Kernel matrix learning with a general regularized risk functional criterion

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.

Risk evaluation of natural disasters based on connection function

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.

Wind Power System Risk Assessment Based on Fuzzy Clustering and Copula Function Modeling

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.

Application of Dangerous Probability Function in Risk Estimation of Surface Rupture Earthquake: Taking the Lingwu Fault in Ningxia Region as an Example

Based on the fitting on paleoearthquake data of intra-plate regions in the northern part of China and giving a statistical model of time interdependence, the potential damage earthquakes in a definite future period and characteristics of present shocks along the Lingwu fault have been analyzed by using dangerous probability function and some new data concerned. We have inferred that the fault has entered a period that earthquakes will probably occur. There exists a potential danger that a strong earthquake with M\-S7.0～7.5 will occur in 10～100a.

Association of Fibrin Monomer Polymerization Function, Cerebrovascular Risk Factors and Ischemic Cerebrovascular Disease in Old People

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.

Randomized Objective Function Linear Programming in Risk Management

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.

吸烟和轻中度饮酒与社区老年男性认知功能相关性研究

目的探讨吸烟和轻中度饮酒与社区老年男性认知功能的相关性。方法采用中国纵向老龄队列数据库,纳入其中1201名社区老年男性(不含重度饮酒者),根据自我汇报的吸烟饮酒史将其分为吸烟饮酒组(n=332)、不吸烟但饮酒组(n=126)、吸烟但不饮酒组(n=308)、不吸烟不饮酒组(n=435),应用北京版蒙特利尔认知评估量表(Montreal cognitive assessment,MoCA)评估认知功能。采用双因素方差分析比较认知功能的差异,多因素线性回归模型分析认知功能下降的风险因素。结果主效应分析提示,轻中度饮酒对MoCA总分(F=6.076,P=0.014)、MoCA命名(F=11.179,P=0.001)和MoCA抽象(F=7.718,P=0.006)的影响具有统计学意义,轻中度饮酒者比不饮酒者的MoCA总分(22.50±5.27 vs.23.30±5.28)、MoCA命名(2.41±0.85 vs.2.58±0.76)、MoCA抽象(0.93±0.84 vs.1.10±0.82)更低。吸烟对MoCA总分(F=0.234,P=0.628)、MoCA命名(F=0.110,P=0.741)和MoCA抽象(F=1.335,P=0.248)的影响无统计学意义,吸烟和轻中度饮酒对MoCA评分的影响不存在交互作用(P>0.05)。多因素线性回归分析结果显示,无轻中度饮酒史与MoCA命名呈正相关(B=0.125,P=0.008)。对非痴呆者进行分层分析显示,无轻中度饮酒史与MoCA总分(B=0.550,P=0.011)、MoCA命名(B=0.134,P=0.002)呈正相关。结论吸烟和轻中度饮酒对社区老年男性认知功能无明显的相互影响,轻中度饮酒可能与总体认知功能和命名功能下降相关。

NONPARAMETRIC ESTIMATION OF BIVARIATE SURVIVAL FUNCTION UNDER THE COMPETING RISKS CASE

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.

ON THE EXPECTED DISCOUNTED PENALTY FUNCTION IN A MARKOV-DEPENDENT RISK MODEL WITH CONSTANT DIVIDEND BARRIER

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.

不同术式喉癌患者术后吞咽功能、认知功能的比较及其危险因素分析

目的 分析不同术式喉癌患者术后吞咽功能、认知功能的比较及其危险因素。方法 回顾性选取2020年4月至2023年4月新疆医科大学第一附属医院收治的80例喉癌患者进行研究,根据手术方法不同将患者分为A组(n=26)、B组(n=27)、C组(n=27)。A组行水平半喉切除术,B组行环状软骨-舌骨-会厌固定术(CHEP),C组行垂直半喉切除术。对比三组患者的术后吞咽功能、认知功能,并建立多因素Logistic模型,分析影响术后吞咽功能、认知功能的独立危险因素。结果 C组术后吞咽功能0级率为70.37%,明显高于A(38.46%)、B组(37.04%),差异有统计学意义(P<0.05)。C组术后MOCA、MMSE评分分别为(26.14±2.25)、(27.16±2.14)分,均明显高于A[(24.12±2.34)、(25.14±2.24)分]、B组[(24.45±2.45)、(25.36±2.18)分],差异均有统计学意义(P<0.05)。80例患者的术后吞咽功能障碍发生率为51.25%,认知功能障碍发生率为60.00%。与吞咽功能正常组比较,障碍组年龄≥60岁、TNMⅢ期、颈部淋巴结清扫、气管切开术、水平半喉切除术、CHEP术患者占比较高;与认知功能正常组比较,障碍组年龄≥60岁、初中及以下、水平半喉切除术、CHEP术患者占比较高(P<0.05)。单因素和多因素Logistic分析结果显示,年龄、TNM分期、颈部淋巴结清扫、气管切开术、水平半喉切除术、CHEP术均是导致喉癌患者术后吞咽功能障碍的独立危险因素;年龄、文化程度、水平半喉切除术、CHEP术均是导致喉癌患者术后认知功能障碍的独立危险因素(P<0.05)。结论 垂直半喉切除术手术对喉癌吞咽和认知功能影响较小,水平半喉切除术、CHEP和年龄、文化程度、临床分期等因素均有可能导致喉癌患者引发术后吞咽功能、认知功能障碍。

