Randomized methods of significance testing enable estimation of the probability that an increase in score has occurred simply by chance. In this paper, we examine the accuracy of three randomized methods of significance testing in the context of machine translation: paired bootstrap resampling, bootstrap resampling and approximate randomization. We carry out a large-scale human evaluation of sh...

Model credibility index is defined to be a sample size under which the power of rejection equals 0.5. It applies goodness-of-fit testing thinking and uses a one-number summary statistic as an assessment tool in a false model world. The estimation of the model credibility index involves a bootstrap resampling technique. To assess the consistency of the estimator of model credibility index, we in...

SUMMARY. The block bootstrap for time series consists in randomly resampling blocks of consecutive values of the given data and aligning these blocks into a bootstrap sample. Here we suggest improving the performance of this method by aligning with higher likelihood those blocks which match at their ends. This is achieved by resampling the blocks according to a Markov chain whose transitions de...

A new performance evaluation paradigm for computer vision systems is proposed. In real situation, the complexity of the input data and/or of the computational procedure can make traditional error propagation methods infeasible. The new approach exploits a resampling technique recently introduced in statistics, the bootstrap. Distributions for the output variables are obtained by perturbing the ...

We have recently developed a distance metric for efficiently estimating the number of substitutions per site between unaligned genome sequences. These substitution rates are called "anchor distances" and can be used for phylogeny reconstruction. Most phylogenies come with bootstrap support values, which are computed by resampling with replacement columns of homologous residues from the original...

Bagging is an ensemble method that relies on random resampling of a data set to construct models for the ensemble. When only statistics about the data are available, but no individual examples, the straightforward resampling procedure cannot be implemented. The question is then whether bagging can somehow be simulated. In this paper we propose a method that, instead of computing certain heurist...

Resampling schemes, and especially the bootstrap method, were proposed as a subclass of perturbation methods to ensure statistical conndentiality in statistical databases. Later, a method based on bootstrapping was presented to achieve the more speciic task of anonymising contingency tables. In this paper, we argue that the latter proposal is either ineecient from a computational point of view ...

It is well-known that diversity among base classifiers is crucial for constructing a strong ensemble. Most existing ensemble methods obtain diverse individual learners through resampling the instances or features. In this paper, we propose an alternative way for ensemble construction by resampling pairwise constraints that specify whether a pair of instances belongs to the same class or not. Us...

If a data set suffers from selection bias then, ordinary regression estimators must be adapted in order to achieve consistency. Consequently, it is necessary to modify any goodness of fit tests based on distances between parametric and non-parametric residuals. More particularly, any bootstrap approximation to these tests must be re– designed. Our first approach to this latter task is based on ...

Uncertainty of insurance liabilities has always been the key issue in actuarial theory and practice. This is represented for instance by study and modeling of mortality in life insurance and loss distributions in traditional actuarial science. These models have evolved from early simple deterministic calculations to more sophisticated, probabilistic ones. Such probabilistic models have been tra...