Research On Icon Visualization Technology Used In Clustering Algorithm Method

With the advent of the era of data, large data mining and visualization has become a hotspot for information technology development. Data visualization is theory which use computer graphics and image processing technology to convert data into graphics or images displayed on the screen and interact with users. Icon-based visualization methods is a common visualization method, which uses a single icon to express multivariate data and represent different attributes with different visual elements. With its simple and efficient features, it has been widely used in the fields of data mining.Firstly, this article introduces the basic concept of data visualization and common algorithms of data visualization. To lay the foundation for Research, we give a detailed description of the icons visualization technology. Secondly, it introduces the common data mining clustering algorithms and explains the advantages and disadvantages of each algorithm. Then focuses on the evaluation algorithm for K-means algorithm. Finally, give a new visualization method which can not only set the initial parameters of K-means algorithm, but also evaluate the effect of K-means clustering algorithm.PCA (Principal component analysis) is a commonly methods which used in the field of data mining and the field of data analysis. When the dataset has large amount of attributes, common methods can’t show lots of the details. In order to solve this problem, we propose a visualization method based on PCA and icon based technique. This method not only reduce the dimensions of the dataset set but also use different colors of rectangle icons to display multivariate dataset. Experiments show that this method can not only effectively display complex data relationships and information in a reasonable space layout, but also help setting parameters of K-means clustering algorithm more quickly and appraisal the results of K-means clustering algorithm to get better results.