Kernel-based methods work by embedding the data into a feature space and then searching linear hypothesis among the embedding data points. The performance is mostly affected by which kernel is used. A promising way is...Kernel-based methods work by embedding the data into a feature space and then searching linear hypothesis among the embedding data points. The performance is mostly affected by which kernel is used. A promising way is to learn the kernel from the data automatically. A general regularized risk functional （RRF） criterion for kernel matrix learning is proposed. Compared with the RRF criterion, general RRF criterion takes into account the geometric distributions of the embedding data points. It is proven that the distance between different geometric distdbutions can be estimated by their centroid distance in the reproducing kernel Hilbert space. Using this criterion for kernel matrix learning leads to a convex quadratically constrained quadratic programming （QCQP） problem. For several commonly used loss functions, their mathematical formulations are given. Experiment results on a collection of benchmark data sets demonstrate the effectiveness of the proposed method.展开更多
Single-value prediction such as the End of Life and Remaining Useful Life is a common method of estimating the lifetime of Li-ion batteries.Information from such prediction is limited when the entire degradation patte...Single-value prediction such as the End of Life and Remaining Useful Life is a common method of estimating the lifetime of Li-ion batteries.Information from such prediction is limited when the entire degradation pattern is needed for practical applications such as dynamic adjustment of battery warranty,improved maintenance scheduling,and battery stock management.In this research,a predictive,semi-parametric survival model called the Cox Proportional Hazards is proposed for the prediction of cell degradation in the form of survival probability(battery reliability)and cumulative hazard(battery risk)functions.Once this model is trained,the two functions can be obtained directly for a new cell without having to predict several cogent points.The model is trained on the first 50 cycles of only the voltage profile from either the charge or discharge data regime,implying that our methodology is data region agnostic.The signature method with both desirable mathematical and machine learning properties was adopted as a feature extraction technique.展开更多
Sexual dysfunction and prostate cancer are common among older men. Few studies explored the association between these two illnesses. We examined whether sexual function is associated with prostate cancer risk among ol...Sexual dysfunction and prostate cancer are common among older men. Few studies explored the association between these two illnesses. We examined whether sexual function is associated with prostate cancer risk among older men. Among 448 men undergoing prostate biopsy at the Durham Veterans Affairs Hospital, sexual function was ascertained from the Expanded Prostate Cancer Index Composite sexual assessment. We tested the link between sexual function and prostate cancer risk adjusting for multiple demographic and clinical characteristics using logistic regression. Multinomial logistic regression was used to test the associations with risk of low-grade （Gleason 〈6） and high-grade （Gleason 〉7 or 〉4 ＋ 3） disease versus no cancer. Of 448 men, 209 （47%） had a positive biopsy; these men were less likely to be white （43% vs 55%, P=0.013）, had higher prostate-specific antigen （PSA） （6.0 vs 5.4 ng ml-1, P 〈 0.001）, but with lower mean sexual function score （47 vs 54, P = 0.007）. There was no difference in age, BMI, pack years smoked, history of heart disease and/or diabetes. After adjusting for baseline differences, sexual function was linked with a decreased risk of overall prostate cancer risk （OR. 0.91 per lO-point change in sexual function, P= 0.004） and high-grade disease whether defined as Gleason ≥7 （OR： 0.86, P= 0.001） or 〉4 ＋ 3 （OR： 0.85, P= 0.009）. Sexual function was unrelated to low-grade prostate cancer （OR： 0.94, P = 0.13）. Thus, among men undergoing prostate biopsy, higher sexual function was associated with a decreased risk of overall and high-grade prostate cancer. Confirmatory studies are needed.展开更多
According to the characteristics of the correlation of multiple wind farm output, this paper put forwards a modeling method based on fuzzy c-means clustering and the copula function, and correlation wind farms are ins...According to the characteristics of the correlation of multiple wind farm output, this paper put forwards a modeling method based on fuzzy c-means clustering and the copula function, and correlation wind farms are inserted into IEEE-RTS79 reliability system for risk assessment. By the probabilistic load flow calculated by Monte Carlo simulation method, the probability of the accident is derived, and bus voltage and branch power flow overload risk index are defined in this paper. The results show that this method can realize the modeling of the correlation of wind power output, and the risk index can identify the weakness of the system, which can provide reference for the operation and maintenance personnel.展开更多
