期刊
COMPUTERS & GEOSCIENCES
卷 37, 期 9, 页码 1311-1317出版社
PERGAMON-ELSEVIER SCIENCE LTD
DOI: 10.1016/j.cageo.2011.02.010
关键词
Landslides; Spatial information; Data mining; Rough set theory
资金
- National Natural Science Foundation of China (NSFC) [40772170]
- Natural Science Foundation of Hubei [2009CDA007]
- Huazhong University of Science and Technology
Landslide incidence can be affected by a variety of environmental factors. Past studies have focused on the identification of these environmental factors, but most are based on statistical analysis. In this paper, spatial information techniques were applied to a case study of landslide occurrence in China by combining remote sensing and geographical information systems with an innovative data mining approach (rough set theory) and statistical analyses. Core and reducts of data attributes were obtained by data mining based on rough set theory. Rules for the impact factors, which can contribute to landslide occurrence, were generated from the landslide knowledge database. It was found that all 11 rules can be classified as both exact and approximate rules. In terms of importance, three main rules were then extracted as the key decision-making rules for landslide predictions. Meanwhile, the relationship between landslide occurrence and environmental factors was statistically analyzed to validate the accuracy of rules extracted by the rough set-based method. It was shown that the rough set-based approach is of use in analyzing environmental factors affecting landslide occurrence, and thus facilitates the decision-making process for landslide prediction. (C) 2011 Elsevier Ltd. All rights reserved.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据