Landslides are abundant in mountainous regions.They are responsible for substantial damages and losses in those areas.The A1 Highway,which is an important road in Algeria,was sometimes constructed in mountainous and/o...Landslides are abundant in mountainous regions.They are responsible for substantial damages and losses in those areas.The A1 Highway,which is an important road in Algeria,was sometimes constructed in mountainous and/or semi-mountainous areas.Previous studies of landslide susceptibility mapping conducted near this road using statistical and expert methods have yielded ordinary results.In this research,we are interested in how do machine learning techniques help in increasing accuracy of landslide susceptibility maps in the vicinity of the A1 Highway corridor.To do this,an important section at Ain Bouziane(NE,Algeria) is chosen as a case study to evaluate the landslide susceptibility using three different machine learning methods,namely,random forest(RF),support vector machine(SVM),and boosted regression tree(BRT).First,an inventory map and nine input factors were prepared for landslide susceptibility mapping(LSM) analyses.The three models were constructed to find the most susceptible areas to this phenomenon.The results were assessed by calculating the receiver operating characteristic(ROC) curve,the standard error(Std.error),and the confidence interval(CI) at 95%.The RF model reached the highest predictive accuracy(AUC=97.2%) comparatively to the other models.The outcomes of this research proved that the obtained machine learning models had the ability to predict future landslide locations in this important road section.In addition,their application gives an improvement of the accuracy of LSMs near the road corridor.The machine learning models may become an important prediction tool that will identify landslide alleviation actions.展开更多
Studying spatio-temporal evolution of epidemics can uncover important aspects of interaction among people, infectious diseases, and the environment, providing useful insights and modeling support to facilitate public ...Studying spatio-temporal evolution of epidemics can uncover important aspects of interaction among people, infectious diseases, and the environment, providing useful insights and modeling support to facilitate public health response and possibly prevention measures. This paper presents an empirical spatio-temporal analysis of epidemiological data concerning 2321 SARS-infected patients in Beijing in 2003. We mapped the SARS morbidity data with the spatial data resolution at the level of street and township. Two smoothing methods, Bayesian adjustment and spatial smoothing, were applied to identify the spatial risks and spatial transmission trends. Furthermore, we explored various spatial patterns and spatio-temporal evolution of Beijing 2003 SARS epidemic using spatial statistics such as Moran’s I and LISA. Part of this study is targeted at evaluating the effectiveness of public health control measures implemented during the SARS epidemic. The main findings are as follows. (1) The diffusion speed of SARS in the northwest-southeast direction is weaker than that in northeast-southwest direction. (2) SARS’s spread risk is positively spatially associated and the strength of this spatial association has experienced changes from weak to strong and then back to weak during the lifetime of the Beijing SARS epidemic. (3) Two spatial clusters of disease cases are identified: one in the city center and the other in the eastern suburban area. These two clusters followed different evolutionary paths but interacted with each other as well. (4) Although the government missed the opportunity to contain the early outbreak of SARS in March 2003, the response strategies implemented after the mid of April were effective. These response measures not only controlled the growth of the disease cases, but also mitigated the spatial diffusion.展开更多
文摘Landslides are abundant in mountainous regions.They are responsible for substantial damages and losses in those areas.The A1 Highway,which is an important road in Algeria,was sometimes constructed in mountainous and/or semi-mountainous areas.Previous studies of landslide susceptibility mapping conducted near this road using statistical and expert methods have yielded ordinary results.In this research,we are interested in how do machine learning techniques help in increasing accuracy of landslide susceptibility maps in the vicinity of the A1 Highway corridor.To do this,an important section at Ain Bouziane(NE,Algeria) is chosen as a case study to evaluate the landslide susceptibility using three different machine learning methods,namely,random forest(RF),support vector machine(SVM),and boosted regression tree(BRT).First,an inventory map and nine input factors were prepared for landslide susceptibility mapping(LSM) analyses.The three models were constructed to find the most susceptible areas to this phenomenon.The results were assessed by calculating the receiver operating characteristic(ROC) curve,the standard error(Std.error),and the confidence interval(CI) at 95%.The RF model reached the highest predictive accuracy(AUC=97.2%) comparatively to the other models.The outcomes of this research proved that the obtained machine learning models had the ability to predict future landslide locations in this important road section.In addition,their application gives an improvement of the accuracy of LSMs near the road corridor.The machine learning models may become an important prediction tool that will identify landslide alleviation actions.
基金supported by National Science Foundation of the United State (Grant Nos. IIS-0839990, IIS-0428241)Department of Homeland Security of the United State (Grant No. 2008-ST-061-BS0002)+4 种基金Important National Science & Technology Specific Projects (Grant Nos. 2009ZX10004- 315, 2008ZX10005-013)the Chinese Academy of Sciences (Grants Nos. 2F07C01, 2F08N03)China Postdoctoral Science Fund (Grant No. 20080440559)National High Technology Research and Development Pro-gram of China (Grant No. 2006AA010106)National Natural Science Foundation of China (Grant Nos. 60621001, 40901219, 90924302)
文摘Studying spatio-temporal evolution of epidemics can uncover important aspects of interaction among people, infectious diseases, and the environment, providing useful insights and modeling support to facilitate public health response and possibly prevention measures. This paper presents an empirical spatio-temporal analysis of epidemiological data concerning 2321 SARS-infected patients in Beijing in 2003. We mapped the SARS morbidity data with the spatial data resolution at the level of street and township. Two smoothing methods, Bayesian adjustment and spatial smoothing, were applied to identify the spatial risks and spatial transmission trends. Furthermore, we explored various spatial patterns and spatio-temporal evolution of Beijing 2003 SARS epidemic using spatial statistics such as Moran’s I and LISA. Part of this study is targeted at evaluating the effectiveness of public health control measures implemented during the SARS epidemic. The main findings are as follows. (1) The diffusion speed of SARS in the northwest-southeast direction is weaker than that in northeast-southwest direction. (2) SARS’s spread risk is positively spatially associated and the strength of this spatial association has experienced changes from weak to strong and then back to weak during the lifetime of the Beijing SARS epidemic. (3) Two spatial clusters of disease cases are identified: one in the city center and the other in the eastern suburban area. These two clusters followed different evolutionary paths but interacted with each other as well. (4) Although the government missed the opportunity to contain the early outbreak of SARS in March 2003, the response strategies implemented after the mid of April were effective. These response measures not only controlled the growth of the disease cases, but also mitigated the spatial diffusion.