There are many uncertain factors,such as stochastic,fuzzy and gray information in the risk analysis on natural hazard.The set pair analysis(SPA)deals effectively with the various uncertain factors contributing to eval...There are many uncertain factors,such as stochastic,fuzzy and gray information in the risk analysis on natural hazard.The set pair analysis(SPA)deals effectively with the various uncertain factors contributing to evaluation of the risk level of natural disasters.The evaluation indicators and standards of natural disasters risk are analyzed by identity-discrepancycontrary(IDC).The result,the connection numbers,still has uncertainty information.Thus,yielding the risk evaluation model of natural disasters based on connection function,which construct a set pair relation between the indicators of connection numbers and comprehensive evaluation standards,and describe uncertainty of the connection numbers by using connection function.The study showed that the proposed model takes into account only the uncertainty of risk evaluation indicator identification on natural disasters,also the uncertainty of result connection numbers.This approach gives full consideration to the uncertainty of systematic evaluation process along with the actual meaning of comprehensive evaluation functions.Therefore,this means it is able to reduce the uncertainty of final evaluation results and improve the accuracy and reasonability of evaluation results.This model is capable of reflecting actual situation of the risk evaluation on natural disaster affected by various uncertain factors and has a promotional value in the natural disaster risk assessment.展开更多
This article considers a Markov-dependent risk model with a constant dividend barrier. A system of integro-differential equations with boundary conditions satisfied by the expected discounted penalty function, with gi...This article considers a Markov-dependent risk model with a constant dividend barrier. A system of integro-differential equations with boundary conditions satisfied by the expected discounted penalty function, with given initial environment state, is derived and solved. Explicit formulas for the discounted penalty function are obtained when the initial surplus is zero or when all the claim amount distributions are from rational family. In two state model, numerical illustrations with exponential claim amounts are given.展开更多
Estimation of the bivariate survival function under the competing risks caseis considered.We give an explicit formula for the estimator from a decomposition of thebivariate survival function based on competing risks,w...Estimation of the bivariate survival function under the competing risks caseis considered.We give an explicit formula for the estimator from a decomposition of thebivariate survival function based on competing risks,which is almost sure consistent.展开更多
Urbanization significantly increases the risk of urban flooding. Therefore, quantitative study of urban rainfall-runoff processes can provide a scientific basis for urban planning and management. In this paper, the bu...Urbanization significantly increases the risk of urban flooding. Therefore, quantitative study of urban rainfall-runoff processes can provide a scientific basis for urban planning and management. In this paper, the built-up region within the Fifth Ring Road of Beijing was selected as the study area. The details of land cover and urban function zones(UFZs) were identified using GIS and RS methods. On this basis, the SCS-CN model was adopted to analyze the rainfall-runoff risk characteristics of the study area. The results showed that:(1) UFZs within different levels of runoff risk varied under different rainfall conditions. The area ratio of the UFZs with high runoff risk increased from 18.90%(for rainfall return period of 1 a) to 54.74%(for period of 100 a). Specifically, urban commercial areas tended to have the highest runoff risk, while urban greening spaces had the lowest.(2) The spatial characteristics of the runoff risks showed an obvious circular distribution. Spatial cluster areas with high runoff risk were mainly concentrated in the center of the study area, while those with low runoff risk were mainly distributed between the fourth and fifth ring roads. The results indicated that the spatial clustering characteristic of urban runoff risk and runoff heterogeneity among different UFZs should be fully considered during urban rainwater management.展开更多
