摘要
现代统计模式识别以数据满足一定的统计分布规律(一般为正态分布)为前提。然而现实问题研究中存在大量不满足任何已知统计模型的情况,同时也有很多小样本情况,以上显然不适合统计方法。本文提出对非高斯信息进行基于多元数据多元图表示原理的可视化模式识别方法,并通过算法实现多元数据多元图分析过程的客观化和自动化,最后基于UCI数据对该方法进行了数据实验。
There are many real-world problems the sample is not obey the known Statistical models. It can＇t be fauhlessly solved by statistical approaches. Applied the non-statistical techniques in PR technology, we can avoid the distortion and aberration of information because the statistical condition was insufficiency. The mannual non-statistical PR technique Based on graphical representation of multivariate data is introduced. The automatization of mannual method is realized based on the algorithm of sorted overlap coefficient. In the data experiments base on the UCI databases, we achieved better performance.
出处
《微计算机信息》
2010年第4期38-40,共3页
Control & Automation
基金
基金申请人:徐永红
项目名称:一种基于多元数据多元图形特征表示原理的模式识别新方法研究
基金颁发部门:国家自然科学基金委(60605006)
关键词
模式识别
多元可视化
多元图表示
非高斯
类间重叠系数
Pattern recognition
Multivariate visualization
graphical representation of multivariate data
Non-Gaussian
sorted overlap coefficient
