基于分类数据的可视化改善方法

The Visualization Improvement Method Based on Categorical Data

由于大型、复杂、多维的数据通常具有分类属性,而分类值没有内在的顺序,可视化的技术能以一个好的视觉展示这类数据,从而探索分析复杂问题的多个方面。以仿真的通讯公司网络告警数据为研究对象(该数据具有分类属性:告警名称,产生时间,产生告警设备编号),经过对数据聚类、排序等一系列预处理再进行可视化展示,达到改善可视化效果的目的。

For large, complex, multidimensional data usually has classification properties, classification and value without inner sequence, visual tech- nology to a good visual display this kind of data, so as to explore the analysis of several aspects of the complicated problem. Based on the simulation of communication network alarm data as the research object, the data has classification properties： the name of the alarm time, produce the alarm device number, through the data clustering, sorting, and a series of preprocessing and then to visual display, achieve the goal of improve visualization.