In order to investigate the association of fibrin monomer polymerization function (FMPF) with traditional cerebrovascular risk factors and ischemic cerebrovascular disease in old people. 1∶1 paired case-control compa...In order to investigate the association of fibrin monomer polymerization function (FMPF) with traditional cerebrovascular risk factors and ischemic cerebrovascular disease in old people. 1∶1 paired case-control comparative study was performed for FMPF and traditional cerebrovascular risk factors on 110 cases of old ischemic cerebrovascular disease and 110 controls matched on age, sex and living condition. The results showed that cerebrovascular risk factors were more prevalent in case group than in control group. In the case group, FMPF was significantly higher than in control group. There was a significant positive correlation between hypertension and fibrin monomer polymerization velocity (FMPV), hypertension and fibrinogen (Fbg), alcohol consumption and Fbg, but no significant correlation between diabetic mellitus, smoking and FMPF was found. Among the parameters of blood lipids, there were significant positive correlations between total cholesterol (TC) and parameters of FMPF to varying degrees, triglycerides (TG) and FMPV, TG and Fbg. Our results also showed there were significant linear trends between TC and FMPV (P<0. 001), TC and Fbg (P=0. 0087), TG and FMPV/Amax (maximum absorbance)(P=0. 0143) respectively. Multiple logistic regression analysis revealed that FMPF in case group remained significantly higher than control group after adjustment of all risk factors that were significant in univariate analysis. It was concluded that there is a possible pathophysiological link between FMPF and cerebrovascular risk factors. An elevated FMPF is associated with ischemic cerebrovascular disease and an independent risk factor of this disease. In old people, detection of FMPF might be a useful screening to identify individuals at increased cerebrothrombotic risk.展开更多
The traditional linear programming model is deterministic. The way that uncertainty is handled is to compute the range of optimality. After the optimal solution is obtained, typically by the simplex method, one consid...The traditional linear programming model is deterministic. The way that uncertainty is handled is to compute the range of optimality. After the optimal solution is obtained, typically by the simplex method, one considers the effect of varying each objective function coefficient, one at a time. This yields the range of optimality within which the decision variables remain constant. This sensitivity analysis is useful for helping the analyst get a sense for the problem. However, it is unrealistic because objective function coefficients tend not to stand still. They are typically profit contributions from products sold and are subject to randomly varying selling prices. In this paper, a realistic linear program is created for simultaneously randomizing the coefficients from any probability distribution. Furthermore, we present a novel approach for designing a copula of random objective function coefficients according to a specified rank correlation. The corresponding distribution of objective function values is created. This distribution is examined directly for central tendency, spread, skewness and extreme values for the purpose of risk analysis. This enables risk analysis and business analytics, emerging topics in education and preparation for the knowledge economy.展开更多
基金supported by the National Natural Science Fundation of China (60736021)the Joint Funds of NSFC-Guangdong Province(U0735003)
文摘Kernel-based methods work by embedding the data into a feature space and then searching linear hypothesis among the embedding data points. The performance is mostly affected by which kernel is used. A promising way is to learn the kernel from the data automatically. A general regularized risk functional （RRF） criterion for kernel matrix learning is proposed. Compared with the RRF criterion, general RRF criterion takes into account the geometric distributions of the embedding data points. It is proven that the distance between different geometric distdbutions can be estimated by their centroid distance in the reproducing kernel Hilbert space. Using this criterion for kernel matrix learning leads to a convex quadratically constrained quadratic programming （QCQP） problem. For several commonly used loss functions, their mathematical formulations are given. Experiment results on a collection of benchmark data sets demonstrate the effectiveness of the proposed method.
基金funded by an industry-academia collaborative grant EPSRC EP/R511687/1 awarded by EPSRC&University of Edinburgh,United Kingdom program Impact Acceleration Account(IAA)R.Ibraheem is a Ph.D.student in EPSRC’s MAC-MIGS Centre for Doctoral Training.MAC-MIGS is supported by the UK’s Engineering and Physical Science Research Council(grant number EP/S023291/1)+2 种基金G.dos Reis acknowledges partial support from the FCT-Fundação para a Ciência e a Tecnologia,PortugalI.P.,under the scope of the projects UIDB/00297/2020(https://doi.org/10.54499/UIDB/00297/2020)and UIDP/00297/2020(https://doi.org/10.54499/UIDP/00297/2020)(Center for Mathematics and Applications,NOVA Math)G.dos Reis acknowledges support from the Faraday Institution,United dom King-(grant number FIRG049).
