论文信息 - Visual Clustering of Multidimensional and Large Data Sets Using Parallel Environments

Visual Clustering of Multidimensional and Large Data Sets Using Parallel Environments

A method for visual clustering of large N-dimensional data sets is presented briefly. Its implementation on HP/Convex SPP/1600 enables visualization of data sets consisting of more than 104 multidimensional data vectors. The method was tested in PVM, MPI and data parallel environments. In the paper, the authors compare the parallel algorithm performance for these three interfaces. The results of tests, made to exemplify the algorithm “immunity” from data errors, are presented and discussed.

Witold Dzwinel | Jan Blasiak | W. Dzwinel | J. Blasiak

[1] Jack Sklansky,et al. Experiments on mapping techniques for exploratory pattern analysis , 1988, Pattern Recognit..

[2] Witold Dzwinel,et al. Pattern recognition via molecular dynamics on vector supercomputers and networked workstations , 1995, HPCN Europe.

[3] Anil K. Jain,et al. Algorithms for Clustering Data , 1988 .

[4] Witold Dzwinel,et al. How to make sammon's mapping useful for multidimensional data structures analysis , 1994, Pattern Recognit..

[5] Witold Dzwinel,et al. Virtual particles and search for global minimum , 1997, Future Gener. Comput. Syst..

[6] Witold Dzwinel,et al. Comparison of noise diagnostic systems based on pattern recognition and discriminant methods , 1995 .

[7] Stefan Brode,et al. An optimized MD program for the vector computer cyber 205 , 1986 .

[8] Jack Sklansky,et al. An overview of mapping techniques for exploratory pattern analysis , 1988, Pattern Recognit..