Investigating Spatio-Temporal Pattern of Relative Risk of Tuberculosis in Kenya Using Bayesian Hierarchical Approaches

Proper understanding of global distribution of infectious diseases is an important part of disease management and policy making. However, data are subject to complexities caused by heterogeneities across host classes and space-time epidemic processes. This paper seeks to suggest or propose Bayesian spatio-temporal model for modeling and mapping tuberculosis relative risks in space and time as well identify risks factors associated with the tuberculosis and counties in Kenya with high tuberculosis relative risks. In this paper, we used spatio-temporal Bayesian hierarchical models to study the pattern of tuberculosis relative risks in Kenya. The Markov Chain Monte Carlo method via WinBUGS and R packages were used for simulations and estimation of the parameter estimates. The best fitting model is selected using the Deviance Information Criterion proposed by Spiegelhalter and colleagues. Among the spatio-temporal models used, the Knorr-Held model with space-time interaction type III and IV fit the data well but type IV appears better than type III. Variation in tuberculosis risk is observed among Kenya counties and clustering among counties with high tuberculosis relative risks. The prevalence of HIV is identified as the determinant of TB. We found clustering and heterogeneity of TB risk among high rate counties and the overall tuberculosis risk is slightly decreasing from 2002-2009. We proposed that the Knorr-Held model with interaction type IV should be used to model and map Kenyan tuberculosis relative risks. Interaction of TB relative risk in space and time increases among rural counties that share boundaries with urban counties with high tuberculosis risk. This is due to the ability of models to borrow strength from neighboring counties, such that nearby counties have similar risk. Although the approaches are less than ideal, we hope that our study provide a useful stepping stone in the development of spatial and spatio-temporal methodology for the statistical analysis of risk from tuberculosis in Kenya.

Spectrum-effect relationship between HPLC fingerprints and bioactive components of Radix Hedysari on increasing the peak bone mass of rat

The traditional Chinese medicine of Radix Hedysari plays an important role in invigorating gas for ascending, benefiting blood for promoting production of fluid, and promoting circulation for removing obstruction in collaterals, which is consistent with the principle of treatment for osteoporosis. This study is designed to investigate the bioactive components on increasing peak bone mass (PBM) by exploring the spectrum-effect relationship between chromatography fingerprints and effect. Multiple indicators are selected to evaluate the pharmacological activity. In fingerprints, 21 common peaks are obtained, five of which are identified. Furthermore, gray relational analysis (GRA) is a quantitative method of gray system theory and is used to describe the correlation degree of common peaks and pharmacological activities with relational value. 21 components are then divided into three different regions, of which ononin and calycosin play an extremely significant role in increasing PBM. In addition, factor analysis and hierarchical cluster analysis (HCA) are used to screen the optimal producing area for Radix Hedysari. This provides a comprehensive and efficient method to improve the quality evaluation of Radix Hedysari, confirming the bioactive components for PBM-enhancement and further develop its medicinal value.

Spatial Modeling and Mapping of Tuberculosis Using Bayesian Hierarchical Approaches

Global spread of infectious disease threatens the well-being of human, domestic, and wildlife health. A proper understanding of global distribution of these diseases is an important part of disease management and policy making. However, data are subject to complexities by heterogeneity across host classes. The use of frequentist methods in biostatistics and epidemiology is common and is therefore extensively utilized in answering varied research questions. In this paper, we applied the hierarchical Bayesian approach to study the spatial distribution of tuberculosis in Kenya. The focus was to identify best fitting model for modeling TB relative risk in Kenya. The Markov Chain Monte Carlo (MCMC) method via WinBUGS and R packages was used for simulations. The Deviance Information Criterion (DIC) proposed by [1] was used for models comparison and selection. Among the models considered, unstructured heterogeneity model perfumes better in terms of modeling and mapping TB RR in Kenya. Variation in TB risk is observed among Kenya counties and clustering among counties with high TB Relative Risk (RR). HIV prevalence is identified as the dominant determinant of TB. We find clustering and heterogeneity of risk among high rate counties. Although the approaches are less than ideal, we hope that our formulations provide a useful stepping stone in the development of spatial methodology for the statistical analysis of risk from TB in Kenya.

