基于粗糙集流向图的多维定性数据分析方法

Multidimensional Qualitative Data Analysis Based on the Rough Set Flow Graphs

基于现有多维定性数据分析方法对海量数据分析时存在的局限性,提出采用流向图理论进行多维定性数据的交叉分类分析。以图形的方式表示多个定性变量间的相互关联,并从数量上对定性数据间存在的模式进行推理,克服了高维列联表难以表示的困境,并进行了实例分析。

In view of the limitations of the existing multidimensional qualitative data analysis methods in dealing with massive data, this paper proposes to use the flow graphs theory for the multi-dimensional qualitative data cross-classification analysis. The inter-relations between qualitative variables are expressed with graphs and the pattern related with the qualitative data is analyzed based on inference from the values of the graphs, which can overcome the difficulty in using the high-dimensional contingency tables. Examples are given.