LEARNING CONCEPTS IN PARALLEL BASED UPON THE STRATEGY OF VERSION SPACE

In this paper, we have attempted to apply the technique of parallel pr ocessing to concept learning. A parallel version-space learning algori thm based upon the principle of divide-and-conquer is proposed. Its ti me complexity is analyzed to be O(k log(2) n) with n processors, where n is the number of given training instances and k is a coefficient de pending on application domains. For a bounded number of processors in the real situations, a modified parallel learning algorithm is then pr oposed. Experimental results are then performed on a real learning pro blem, showing our parallel learning algorithm works and being quite co nsistent with results of theoretic analysis. We have finally concluded that when the number of training instances is large, it is worth lear ning in parallel because of its faster execution.