This paper presents an procedure for purifying training data sets （i.e., past occurrences of slope failures） for inverse estimation on unobserved trigger factors of ＂different types of simultaneous slope failures＂...This paper presents an procedure for purifying training data sets （i.e., past occurrences of slope failures） for inverse estimation on unobserved trigger factors of ＂different types of simultaneous slope failures＂. Due to difficulties in pixel-by-pixel observations of trigger factors, as one of the measures, the authors had proposed an inverse analysis algorithm on trigger factors based on SEM （structural equation modeling）. Through a measurement equation, the trigger factor is inversely estimated, and a TFI （trigger factor influence） map can be also produced. As a subsequence subject, a purification procedure of training data set should be constructed to improve the accuracy of TFI map which depends on the representativeness of given training data sets of different types of slope failures. The proposed procedure resamples the matched pixels between original groups of past slope failures （i.e., surface slope failures, deep-seated slope failures, landslides） and classified three groups by K-means clustering for all pixels corresponding to those slope failures. For all cases of three types of slope failures, the improvement of success rates with respect to resampled training data sets was confirmed. As a final outcome, the differences between TFI maps produced by using original and resampled training data sets, respectively, are delineated on a DIF map （difference map） which is useful for analyzing trigger factor influence in terms of ＂risky- and safe-side assessment＂ sub-areas with respect to ＂different types of simultaneous slope failures＂.展开更多
Since the impoundment of Three Gorges Reservoir(TGR)in 2003,numerous slopes have experienced noticeable movement or destabilization owing to reservoir level changes and seasonal rainfall.One case is the Outang landsli...Since the impoundment of Three Gorges Reservoir(TGR)in 2003,numerous slopes have experienced noticeable movement or destabilization owing to reservoir level changes and seasonal rainfall.One case is the Outang landslide,a large-scale and active landslide,on the south bank of the Yangtze River.The latest monitoring data and site investigations available are analyzed to establish spatial and temporal landslide deformation characteristics.Data mining technology,including the two-step clustering and Apriori algorithm,is then used to identify the dominant triggers of landslide movement.In the data mining process,the two-step clustering method clusters the candidate triggers and displacement rate into several groups,and the Apriori algorithm generates correlation criteria for the cause-and-effect.The analysis considers multiple locations of the landslide and incorporates two types of time scales:longterm deformation on a monthly basis and short-term deformation on a daily basis.This analysis shows that the deformations of the Outang landslide are driven by both rainfall and reservoir water while its deformation varies spatiotemporally mainly due to the difference in local responses to hydrological factors.The data mining results reveal different dominant triggering factors depending on the monitoring frequency:the monthly and bi-monthly cumulative rainfall control the monthly deformation,and the 10-d cumulative rainfall and the 5-d cumulative drop of water level in the reservoir dominate the daily deformation of the landslide.It is concluded that the spatiotemporal deformation pattern and data mining rules associated with precipitation and reservoir water level have the potential to be broadly implemented for improving landslide prevention and control in the dam reservoirs and other landslideprone areas.展开更多
AIM To investigate potential triggering factors leading to acute liver failure(ALF) as the initial presentation of autoimmune hepatitis(AIH).METHODS A total of 565 patients treated at our Department between 2005 and 2...AIM To investigate potential triggering factors leading to acute liver failure(ALF) as the initial presentation of autoimmune hepatitis(AIH).METHODS A total of 565 patients treated at our Department between 2005 and 2017 for histologically-proven AIH were retrospectively analyzed. However, 52 patients(9.2%) fulfilled the criteria for ALF defined by the "American Association for the Study of the Liver(AASLD)". According to this definition, patients with "acute-on-chronic" or "acute-on-cirrhosis" liver failure were excluded. Following parameters with focus on potential triggering factors were evaluated: Patients' demographics, causation of liver failure, laboratory data(liver enzymes, MELD-score, autoimmune markers, virus serology), liver