【摘要】 提出一种多维数据多元图图形基元和特征基元提取和表示的方法。首先应用多维数据多元图表示原理实现无结构数据的结构化表示,然后在建立基于多元图图形基元和特征基元表示的一般化表示模型基础上,提取出了表征多元图图形的图形基元和特征基元。为验证该方法的正确性,采用UCI机器学习数据库中的Iris等数据进行了分类实验,不同分类方法实验结果对比证明该方法具有较好的识别效果。更多还原

【Abstract】 A novel method for pattern recognition is introduced, which isbased on graph and feature primitives presentation of mul- tivariate graph. Firstly, the structural information of multivariate should be displayed based on multivariate graphical presentation. Then the graph primitives and feature primitives of multivariate graph is extracted. Finally, the object is classified based on the primitives ofmultivariategraph information. Some dataexperimentsusing theIrisdatasetofthe UCI repository of machine learning databases are accomplished, and it is shown that this method achieves better recognition result.更多还原