The Community for Technology Leaders
Green Image
<p><b>Abstract</b>—We formally define the Multiprocessor Document Allocation Problem (MDAP) and prove it to be computationally intractable (NP Complete). Once it is shown that MDAP is NP Complete, we describe a document allocation algorithm based on genetic algorithms. This algorithm assumes that the documents are clustered using any one of the many clustering techniques. We later show that our allocation algorithm probabilistically converges to a good solution. For a behavioral evaluation, we present sample experimental results.</p>
Genetic algorithms, information systems, information retrieval, multiprocessor, parallel processing, data placement, data allocation.
Ophir Frieder, Hava T. Siegelmann, "Multiprocessor Document Allocation: A Genetic Algorithm Approach", IEEE Transactions on Knowledge & Data Engineering, vol. 9, no. , pp. 640-642, July-August 1997, doi:10.1109/69.617055
90 ms
(Ver 3.3 (11022016))