高级检索

    一种基于平行坐标的度量模型及其应用

    A Parallel Coordinates-Based Measure Model and Its Applications

    • 摘要: 分析了数据挖掘中可视化技术应用的特点与方法,给出了数据挖掘中可视对象与参数的确定及算法分解的方法,提出了一种基于平行坐标技术的度量指标体系,证明了其中的相关性质与结论,并给出基于平行坐标技术的度量模型以及在实验数据上的应用例.结果表明这种方法对于数据挖掘中的数据可视化表示是有效的.基于度量指标的可视化技术在对可视对象的分析处理上可以借助适用的数学方法建模与评测,这有助于数据挖掘可视化的研究与应用.

       

      Abstract: To apply visualization in data mining, or to establish visible data mining method is a cross research subject about visualization and data mining. This type of research is required to be established on reasonable acknowledge basement. On one hand, it requires to analyse the theory and technology basement of this method; on the other hand, it also requires to consider the visualization characteristics of the property of data mining subjects and the observers awareness of visualization characteristics. Two aspects are mainly needed to be considered during applying visualization to data mining. One is the separability of the mining algorithm process, that is, to split the process of the mining algorithm to inaffect the result of the data mining. The other is to determine the key factors in the mining algorithm and measuring standard, and find out their influences to the result of data mining. In this paper, the characteristic and method of visualization techniques applications are analyzed. The method of determining visualization data object and resolution of data mining algorithm is proposed. This paper proposes a parallel coordinates-based measurement index system and a measure model, proves some related properties and conclusion. Finally, it gives an application case. The results show that the methods are simple and valid for visualization of data in data mining. In analysing and processing the visible objects, visualization techniques based on measuring index can get help from proper mathematic method to model and evaluate, contributing to the research and application of visualization of data mining.

       

    /

    返回文章
    返回