摘要
模式识别技术包括主成分分析和非线性映射法,被应用于矿产预报。湖南南部的不同矿点。特征变量是化学成分Cu、Sn、Mo、As、Sb、Ni、Pb以及物理和地质变量。训练样本的分类正确率达100％。据此模型预报了26个矿点为铅矿区。
Pattern recognition technique including principal component analysis, non-linear mapping has been applied to the prediction of mineral resources. The samples are taken from different prospects in the south of Hunan. The distinct variation are their chemical compositions such as Cu、Sn、Mo、As、Sb、Ni、Pb etc, as well as their physical and geological characteristics. For training samples their correct classification rate is 100%. Total 26 targets are predicted as potential Pb ore-field.
出处
《矿产与地质》
1990年第4期64-68,共5页
Mineral Resources and Geology
