Research on Similarity Classifier Based on Multivariate Graph Characteristic Enhancement Techniques
A novel method for the classifier based on radar chart combined similarity is proposed. The method uses the radar chart to show multivariate data,makes the different category data constitute the different pattern radar chart. After enhancing the graph characteristics which having prominent contribution to the classification, similarity classifier is used to recognize the radar chart, and automatic classification has been realized. In this paper we used ref. 1vegetables oil classified data to carry on the confirmation experiment, after training with 95 oil samples, the proposed method was able to ascribe correct classes to all samples. The assessment of the model by the leave-one-out cross validation which resulted in 100％ correct predictions for 95 samples of known type of oil
radar chart similarity classification characteristic enhancement
Wenxue Hong Fengxiang Chang Yonghong Xu Jialin Song
Biomedical Engineering department, Yanshan University,Qinhuangdao, 066004
国际会议
The International Colloquium on Onformation Fusion 2007(2007年国际信息融合研讨会)
西安
英文
307-310
2007-08-22（万方平台首次上网日期，不代表论文的发表时间）