一种新的面向普通用户的多值属性关联规则可视化挖掘方法

摘要

Considering the problems aroused by the traditional association rules visualization mining methods which are lack of dealing with multi-valued attribute data ,especially not conducive to expressing the frequent pattern between items and represent-ing multi-schema association rules ,this paper ,which presents the redefinition and classification of multi-valued attribute data by us-ing conceptual lattice ,proposes an improvement of Apriori algorithm based on the KAF factor and the CHF factor to mine multi-val-ued attribute association rules as well as introduces a novel visualizing approach for multi-valued association rules based on concept lattice ,and establishes a complete mining course parameters adjustment strategy acting very well in improving the speed and effi-ciency of mining algorithm ,which is convenient for users to select key attribute values to mine and analyze rules .This methodology organically organizes the multi-valued attribute data with concept lattice structure ,which has achieved frequent itemset visualization and multi-schema visualization of association rules .The experimental results turn out that the improved mining algorithm has a better performance and the schema has much excellent visual effects for multi-schema association rules visualization .%针对传统关联规则可视化挖掘方法不利于处理多值属性数据、缺乏展现数据间的频繁模式和关联模式以及效率低下等问题，提出了基于KAF因子和CHF因子的Apriori改进算法进行多值属性关联规则挖掘，实现了一种新的基于概念格的多值属性关联规则可视化方法。运用概念格理论对多值属性数据进行了重新定义和分类，建立了较为完整的挖掘过程参数调整策略，方便用户选择关键属性值进行规则挖掘分析，提高了算法运行速度和挖掘效率。以概念格结构将多值数据组织起来，实现了对频繁项集的可视化展示，以及关联规则的多模式可视化展示。实验结果表明，改进后的挖掘算法具有更好的性能，所提出的可视化形式和已有成果相比具有良好的展现效果。