtures of local experts

Neural Computation

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper proposes a voting strategy for knowledge integration and decision making systems with information uncertainty. As ensemble learning methods have recently attracted growing attention from both academia and industry, it is critical to understand the fundamental problem of voting strategy for such learning methodologies. Motivated by the signal to noise ratio (SNR) concept, we propose a method that can vote optimally according to the knowledge level of each hypothesis. The mathematical framework based on gradient analysis is used to find the optimal weights, and a voting algorithm, BoostVote, is presented in detail in this paper. Simulation analyses based on synthetic data and real-world data sets with comparison to the existing voting rules demonstrate the effectiveness of this method.