%0 Journal Article
%T 基于属性偏序结构图的关联规则提取方法<br>An Association Rule Extraction Method Based on Attribute Partial Order Structure Diagrams
%A 王秋婷
%A 王诗慧
%J Hans Journal of Data Mining
%P 112-120
%@ 2163-1468
%D 2021
%I Hans Publishing
%R 10.12677/HJDM.2021.112011
%X 针对Apriori算法生成大量冗余关联规则的问题，本文提出了一种基于属性偏序结构图的关联规则提取方法。该方法旨在寻找相同支持度下的最大频繁项目集，进而提取无冗余关联规则。本文提出的方法不仅减少了挖掘频繁项目集的数量，从而提高关联规则提取的效率，而且将关联规则转换成属性偏序结构图中的知识表示形式，实现了频繁项分层的关联规则可视化展示。具有较强的可读性，有助于用户对关联规则进行深入分析，提高对潜在知识的利用和发掘程度。<br />
Aiming at the problem of Apriori algorithm generating a large number of redundant association rules, this paper proposes an association rule extraction method based on attribute partial order structure diagrams. It can extract non-redundant association rules by finding the maximum fre-quent item sets under the same support. The method proposed in this paper can reduce the num-ber of mining frequent item sets, so as to improve the efficiency of extracting association rules. Moreover, the association rules are converted into the knowledge representation in the attribute partial order structure diagrams, which realizes the visualization of the association rules of the fre-quent item hierarchy. With strong readability, it is helpful for users to conduct in-depth analysis of association rules and improve the utilization and exploitation of potential knowledge.
%K Apriori算法，属性偏序结构图，无冗余关联规则，可视化<br>Apriori Algorithm
%K Attribute Partial Order Structure Diagrams
%K Non-Redundant Association Rules
%K Visualization
%U http://www.hanspub.org/journal/PaperInformation.aspx?PaperID=41972