Evolutionary Method to Discover Itemsets with Statistically Distinctive Backgrounds
Evolutionary Method to Discover Itemsets with Statistically Distinctive Backgrounds
复制标题
发现具有统计特征背景的项集的进化方法
DOI：
10.1109/aike52691.2021.00024
复制
发表时间：
2021
期刊：
影响因子：
--
通讯作者：
Shogo Matsuno
中科院分区：
文献类型：
--
作者：
Kaoru Shimada;Takaaki Arahira;Shogo Matsuno
在本文中，我们提出了一种方法发现的组合属性（项目集）的统计特征的背景下，不获得频繁项目集。该方法由具有众多属性的数据库组成，并且可以在不完整的数据库中直接找到两个连续的感兴趣变量中与小总体高度相关的属性组合。它确定大规模数据中的局部模式，并有望成为大规模数据分析的基础。它使用以网络结构和策略为特征的进化计算，以在整个世代中汇集解决方案。此外，它使用关联规则来推广的分析方法。用于分类的类关联规则是一种属性组合的发现方法，当类属性的比例被注意时，这是特征。所提出的方法可以定位为二元统计分析方法的扩展。评价实验结果表明了该方法的特点和有效性。
In this paper, we propose a method for discovering combinations of attributes (itemsets) against a background of statistical characteristics without obtaining frequent itemsets. The method consists of a database with numerous attributes and can directly find a combination of attributes that are highly correlated with small populations in two consecutive variables of interest in an incomplete database. It determines locally found patterns in large-scale data, and is expected to be the basis of large-scale data analysis. It uses evolutionary computations characterized by a network structure and a strategy to pool solutions throughout the generations. Moreover, it uses association rules to generalize the analysis method. The class-association rules used for classification are a discovery method of attribute combinations, which are characteristic when the ratio of class attributes is noted. The proposed method can be positioned as an extension to the statistical analysis method of the bivariate. The results of the evaluation experiment show the characteristics and effectiveness of the method.