帕金森病患者出现冻结步态的危险因素分析

目的探讨帕金森病(PD)患者出现冻结步态(FOG)的危险因素。方法选取2019年7月至2022年7月中日友好医院收治的52例PD患者为研究对象,根据是否合并FOG将研究对象分为FOG组(合并FOG,14例)和非FOG组(未合并FOG,38例)。比较分析非FOG组与FOG组患者的临床特征与影像学特征。比较分析非FOG组与FOG组患者的基底前脑功能连接。采用多因素logistic回归方法分析PD患者出现FOG的独立影响因素。结果FOG组患者的国际运动障碍学会统一帕金森病评定量表Ⅱ(MDS-UPDRSⅡ)评分高于非FOG组,差异有统计学意义(P<0.05),且Meynert基底核(Ch4)与双侧中扣带回、中央前回及辅助运动区的功能连接低于非FOG组。多因素logistic回归分析结果显示,MDS-UPDRSⅡ评分、Ch4功能连接是PD患者出现FOG的独立影响因素(P<0.05)。结论MDS-UPDRSⅡ评分升高、Ch4功能连接减弱为PD患者出现FOG的独立影响因素。

再植指功能恢复效果的列线图预测模型构建

目的探讨再植指功能恢复差的危险因素,建立患指功能恢复差的预测模型。方法回顾性分析2017年5月至2022年5月在唐山市人民医院行断指再植术的309例(断指376个)手指离断患者的资料,其中男性231例断指274个,女性78例断指102个;年龄19~67岁,平均(41.57±8.59)岁。术后1年对患指功能的恢复效果进行判定,并分成恢复差组和对照组。收集患者一般人口特征、病情信息、手术信息、术前血常规指标行单变量筛查,通过多因素分析确定患指功能恢复差的危险因素,以森林图展现各指标效应大小。依据预测变量构建列线图模型,并通过受试者工作特征(receiver operating characteristic,ROC)曲线下面积(area under curve,AUC)、校准曲线及决策曲线(decision curve analysis,DCA)检验其效能。结果309例(断指376个)患者中,有68例(断指81个)术后功能恢复差,占比21.54%。两组吸烟史、损伤性质、离断程度、离断平面、断指暴露时间、断指保存方式、是否遵医开展康复锻炼情况比较差异有统计学意义(P<0.05)。Logistic回归分析显示,术后患指功能恢复差的危险因素主要为有吸烟史、撕脱伤、离断平面为非末节指骨、断指暴露时间≥8 h、未遵医开展康复锻炼。内部验证显示,AUC为0.819(95%CI:0.762~0.875),模型预测患指功能恢复差的概率与实际概率高度吻合,拟合优度HL检验χ^(2)=7.389,P=0.286。DCA显示模型在风险阈值12%~75%时具有较好的实用性。结论根据有吸烟史、损伤性质、离断平面、断指暴露时间、是否遵医开展康复锻炼构建的列线图模型对断指再植术后患指功能恢复差的预测效果较好。

采用三次样条函数拟合蚊媒密度概率分布及其风险评估

目的研究登革热蚊媒MOI风险的概率分布,为更加科学精准开展MOI风险评估提供新方法。方法采用三次样条函数、累积概率分布和Python语言编程对广州市2016—2019年各月MOI监测数据进行拟合分析。结果本研究绘制了广州市1-12月的MOI风险评估概率分布表和分布图。1-12月各月累积概率小于0.01和0.05的MOI临界值分别在7.54~60.12、5.02~34.10之间。该临界值均在1月最低,其次是12月和2月;均在6月份最高,其次是5月和7月。1-12月各月MOI风险等级为传播、暴发和流行的概率分别在5.07%~66.60%、0.25%~36.19%、0.00%~16.80%之间。传播、暴发和流行的风险概率最低值均出现在1月,其次均在2月;最高值分别出现在7月、7月和6月,其次是6月、6月和5月。结论广州市各月的MOI风险概率分布具有显著的季节性消长规律和特征,在开展登革热蚊媒监测评估与控制工作中值得重视。三次样条函数和累积概率分布拟合法为开展MOI风险评估提供了新的分析方法。

