Corpus ID: 2398

Linear and Order Statistics Combiners for Pattern Classification

@article{Tumer1999LinearAO,
  title={Linear and Order Statistics Combiners for Pattern Classification},
  author={Kagan Tumer and Joydeep Ghosh},
  journal={ArXiv},
  year={1999},
  volume={cs.NE/9905012}
}
  • Kagan Tumer, Joydeep Ghosh
  • Published 1999
  • Computer Science
  • ArXiv
  • Several researchers have experimentally shown that substantial improvements can be obtained in difficult pattern recognition problems by combining or integrating the outputs of multiple classifiers. This chapter provides an analytical framework to quantify the improvements in classification results due to combining. The results apply to both linear combiners and order statistics combiners. We first show that to a first order approximation, the error rate obtained over and above the Bayes error… CONTINUE READING
    A theoretical and experimental analysis of linear combiners for multiple classifier systems
    300
    Analysis of error-reject trade-off in linearly combined multiple classifiers
    51
    A weighted voting framework for classifiers ensembles
    121
    Experiments on Linear Combiners
    23
    Input decimated ensembles
    88
    The impact of diversity on the accuracy of evidential classifier ensembles
    • Y. Bi
    • Mathematics, Computer Science
    • 2012
    58

    References

    Publications referenced by this paper.
    SHOWING 1-10 OF 102 REFERENCES
    Analysis of decision boundaries in linearly combined neural classifiers
    350
    Classifier combining through trimmed means and order statistics
    13
    Error Correlation and Error Reduction in Ensemble Classifiers
    616
    On the Link between Error Correlation and Error Reduction in Decision Tree Ensembles
    63
    Bootstrap Techniques for Error Estimation
    177
    Limits to performance gains in combined neural classifiers
    14
    Learning ranks with neural networks
    37
    Integration Of Neural Classifiers For Passive Sonar Signals
    27
    Estimating the Bayes error rate through classifier combining
    59