文摘Single-value prediction such as the End of Life and Remaining Useful Life is a common method of estimating the lifetime of Li-ion batteries.Information from such prediction is limited when the entire degradation pattern is needed for practical applications such as dynamic adjustment of battery warranty,improved maintenance scheduling,and battery stock management.In this research,a predictive,semi-parametric survival model called the Cox Proportional Hazards is proposed for the prediction of cell degradation in the form of survival probability(battery reliability)and cumulative hazard(battery risk)functions.Once this model is trained,the two functions can be obtained directly for a new cell without having to predict several cogent points.The model is trained on the first 50 cycles of only the voltage profile from either the charge or discharge data regime,implying that our methodology is data region agnostic.The signature method with both desirable mathematical and machine learning properties was adopted as a feature extraction technique.
文摘Sexual dysfunction and prostate cancer are common among older men. Few studies explored the association between these two illnesses. We examined whether sexual function is associated with prostate cancer risk among older men. Among 448 men undergoing prostate biopsy at the Durham Veterans Affairs Hospital, sexual function was ascertained from the Expanded Prostate Cancer Index Composite sexual assessment. We tested the link between sexual function and prostate cancer risk adjusting for multiple demographic and clinical characteristics using logistic regression. Multinomial logistic regression was used to test the associations with risk of low-grade （Gleason 〈6） and high-grade （Gleason 〉7 or 〉4 ＋ 3） disease versus no cancer. Of 448 men, 209 （47%） had a positive biopsy; these men were less likely to be white （43% vs 55%, P=0.013）, had higher prostate-specific antigen （PSA） （6.0 vs 5.4 ng ml-1, P 〈 0.001）, but with lower mean sexual function score （47 vs 54, P = 0.007）. There was no difference in age, BMI, pack years smoked, history of heart disease and/or diabetes. After adjusting for baseline differences, sexual function was linked with a decreased risk of overall prostate cancer risk （OR. 0.91 per lO-point change in sexual function, P= 0.004） and high-grade disease whether defined as Gleason ≥7 （OR： 0.86, P= 0.001） or 〉4 ＋ 3 （OR： 0.85, P= 0.009）. Sexual function was unrelated to low-grade prostate cancer （OR： 0.94, P = 0.13）. Thus, among men undergoing prostate biopsy, higher sexual function was associated with a decreased risk of overall and high-grade prostate cancer. Confirmatory studies are needed.
文摘According to the characteristics of the correlation of multiple wind farm output, this paper put forwards a modeling method based on fuzzy c-means clustering and the copula function, and correlation wind farms are inserted into IEEE-RTS79 reliability system for risk assessment. By the probabilistic load flow calculated by Monte Carlo simulation method, the probability of the accident is derived, and bus voltage and branch power flow overload risk index are defined in this paper. The results show that this method can realize the modeling of the correlation of wind power output, and the risk index can identify the weakness of the system, which can provide reference for the operation and maintenance personnel.
基金supported by the auspices of National Natural Science Foundation of China(No.51079037)Opening Foundation of Chengdu Institute of Plateau Meteorology,China Meteorological Administration(LPM2011002)Natural Science Foundation of Anhui Province(No.1208085ME75)
文摘There are many uncertain factors,such as stochastic,fuzzy and gray information in the risk analysis on natural hazard.The set pair analysis(SPA)deals effectively with the various uncertain factors contributing to evaluation of the risk level of natural disasters.The evaluation indicators and standards of natural disasters risk are analyzed by identity-discrepancycontrary(IDC).The result,the connection numbers,still has uncertainty information.Thus,yielding the risk evaluation model of natural disasters based on connection function,which construct a set pair relation between the indicators of connection numbers and comprehensive evaluation standards,and describe uncertainty of the connection numbers by using connection function.The study showed that the proposed model takes into account only the uncertainty of risk evaluation indicator identification on natural disasters,also the uncertainty of result connection numbers.This approach gives full consideration to the uncertainty of systematic evaluation process along with the actual meaning of comprehensive evaluation functions.Therefore,this means it is able to reduce the uncertainty of final evaluation results and improve the accuracy and reasonability of evaluation results.This model is capable of reflecting actual situation of the risk evaluation on natural disaster affected by various uncertain factors and has a promotional value in the natural disaster risk assessment.