histology, immunosuppressive regime, and finally, outcome of our patients.RESULTS The majority of patients with ALF were female(84.6%) and mean age was 43.6 ± 14.9 years. Interestingly, none of the patients with ALF was positive for antiliver kidney microsomal antibody(LKM). We could identify potential triggering factors in 26/52(50.0%) of previously healthy patients presenting ALF as their first manifestation of AIH. These were drug-induced ALF(57.7%), virus-induced ALF(30.8%), and preceding surgery in general anesthesia(11.5%), respectively. Unfortunately, 6 out of 52 patients(11.5%) did not survive ALF and 3 patients(5.7%) underwent liver transplantation(LT). Comparing data of survivors and patients with non-recovery following treatment, MELDscore(P < 0.001), age(P < 0.05), creatinine(P < 0.01), and finally, ALT-values(P < 0.05) reached statistical significance. CONCLUSION Drugs, viral infections, and previous surgery may trigger ALF as the initial presentation of AIH. Advanced age and high MELD-score were associated with lethal outcome.展开更多
触发执行编程(Trigger-Action Programming,TAP)为用户联动物联网(Internet of Things,IoT)设备提供了便捷的编程范式。利用机器学习对用户已编辑的TAP规则进行分析,实现TAP规则推荐和生成等功能可以提升用户体验。但TAP规则可能包含个...触发执行编程(Trigger-Action Programming,TAP)为用户联动物联网(Internet of Things,IoT)设备提供了便捷的编程范式。利用机器学习对用户已编辑的TAP规则进行分析,实现TAP规则推荐和生成等功能可以提升用户体验。但TAP规则可能包含个人隐私信息,用户对上传和分享TAP信息存在顾虑。文章提出了基于联邦学习和区块链技术的TAP规则处理系统,用户可在本地进行TAP模型训练,无需上传隐私数据。为解决集中式服务器单点故障和防范恶意模型参数上传的问题,文章利用区块链技术改进集中式TAP联邦学习架构。用户将本地模型更新的累积梯度传输给区块链中的矿工,进行异常识别和交叉验证。矿工委员会整合正常用户提供的累积梯度,得到的全局模型作为一个新区块的数据,链接到区块链上,供用户下载使用。文章采用轻量级无监督的非负矩阵分解方法验证了提出的基于联邦学习和区块链的分布式学习架构的有效性。实验证明该联邦学习架构能有效保护TAP数据中的隐私,并且区块链中的矿工能够很好地识别恶意模型参数,确保了模型的稳定性。展开更多
Severe disasters caused by extreme precipitation events have attracted more and more attention. The relationship between climate change and extreme precipitation has become the hottest scientific frontier issue. The s...Severe disasters caused by extreme precipitation events have attracted more and more attention. The relationship between climate change and extreme precipitation has become the hottest scientific frontier issue. The study of daily torrential rain observations from 659 meteorological stations in China from 1951 to 2010 shows that rapid urbanization may have triggered a significant increase in heavy rains in China. It reached following conclusions: China’s interdecadal heavy rainfall amount,rainy days and rain intensity increased significantly,with an increase of 68. 71%,60. 15% and 11. 52%,respectively. The increase in the number of stations was 84. 22%,84. 22% and 54. 48%,respectively. It showed time change of " rapid-slow-rapid increase" and spatial change of gradual increase from southeastern coast to central China,southwest,north China,and northeastern regions. Rapid urbanization factors,including secondary industry output( GDP2),urban population ratio( UP),annual average haze days( HD),are likely to be the main causes of the increase in heavy rains in China. Their explanations of the variance of heavy rainfall amount( HRA),rainy day( RD) and rain intensity( RI) in China reached 61. 54%,58. 48% and 65. 54%,respectively,of which only the explanation of variance of heavy rainfall amount,rainy days and rain intensity was as high as 25. 93%,22. 98%and 26. 64%,respectively. However,explanation of variance of climatic factors including WPSH( West Pacific Subtropical High),ENSO( El Ni1 o-Southern Oscillation) AMO( Atlantic Interdecadal Oscillation),and AAO( Antarctic Oscillation) was only 24. 30%,26. 23%,and 21. 92%,respectively. Compared with the rapid urbanization forcing factor,the impact of these climatic factors was only one third of the former. The panel data of China’s county-level total population and annual average of visibility days were significantly correlated with China’s interdecadal heavy rainfall amount,rainy days and rain intensity. Their spatial correlation coefficient increased gradually from 1951-1960 to 2001-2010,that is,the total population of the county level increased from 0. 35,0. 36,and 0. 40 to 0. 54,0. 55,and 0. 58,respectively.The annual average of visibility days increased from 0. 36,0. 38,and 0. 48 to 0. 55. 0. 57,0. 58,further indicating that rapid urbanization triggered a significant increase in interdecadal large-area heavy rains in China.展开更多
