The effect of distance metrics on boosting with dynamic weighting schemes

Authors:
Xinzhu Yang;Bo Yuan;Wenhuang Liu
Affiliations:
Graduate School at Shenzhen, Tsinghua University, Shenzhen, P.R. China;Graduate School at Shenzhen, Tsinghua University, Shenzhen, P.R. China;Graduate School at Shenzhen, Tsinghua University, Shenzhen, P.R. China
Venue:
FSKD'09 Proceedings of the 6th international conference on Fuzzy systems and knowledge discovery - Volume 1
Year:
2009

Citing 10
Cited 0

The Strength of Weak Learnability

Machine Learning
Hierarchical mixtures of experts and the EM algorithm

Neural Computation
Bagging predictors

Machine Learning
A decision-theoretic generalization of on-line learning and an application to boosting

Journal of Computer and System Sciences - Special issue: 26th annual ACM symposium on the theory of computing & STOC'94, May 23–25, 1994, and second annual Europe an conference on computational learning theory (EuroCOLT'95), March 13–15, 1995
Boosting classifiers regionally

AAAI '98/IAAI '98 Proceedings of the fifteenth national/tenth conference on Artificial intelligence/Innovative applications of artificial intelligence
Random Forests

Machine Learning
iBoost: Boosting Using an i nstance-Based Exponential Weighting Scheme

ECML '02 Proceedings of the 13th European Conference on Machine Learning
A local boosting algorithm for solving classification problems

Computational Statistics & Data Analysis
Issues in stacked generalization

Journal of Artificial Intelligence Research
Fractional distance measures for content-based image retrieval

ECIR'05 Proceedings of the 27th European conference on Advances in Information Retrieval Research

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper presents some preliminary experimental results on RegionBoost, which is a typical example of a class of Boosting algorithms based on dynamic weighting schemes. It is shown that the performance of RegionBoost with the k-Nearest Neighbor (kNN) algorithm as the competency predictor of its basic classifiers can be significantly improved on a variety of standard UCI benchmark datasets by using non-Euclidean distance metrics.