Parallel Algorithm for Frequent Itemset Mining on Intel Many-core Systems

TL;DR

This paper presents a parallel implementation of the Dynamic Itemset Counting (DIC) algorithm for frequent itemset mining on Intel Xeon Phi many-core systems, achieving improved performance and scalability by exploiting thread-level parallelism and vectorization.

Contribution

The paper introduces a novel parallel implementation of DIC optimized for many-core systems using OpenMP and bit-based data layouts, enhancing efficiency and scalability.

Findings

Good performance on Intel Xeon and Xeon Phi platforms

Scalability with large synthetic and real databases

Effective use of vectorization and thread parallelism

Abstract

Frequent itemset mining leads to the discovery of associations and correlations among items in large transactional databases. Apriori is a classical frequent itemset mining algorithm, which employs iterative passes over database combining with generation of candidate itemsets based on frequent itemsets found at the previous iteration, and pruning of clearly infrequent itemsets. The Dynamic Itemset Counting (DIC) algorithm is a variation of Apriori, which tries to reduce the number of passes made over a transactional database while keeping the number of itemsets counted in a pass relatively low. In this paper, we address the problem of accelerating DIC on the Intel Xeon Phi many-core system for the case when the transactional database fits in main memory. Intel Xeon Phi provides a large number of small compute cores with vector processing units. The paper presents a parallel implementation of DIC based on OpenMP technology and thread-level parallelism. We exploit the bit-based internal layout for transactions and itemsets. This technique reduces the memory space for storing the transactional database, simplifies the support count via logical bitwise operation, and allows for vectorization of such a step. Experimental evaluation on the platforms of the Intel Xeon CPU and the Intel Xeon Phi coprocessor with large synthetic and real databases showed good performance and scalability of the proposed algorithm.