文摘This paper presents an procedure for purifying training data sets （i.e., past occurrences of slope failures） for inverse estimation on unobserved trigger factors of ＂different types of simultaneous slope failures＂. Due to difficulties in pixel-by-pixel observations of trigger factors, as one of the measures, the authors had proposed an inverse analysis algorithm on trigger factors based on SEM （structural equation modeling）. Through a measurement equation, the trigger factor is inversely estimated, and a TFI （trigger factor influence） map can be also produced. As a subsequence subject, a purification procedure of training data set should be constructed to improve the accuracy of TFI map which depends on the representativeness of given training data sets of different types of slope failures. The proposed procedure resamples the matched pixels between original groups of past slope failures （i.e., surface slope failures, deep-seated slope failures, landslides） and classified three groups by K-means clustering for all pixels corresponding to those slope failures. For all cases of three types of slope failures, the improvement of success rates with respect to resampled training data sets was confirmed. As a final outcome, the differences between TFI maps produced by using original and resampled training data sets, respectively, are delineated on a DIF map （difference map） which is useful for analyzing trigger factor influence in terms of ＂risky- and safe-side assessment＂ sub-areas with respect to ＂different types of simultaneous slope failures＂.
基金supported by the Natural Science Foundation of Shandong Province,China(Grant No.ZR2021QD032)。
文摘Since the impoundment of Three Gorges Reservoir(TGR)in 2003,numerous slopes have experienced noticeable movement or destabilization owing to reservoir level changes and seasonal rainfall.One case is the Outang landslide,a large-scale and active landslide,on the south bank of the Yangtze River.The latest monitoring data and site investigations available are analyzed to establish spatial and temporal landslide deformation characteristics.Data mining technology,including the two-step clustering and Apriori algorithm,is then used to identify the dominant triggers of landslide movement.In the data mining process,the two-step clustering method clusters the candidate triggers and displacement rate into several groups,and the Apriori algorithm generates correlation criteria for the cause-and-effect.The analysis considers multiple locations of the landslide and incorporates two types of time scales:longterm deformation on a monthly basis and short-term deformation on a daily basis.This analysis shows that the deformations of the Outang landslide are driven by both rainfall and reservoir water while its deformation varies spatiotemporally mainly due to the difference in local responses to hydrological factors.The data mining results reveal different dominant triggering factors depending on the monitoring frequency:the monthly and bi-monthly cumulative rainfall control the monthly deformation,and the 10-d cumulative rainfall and the 5-d cumulative drop of water level in the reservoir dominate the daily deformation of the landslide.It is concluded that the spatiotemporal deformation pattern and data mining rules associated with precipitation and reservoir water level have the potential to be broadly implemented for improving landslide prevention and control in the dam reservoirs and other landslideprone areas.
文摘AIM To investigate potential triggering factors leading to acute liver failure(ALF) as the initial presentation of autoimmune hepatitis(AIH).METHODS A total of 565 patients treated at our Department between 2005 and 2017 for histologically-proven AIH were retrospectively analyzed. However, 52 patients(9.2%) fulfilled the criteria for ALF defined by the "American Association for the Study of the Liver(AASLD)". According to this definition, patients with "acute-on-chronic" or "acute-on-cirrhosis" liver failure were excluded. Following parameters with focus on potential triggering factors were evaluated: Patients' demographics, causation of liver failure, laboratory data(liver enzymes, MELD-score, autoimmune markers, virus serology), liver histology, immunosuppressive regime, and finally, outcome of our patients.RESULTS The majority of patients with ALF were female(84.6%) and mean age was 43.6 ± 14.9 years. Interestingly, none of the patients with ALF was positive for antiliver kidney microsomal antibody(LKM). We could identify potential triggering factors in 26/52(50.0%) of previously healthy patients presenting ALF as their first manifestation of AIH. These were drug-induced ALF(57.7%), virus-induced ALF(30.8%), and preceding surgery in general anesthesia(11.5%), respectively. Unfortunately, 6 out of 52 patients(11.5%) did not survive ALF and 3 patients(5.7%) underwent liver transplantation(LT). Comparing data of survivors and patients with non-recovery following treatment, MELDscore(P < 0.001), age(P < 0.05), creatinine(P < 0.01), and finally, ALT-values(P < 0.05) reached statistical significance. CONCLUSION Drugs, viral infections, and previous surgery may trigger ALF as the initial presentation of AIH. Advanced age and high MELD-score were associated with lethal outcome.
