The dominant method for evaluating search engines is the Cranfield paradigm, but the existing metrics do not consider some modern search engines features, such as document snippets...
Given a single program (i.e., bit string), one may assume that the program's behaviors can be determined by first identifying the native runtime architecture and then executi...
Sang Kil Cha, Brian Pak, David Brumley, Richard Ja...
Abstract-- In this paper, a segmentation technique of multispectral magnetic resonance image of the brain using a new differential evolution based crisp clustering is proposed. Rea...
The objective of this work is to derive quantitative statements about what fraction of web search queries issued to the state-of-the-art commercial search engines lead to excellen...
Hugo Zaragoza, Berkant Barla Cambazoglu, Ricardo A...
We consider the problem of finding the best arm in a stochastic multi-armed bandit game. The regret of a forecaster is here defined by the gap between the mean reward of the optim...