决策树技术在成矿预测中的应用--以新疆中天山沙垄以西铅锌矿远景区优选为例  

Application of Decision-Making Tree Technology in Metallogenic Predication——Lead-zinc ore prospective preference in west Shalong in Tianshan Mountains of Xinjiang

作　　者：高景刚[1] 高云霞[1] 高凤亮[2] 梁婷[2] 胡长安 王晓虎[2] 

机构地区：[1]长安大学信息工程学院 [2]长安大学地球科学与国土资源学院,西安710054 [3]新疆地质矿产开发局第一地质大队,新疆昌吉831100

出　　处：《桂林工学院学报》2008年第1期15-19,共5页Journal of Guilin University of Technology

基　　金：中国地质调查局项目(200420190004);新疆地质矿产勘查开发局项目(XGMB2004001)

摘　　要：选取新疆中天山沙垄以西地区开展铅锌矿成矿预测研究,将决策树技术引入到成矿预测中。通过区域成矿规律分析,选取不同构造单元中数据较为完整、具有代表性的典型矿床,建立综合信息预测模型,采用决策树技术中的ID3算法构建决策树,在ERDAS IMAG ING 8.7平台上建立分类决策树,开展铅锌远景区预测。经野外验证,预测区与已知矿化区吻合度较高,证实该方法在成矿预测中是可行的。Lead-zinc ore metallogenic predication is introduced. In the west of Shalong of Tianshan mountains in Xinjiang, decision-making tree technology is applied into the metallogenic predication. In the analysis of the regional metallogenic regularity and the typical ore deposit with datum integral very much in different constructional units, general information predication model is set up by decision-making tree technology based on ERDAS IMAGING 8.7 to develop the lead-zinc ore potential region metallogenic predication. After field practice verification, it is found that the predication area and the known mineralizing area highly coincide. This method is feasible in metaUogenic predication.

关 键 词：铅锌矿 成矿预测 决策树 新疆 

分 类 号：P612[天文地球—矿床学] P208[天文地球—地质学]