文摘触发执行编程(Trigger-Action Programming,TAP)为用户联动物联网(Internet of Things,IoT)设备提供了便捷的编程范式。利用机器学习对用户已编辑的TAP规则进行分析,实现TAP规则推荐和生成等功能可以提升用户体验。但TAP规则可能包含个人隐私信息,用户对上传和分享TAP信息存在顾虑。文章提出了基于联邦学习和区块链技术的TAP规则处理系统,用户可在本地进行TAP模型训练,无需上传隐私数据。为解决集中式服务器单点故障和防范恶意模型参数上传的问题,文章利用区块链技术改进集中式TAP联邦学习架构。用户将本地模型更新的累积梯度传输给区块链中的矿工,进行异常识别和交叉验证。矿工委员会整合正常用户提供的累积梯度,得到的全局模型作为一个新区块的数据,链接到区块链上,供用户下载使用。文章采用轻量级无监督的非负矩阵分解方法验证了提出的基于联邦学习和区块链的分布式学习架构的有效性。实验证明该联邦学习架构能有效保护TAP数据中的隐私,并且区块链中的矿工能够很好地识别恶意模型参数,确保了模型的稳定性。
基金Supported by Project of National Natural Science Foundation of China(41801064)China Postdoctoral Science Foundation(2019T120114+1 种基金2019M650756)Central Asian Atmospheric Science Research Fund(CAAS201804)
文摘Severe disasters caused by extreme precipitation events have attracted more and more attention. The relationship between climate change and extreme precipitation has become the hottest scientific frontier issue. The study of daily torrential rain observations from 659 meteorological stations in China from 1951 to 2010 shows that rapid urbanization may have triggered a significant increase in heavy rains in China. It reached following conclusions: China’s interdecadal heavy rainfall amount,rainy days and rain intensity increased significantly,with an increase of 68. 71%,60. 15% and 11. 52%,respectively. The increase in the number of stations was 84. 22%,84. 22% and 54. 48%,respectively. It showed time change of " rapid-slow-rapid increase" and spatial change of gradual increase from southeastern coast to central China,southwest,north China,and northeastern regions. Rapid urbanization factors,including secondary industry output( GDP2),urban population ratio( UP),annual average haze days( HD),are likely to be the main causes of the increase in heavy rains in China. Their explanations of the variance of heavy rainfall amount( HRA),rainy day( RD) and rain intensity( RI) in China reached 61. 54%,58. 48% and 65. 54%,respectively,of which only the explanation of variance of heavy rainfall amount,rainy days and rain intensity was as high as 25. 93%,22. 98%and 26. 64%,respectively. However,explanation of variance of climatic factors including WPSH( West Pacific Subtropical High),ENSO( El Ni1 o-Southern Oscillation) AMO( Atlantic Interdecadal Oscillation),and AAO( Antarctic Oscillation) was only 24. 30%,26. 23%,and 21. 92%,respectively. Compared with the rapid urbanization forcing factor,the impact of these climatic factors was only one third of the former. The panel data of China’s county-level total population and annual average of visibility days were significantly correlated with China’s interdecadal heavy rainfall amount,rainy days and rain intensity. Their spatial correlation coefficient increased gradually from 1951-1960 to 2001-2010,that is,the total population of the county level increased from 0. 35,0. 36,and 0. 40 to 0. 54,0. 55,and 0. 58,respectively.The annual average of visibility days increased from 0. 36,0. 38,and 0. 48 to 0. 55. 0. 57,0. 58,further indicating that rapid urbanization triggered a significant increase in interdecadal large-area heavy rains in China.
