Partition-Based Frequent Closed Pattern Miner

被引：0

作者：

Soni, Anu ^{[1
]}

Goel, Mukta ^{[1
]}

Goel, Rohit ^{[1
]}

机构：

[1] Technol Inst Text & Sci, Bhiwani, India

来源：

PROCEEDINGS OF INTERNATIONAL CONFERENCE ON ICT FOR SUSTAINABLE DEVELOPMENT ICT4SD 2015, VOL 2 | 2016年 / 409卷

关键词：

Frequent closed pattern; Mining algorithm; Parallel mining;

D O I：

10.1007/978-981-10-0135-2_45

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

Frequent closed pattern (FCP) mining has been an important step in data mining research. This paper introduces an algorithm to deal with the problem of finding out (FCP) from a given set of transactions. The miner works on a parallel approach based on compact matrix division to partition the data set. To filter these subtasks two methods are adopted (1) transaction set redundancy removal method and (2) itemset redundancy removal method. Mining of filtered subtasks are done separately. Consolidated result obtained from mining of these filtered independent partitions show all FCPs present in transactions.

引用

页码：459 / 470

页数：12