欢迎您！东篱公司
退出
-
申报数据库
-
立项数据库
-
成果数据库
-
项目获奖数据库
中文摘要：
星座图是多元数据可视化的一种常用方法，具有直观、形象的特点，可以通过调整权系数来对数据进行交互式挖掘。但是传统的星座图缺乏自动调整权系数的较好方法，因而限制了其在可视化数据分析和模式识别的进一步应用。本文将传统的实系数星座图推广为复系数星座图，并且提出了基于复线性判别分析算法对星座图权系数进行自动优化的方法。对4个数据集的实验结果表明，复系数星座图可以较好地表达高维数据的结构关系，并且可以和有关机器算法结合对数据进行可视化分析。
英文摘要：
The constellation graph is a popular visualization method of multivariate data. This method has merits of straightforward, intuitive and can interaetively explore the structural characteristics of the data by adjusting the weights. One of the hard problems of the constellation graph is the lack of effective approach to determine the weights. In this paper, a visual analysis method called complex constellation graph is proposed, which allow the weights to be complex numbers and determine the weights automatically based on complex linear discriminant analysis of the data. The visualization results of for datasets indicates that the method reveal the structural relationship of original data well and can be used combining machine algorithms as a tool for visual data analysis.
同期刊论文项目
同项目期刊论文
期刊信息
位置：