Programmers and users of compute intensive scientific applications often do not want to (or even cannot) code load balancing and fault tolerance into their programs. The PBEAM syst...
We developa clustereddithering methodthatusesstochasticscreening and is able to perform an adaptive variation of the cluster size. This makes it possible to achieve optimal rendit...
Information visualization is essential in making sense out of large data sets. Often, high-dimensional data are visualized as a collection of points in 2-dimensional space through...
Ambiguous person names are a problem in many forms of written text, including that which is found on the Web. In this paper we explore the use of unsupervised clustering techniques...