A Parallel Coordinates-Based Measure Model and Its Applications

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.