摘要
以浙江庆元县境内滑坡发生集中区为试验区,获取与滑坡危险性有关的因素集,包括数字高程、坡度、对应分析因子、地质图等,采用决策树算法对因素集数据进行数据挖掘,开展滑坡危险性区划,取得了与实际较为一致的结果.因此,利用决策树归纳技术可以快速准确地实现滑坡灾害区域评价与预测,是一种较好的滑坡灾害危险性评价方法.
The landslides take place frequently in Qingyuan, Zhejiang Province and account each year for enormous property damage in terms of both direct and indirect costs. Image-map database related to landslides including digital elevation map, the slop map, the component data of correspondence analysis and the geological map are set up. The landslide hazard assessment are obtained according to the analysis of the image-map database with decision tree algorithm. The result is close to actual situation. So the technology of inductive learning can be used in landslide hazard assessment and prediction very well.
出处
《浙江大学学报（理学版）》
CAS
CSCD
2004年第4期465-470,共6页
Journal of Zhejiang University（Science Edition）
基金
国家自然科学基金资助项目(40072096)
教育部高校优秀青年教师科研与教学奖励计划资助项目.
