Secure sketches are useful in extending cryptographic schemes to biometric data since they allow recovery of fuzzy secrets under inevitable noise. In practice, secrets derived from...
With very noisy data, having plentiful samples eliminates overfitting in nonlinear regression, but not in nonlinear principal component analysis (NLPCA). To overcome this problem...
Background: In testing for differential gene expression involving multiple serial analysis of gene expression (SAGE) libraries, it is critical to account for both between and with...
Usually unequal error protection schemes mostly focus on protecting video data parameters with unequal rates or levels depending on their sensitivities to errors. This paper first...
Clustering has been one of the most popular approaches used in gene expression data analysis. A clustering method is typically used to partition genes according to their similarity...