北京社区55~75岁高、低痴呆风险人群身体素质及其认知功能特点

目的分析预测社区中处于生命中期健康老年人痴呆风险因素。方法2021年7月至2023年4月招募北京市两个社区常住1年以上的55~75岁中老年人175例,收集心血管危险因素、衰老和痴呆(CAIDE)相关危险因素和其他一般人口资料,CAIDE风险评分≥9分为高危组,<9分为低危组。采用Stroop色词测验(SCWT)、两元素1-back任务范式、连线测验(TMT)进行评估;测量握力、30 s前臂屈曲测验和5次坐站试验;记录平地步行10 m的平均步速和步长。结果高危组CAIDE总分平均9.86分,低危组4.95分。两组年龄无显著性差异(P=0.188)。男性,受教育程度<7年、收缩压>140 mmHg、胆固醇>6.5 mmol/L、体质量指数>30 kg/m^(2)和缺乏体力活动人群CAIDE痴呆风险评分更高(χ^(2)>3.116,P<0.05)。高危组握力(t=-4.174)、步速(t=-2.414)、SCWT正确率(Z=-2.684)均差于低危组(P<0.05)。Logistic回归分析显示,步速(OR=25.483)、握力(OR=1.133),SCWT正确率(OR=37.430)是痴呆的独立危险因素(P<0.05)。结论握力弱、步速慢、抑制控制能力差可能预示痴呆风险增加。

老年急性脑梗死患者术后神经功能恢复不良的影响因素分析

目的分析老年急性脑梗死患者术后神经功能恢复不良的影响因素,并构建风险列线图预测模型。方法选取2021-01—2022-12于发病4.5 h内在济宁医学院附属医院进行静脉溶栓术的174例老年急性脑梗死患者,根据术后神经功能恢复情况分为恢复良好组(n=113)与恢复不良组(n=61),采用Logistic回归分析影响老年急性脑梗死患者术后神经功能恢复的危险因素;采用R软件构建预测老年急性脑梗死患者术后神经功能恢复不良的列线图模型;采用ROC曲线、校准曲线、Hosmer-Lemeshow拟合优度检验评估列线图模型的效能。结果174例老年急性脑梗死患者,61例(35.06%)术后神经功能恢复不良,113例(64.94%)术后神经功能恢复良好。多因素Logistic回归分析显示,糖尿病(95%CI:2.466~21.685,P<0.001)、心房颤动(95%CI:1.485~24.484,P=0.012)、基线NIHSS评分≥12.50分(95%CI:5.945~53.525,P<0.001)、出血性转化(95%CI:1.823~15.328,P=0.002)、责任血管中度狭窄(95%CI:1.686~17.834,P=0.005)、责任血管重度狭窄(95%CI:2.769~40.737,P=0.001)为老年急性脑梗死患者术后神经功能恢复不良的独立危险因素。该列线图模型预测老年急性脑梗死患者术后神经功能恢复不良的ROC曲线下面积0.872(95%CI:0.816~0.928),绘制校准曲线显示预测概率曲线与实际概率曲线基本吻合,且Hosmer-Lemeshow拟合优度检验χ^(2)=7.943,P=0.439。结论糖尿病、心房颤动、基线NIHSS评分≥12.50分、出血性转化、责任血管中/重度狭窄是老年急性脑梗死患者术后神经功能恢复不良的独立危险因素,依此构建的风险列线图预测模型经验有较好的预测效能,能准确、有效预测老年急性脑梗死患者术后神经功能恢复不良发生风险。