Mining Hierarchical Decision Rules from Hybrid Data with Categorical and Continuous Valued Attributes

Decision rules mining is an important issue in machine learning and data mining.However,most proposed algorithms mine categorical data at single level,and these rules are not easily understandable and really useful for users.Thus,a new approach to hierarchical decision rules mining is provided in this paper,in which similarity direction measure is introduced to deal with hybrid data.This approach can mine hierarchical decision rules by adjusting similarity measure parameters and the level of concept hierarchy trees.

基于异构图分层学习的细粒度多文档摘要抽取

抽取的目标是在多个文档中提取共有关键信息,其对简洁性的要求高于单文档摘要抽取。现有的多文档摘要抽取方法通常在句子级别进行建模,容易引入较多的冗余信息。为了解决上述问题,提出一种基于异构图分层学习的多文档摘要抽取框架,通过层次化构建单词层级图和子句层级图来有效建模语义关系和结构关系。针对单词层级图和子句层级图这2个异构图的学习问题,设计具有不同层次更新机制的两层学习层来降低学习多种结构关系的难度。在单词层级图学习层,提出交替更新机制更新不同的粒度节点,以单词节点为载体通过图注意网络进行语义信息传递;在子句层级图学习层,提出两阶段分步学习更新机制聚合多种结构关系,第一阶段聚合同构关系,第二阶段基于注意力聚合异构关系。实验结果表明,与抽取式基准模型相比,该框架在Multinews数据集上取得了显著的性能提升,ROUGE-1、ROUGE-2和ROUGE-L分别提高0.88、0.23和2.27,消融实验结果也验证了两层学习层及其层次更新机制的有效性。

Corporate Credit Ratings Based on Hierarchical Heterogeneous Graph Neural Networks

order to help investors understand the credit status of target corporations and reduce investment risks,the corporate credit rating model has become an important evaluation tool in the financial market.These models are based on statistical learning,machine learning and deep learning especially graph neural networks(GNNs).However,we found that only few models take the hierarchy,heterogeneity or unlabeled data into account in the actual corporate credit rating process.Therefore,we propose a novel framework named hierarchical heterogeneous graph neural networks(HHGNN),which can fully model the hierarchy of corporate features and the heterogeneity of relationships between corporations.In addition,we design an adversarial learning block to make full use of the rich unlabeled samples in the financial data.Extensive experiments conducted on the public-listed corporate rating dataset prove that HHGNN achieves SOTA compared to the baseline methods.

电网分析模型数据的图表组织结构对比与适用性研究

为了明晰电网分析模型的数据组织结构选择标准,对其图表组织方式的适用性研究。基于本体思想,构建电网分析模型的元件表、元件图和层次化图数据模型,并在不同规模电网下进行测试和对比性能。测试结果给出了三者的检索性能定量关系,表明所构建层次图的检索性能在多层拓扑检索中分别约为元件图和元件表的10倍和40倍,可为电网数字底座的数据组织结构选配提供依据。

后申克理论的原则、方法及思考

与传统申克理论相较,后申克理论发展变化主要体现在三方面:一是“域扩展”,将申克理论的分析应用扩展至广阔的后调性音乐之中;二是“态拓展”,为适应后调性音乐之特点,后申克理论对基本结构及延长形态作了相应变化与拓展处理;三是理论综合,将申克理论与其他分析方法综合起来,以混合策略来解析作品。有机统一、层次延展是申克理论与后申克理论的共同原则,一言蔽之,道未变,乃象变也。

