Parallel mining of association rules from text databases

摘要

In this paper, we propose a new algorithm named Parallel Multipass with Inverted Hashing and Pruning (PMIHP) for mining association rules between words in text databases. The characteristics of text databases are quite different from those of retail transaction databases, and existing mining algorithms cannot handle text databases efficiently because of the large number of itemsets (i.e., sets of words) that need to be counted. The new PMIHP algorithm is a parallel version of our Multipass with Inverted Hashing and Pruning (MIHP) algorithm (Holt, Chung in: Proc of the 14th IEEE int'l conf on tools with artificial intelligence, 2002, pp 49-56), which was shown to be quite efficient than other existing algorithms in the context of mining text databases. The PMIHP algorithm reduces the overhead of communication between miners running on different processors because they are mining local data-. bases asynchronously and prune the global candidates by using the Inverted Hashing and Pruning technique. Compared with the well-known Count Distribution algorithm (Agrawal, Shafer in: (1996) IEEE Trans Knowl Data Eng 8(6):962-969), PMIHP demonstrates superior performance characteristics for mining association rules in large text databases, and when the minimum support level is low, its speedup is superlinear as the number of processors increases. These experiments were performed on a cluster of Linux workstations using a collection of Wall Street Journal articles.