聊城市农田土壤污染评价及农业功能分区

为探索聊城市农田土壤污染状况及农业功能分区,以2017~2020年采集的74个表层土样为研究对象,测定了8种重金属元素(Cd、Hg、As、Cu、Pb、Cr、Zn和Ni)、BHC_(s)、DDT_(s)和BaP的含量。依据《土壤环境质量农用地土壤污染风险管控标准》和《绿色食品产地环境质量》作为评价标准,采用污染指数法开展了土壤环境质量评价和农业功能分区研究。采用ANOVA和LSD多重比较,确定了污染物的空间分布差异及其来源。依据山东省土壤背景值进行了重金属潜在生态风险评价。结果表明,除3个DDTs污染样点和1个Cd污染样点外,其他样点均处于清洁水平。在空间分布上,DDTs和Cr含量存在极显著差异,BaP和Hg存在显著差异。农业功能分区结果表明,绿色食品种植区、无公害食品种植区、观察种植区和污染区面积分别占研究区总面积的78.13%、9.45%、6.52%和5.90%。土壤环境生态风险水平以轻微为主,占总面积的77.03%。Hg和Cd在生态风险评估中占有较大比率,应高度警惕和严格监管。

基于AHP耦合Sigmoid函数的重载铁路黄土边坡风险分析

重载铁路具有大轴重、高密度、大运量的运输特点,黄土边坡路基段一直是重载铁路运营的高风险区段,现有风险评价方法尚无法满足其风险量化分析的需求。利用层次分析法AHP耦合Sigmoid函数的方法,围绕事故发生概率和损失两个方面对重载铁路黄土边坡工程展开风险定量评价。首先以边坡滑塌事故调研和稳定性控制理论为基础,明确黄土边坡的失稳影响因素,建立重载铁路黄土边坡稳定性评价模型和边坡失稳危害性评价模型;然后应用层次分析法建立黄土边坡稳定性评价模型指标权重,并提出各级影响因素的指标赋值方法;而后引入基于朴素贝叶斯理论的Sigmoid函数,构建边坡稳定服役概率、风险判断矩阵和黄土边坡的风险等级确定方法;最后以朔黄铁路某站高陡黄土边坡为工程背景,应用所得理论方法对其进行风险定量分析,计算得出该边坡失稳概率为0.64,造成经济损失约8611万元,隶属于高风险等级,并针对坡体几何特征、渗透能力、基岩特性等低分指标提出了具体的安全性提升措施建议。

基于快速进行平方算法的巡检机器人路径规划研究

针对目前快速行进平方法在实际场景中对Sat值选取缺乏评价尺度的问题,提出一种改进的快速行进平方法以应用于巡检机器人的路径规划。该改进方法引入路径评价函数,通过对机器人行驶距离、行驶角度变化量和与障碍物的距离进行无量纲化处理,以获取当前环境下最优的截止阈值参数Sat。构建了PID控制算法,校正了机器人的动作误差。使用MATLAB软件绘制实际运动轨迹,比较相关路径参数。仿真实验表明,改进后的快速行进平方法可根据截止阈值参数Sat生成适用于不同车型和环境特点的最优路径。该研究的改进方法为巡检机器人在实际应用中的路径规划提供了参考。

基于不确定性的机坪管制系统风险评估

对机坪管制系统进行风险评估是提高机场安全水平的一个重要措施,但如何科学客观地评价机坪管制系统,得到贴近现实的评估结果是一个难题。基于此,建立了机坪管制系统风险评价体系,从组合权重法的观点出发,通过乘法归一化合成评价权重,并在模糊集理论的基础上引入DS证据论对某机场机坪管制系统进行实例评估,得出人员因素是机坪管制系统安全体系的重点这一结论,对机坪管制系统安全改进有较好的指导意义。

数据-知识融合的水利工程建设安全风险灰色因子分解机预测模型

【目的】已有的数据驱动的水利工程建设安全风险预测方法对领域知识的挖掘和利用不足,预测结果的准确性和可解释性有待进一步提高。为了构建数据-知识融合的水利工程建设安全风险预测模型,【方法】将灰色聚类与因子分解机相结合,提出了一种融合领域知识的灰色因子分解机。首先,引入基于可能度函数的灰色聚类表征水利工程建设领域专家有关安全风险的先验知识。然后,将先验知识以参数的形式嵌入到因子分解机中,构建出数据-知识融合的灰色因子分解机。最后,基于随机梯度下降构造模型参数的求解算法,并结合实例对模型的有效性进行验证。【结果】实例应用结果显示,与传统因子分解机相比,灰色因子分解机的准确率、精确率、召回率和F_(1)值均得到了不同程度的提升。与支持向量机、深度因子分解机等其他基准模型相比,灰色因子分解机同样具有更好的预测性能。【结论】这表明,数据-知识融合驱动的灰色因子分解机模型能够更加准确地预测出安全风险,从而为水利工程建设安全风险管控提供更好的决策支持。