基金supported in part by Hubei Normal University Post-graduate Foundation(2007D59 and 2007D60)the Science and Technology foundation of Hubei(D20092207)the National Natural Science Foundation of China(10671149)
文摘This article considers a Markov-dependent risk model with a constant dividend barrier. A system of integro-differential equations with boundary conditions satisfied by the expected discounted penalty function, with given initial environment state, is derived and solved. Explicit formulas for the discounted penalty function are obtained when the initial surplus is zero or when all the claim amount distributions are from rational family. In two state model, numerical illustrations with exponential claim amounts are given.
文摘Estimation of the bivariate survival function under the competing risks caseis considered.We give an explicit formula for the estimator from a decomposition of thebivariate survival function based on competing risks,which is almost sure consistent.
基金Major Program of National Natural Science Foundation of China,No.41590841Key Project for National Natural Science Foundation of China,No.41230633
文摘Urbanization significantly increases the risk of urban flooding. Therefore, quantitative study of urban rainfall-runoff processes can provide a scientific basis for urban planning and management. In this paper, the built-up region within the Fifth Ring Road of Beijing was selected as the study area. The details of land cover and urban function zones(UFZs) were identified using GIS and RS methods. On this basis, the SCS-CN model was adopted to analyze the rainfall-runoff risk characteristics of the study area. The results showed that:(1) UFZs within different levels of runoff risk varied under different rainfall conditions. The area ratio of the UFZs with high runoff risk increased from 18.90%(for rainfall return period of 1 a) to 54.74%(for period of 100 a). Specifically, urban commercial areas tended to have the highest runoff risk, while urban greening spaces had the lowest.(2) The spatial characteristics of the runoff risks showed an obvious circular distribution. Spatial cluster areas with high runoff risk were mainly concentrated in the center of the study area, while those with low runoff risk were mainly distributed between the fourth and fifth ring roads. The results indicated that the spatial clustering characteristic of urban runoff risk and runoff heterogeneity among different UFZs should be fully considered during urban rainwater management.
文摘In order to investigate the association of fibrin monomer polymerization function (FMPF) with traditional cerebrovascular risk factors and ischemic cerebrovascular disease in old people. 1∶1 paired case-control comparative study was performed for FMPF and traditional cerebrovascular risk factors on 110 cases of old ischemic cerebrovascular disease and 110 controls matched on age, sex and living condition. The results showed that cerebrovascular risk factors were more prevalent in case group than in control group. In the case group, FMPF was significantly higher than in control group. There was a significant positive correlation between hypertension and fibrin monomer polymerization velocity (FMPV), hypertension and fibrinogen (Fbg), alcohol consumption and Fbg, but no significant correlation between diabetic mellitus, smoking and FMPF was found. Among the parameters of blood lipids, there were significant positive correlations between total cholesterol (TC) and parameters of FMPF to varying degrees, triglycerides (TG) and FMPV, TG and Fbg. Our results also showed there were significant linear trends between TC and FMPV (P<0. 001), TC and Fbg (P=0. 0087), TG and FMPV/Amax (maximum absorbance)(P=0. 0143) respectively. Multiple logistic regression analysis revealed that FMPF in case group remained significantly higher than control group after adjustment of all risk factors that were significant in univariate analysis. It was concluded that there is a possible pathophysiological link between FMPF and cerebrovascular risk factors. An elevated FMPF is associated with ischemic cerebrovascular disease and an independent risk factor of this disease. In old people, detection of FMPF might be a useful screening to identify individuals at increased cerebrothrombotic risk.
文摘The traditional linear programming model is deterministic. The way that uncertainty is handled is to compute the range of optimality. After the optimal solution is obtained, typically by the simplex method, one considers the effect of varying each objective function coefficient, one at a time. This yields the range of optimality within which the decision variables remain constant. This sensitivity analysis is useful for helping the analyst get a sense for the problem. However, it is unrealistic because objective function coefficients tend not to stand still. They are typically profit contributions from products sold and are subject to randomly varying selling prices. In this paper, a realistic linear program is created for simultaneously randomizing the coefficients from any probability distribution. Furthermore, we present a novel approach for designing a copula of random objective function coefficients according to a specified rank correlation. The corresponding distribution of objective function values is created. This distribution is examined directly for central tendency, spread, skewness and extreme values for the purpose of risk analysis. This enables risk analysis and business analytics, emerging topics in education and preparation for the knowledge economy.
