一种基于图形处理器的频繁模式挖掘算法.doc

文档介绍：一种基于图形处理器的频繁模式挖掘算法*
白洪涛1,2,欧阳丹彤1,2,何丽莉1,2
(1 吉林大学计算机科学与技术学院长春 130012;
2 吉林大学符号计算与知识工程教育部重点实验室长春 130012)
摘要:频繁模式挖掘是数据挖掘的核心问题。传统上,频繁模式并行挖掘主要是在集群上进行的,较少涉及共享内存多处理系统上的并行挖掘。基于广度优先搜索和直接计数策略研究了一种并行挖掘方法,并在图形处理器(graphics processing unit, GPU)pute unified device architecture)下进行了实现。GPU-based FPMA用CPU控制搜索进程;在GPU的多处理器上,采用数据划分的计算策略,以适合GPU的顺序数据流方式计数,并根据候选项的长度动态剪枝事务数据集。实验结果表明,GPU-based FPMA比CPU版本平均加速了10倍以上。
关键词:关联规则;频繁模式;图形处理器;并行计算;统一计算设备架构
中图分类号: 文献标识码:A 国家标准学科分类代码:
GPU-based frequent pattern mining algorithm
Bai Hongtao1,2, Ouyang Dantong1,2, He Li1i1,2
(1 College puter Science and Technology, Jilin University, Changchun 130012, China; 2 Key Laboratory of putation and Knowledge Engineering of Ministry of Education, Jilin University, Changchun 130012, China)
Abstract:Frequent pattern mining is an important issue in data mining area. Traditionally, parallel frequent pattern mining is carried out in PC clusters, and seldom related to multi-processors or massive cores with shared memories. In this paper, we propose a parallel frequent pattern mining algorithm suitable for GPU (graphics processing unit) based on width search and direct support strategy. It is implemented pute unified device architecture (CUDA) of GPU. In this algorithm, CPU takes charge of search process and GPU is responsible for counting using data partition. In addition, transactions are dynamically pr