6533b857fe1ef96bd12b4497
RESEARCH PRODUCT
PGAC: A Parallel Genetic Algorithm for Data Clustering
G. Lo Boscosubject
IntranetCorrectnessTheoretical computer scienceParallel processing (DSP implementation)Artificial neural networkData Clustering Evolutionary Aglorithms Parallel processingSettore INF/01 - InformaticaComputer scienceParallel algorithmA priori and a posterioriAlgorithm designParallel computingCluster analysisdescription
Cluster analysis is a valuable tool for exploratory pattern analysis, especially when very little a priori knowledge about the data is available. Distributed systems, based on high speed intranet connections, provide new tools in order to design new and faster clustering algorithms. Here, a parallel genetic algorithm for clustering called PGAC is described. The used strategy of parallelization is the island model paradigm where different populations of chromosomes (called demes) evolve locally to each processor and from time to time some individuals are moved from one deme to another. Experiments have been performed for testing the benefits of the parallelisation paradigm in terms of computation time and correctness of the solution.
| year | journal | country | edition | language |
|---|---|---|---|---|
| 2005-01-01 |