基于标签注意力的分层ICD自动编码方法

针对目前自动ICD(international classification of diseases)编码任务存在标签空间大、诊断代码分布不均衡与临床文本表征差的问题,提出一种融合Longformer与标签注意力的分层ICD自动编码模型。借助Clinical-Longformer预训练语言模型获得融合长文本语境的词向量表征。通过将标签的语义表示与注意力机制相结合,捕捉临床文本中与诊断代码相关的关键特征信息,获取更精准的文本表示。引入分层联合学习机制,建立分层预测层解码输出ICD编码。实验结果表明,该模型的准确率、召回率与F1值均高于现有模型,验证了该方法进行自动ICD编码的有效性,为实施疾病诊断相关分组提供高质量的数据支撑。

基于嵌套集合模型的时态层次数据管理方法

时态层次数据是层次数据在时间维度的扩展,用于描述随时间变化的层次结构。相较于非时态层次数据,现有的时态层次数据管理方法仍存在存储方案复杂以及查询和更新效率低下等问题。针对上述问题,提出了一种基于嵌套集合模型的时态层次数据管理方法。首先从节点变化角度分析了层次数据变化的4种类型,在此基础上通过扩展时间标签字段实现了多版本节点在关系数据库中的存储和查询功能,最后提出了一种基于存量空间的嵌套集合模型(Abundantly Gapped Nested Intervals Scheme,AGNIS),用于解决主流嵌套集合模型插入数据记录效率较低的问题。基于我国2021-2022年行政区划及其调整数据的实验结果表明:提出的数据管理方法能够实现历史层次数据的存储和任意时刻层次结构快照的查询,且兼顾了时态层次数据查询和更新操作的高效性。

一种基于分层结构的设备故障领域本体实例关系抽取方法

优质的设备故障领域本体需要实例之间拥有准确的关系。传统的基于神经网络的实例关系抽取方法容易使意义相近的关系误分,且由于故障领域文本中实例间隔较远,使用该类方法抽取故障领域实例关系效果不佳。针对上述问题,基于分层结构提出了一种从非结构化文本中抽取设备故障领域实例关系的方法。第一层通过对实例对中实例类型的识别,进行初步关系抽取;第二层使用融合自注意力机制的BiLSTM模型对包含实例对的句子进行解析,进行精确分类;通过统计的方法完成实例关系抽取。实验结果表明,此方法与现有的基于神经网络的实例关系抽取方法相比,取得了较好的精确度与召回率,提高了故障本体实例关系抽取的效果(F1值)。

基于路径与层次注意力的关系预测方法

目前,大多数基于路径的关系预测方法忽略了路径和关系之间的语义相关性,针对此问题提出一种层次注意力机制,捕获不同粒度的信息,包括关系级特征和路径级特征,在多跳路径和关系之间提供一种补充的语义关联。关系级注意力是以目标关系为导向,更多关注与目标关系相匹配的关系。路径级注意力是通过捕获给定实体对的邻域子图结构实现,为结果提供可解释性。该模型可以有效缓解噪声路径和冗余路径对关系预测任务的影响。实验结果表明,在相关数据集上,该模型的关系预测性能优于现有模型。

基于实体层次结构的文档级别关系抽取

针对目前文档级别关系抽取主要关注实体间的逻辑推理,未充分利用实体间的层次语义信息问题,提出一种基于实体层次结构的文档级别关系抽取模型。考虑多句文本中实体间的交互,将实体构建为文档图并使用图卷积神经进行信息传播;通过实体间的上下位关联构建实体层次树,使用注意力机制将层次语义信息融入实体;为降低模型对实体表面信息的关注,使用实体类型对实体词进行替换。实验结果表明,在大规模文档级别关系抽取数据集上实体语义信息增强的方案能够有效提高文档级别关系抽取的效果。

医联体模式下DRG知识教学对基层全科医师转诊行为的影响

目的探究医联体模式下疾病诊断相关分组(DRG)学习对基层全科医师转诊行为的影响作用。方法以河南省内某三甲医院及其医联体内的某二甲医院为研究对象,通过线上线下结合的方式对二甲医院全科医师进行DRG知识培训。培训结束后利用病案首页提取培训前后各1年的DRG相关数据,通过聚类分析了解这一时间内由该二甲医院全科医学科向上转诊至该三甲医院的所有患者病案首页相对权重(RW)值数据,并比较培训前后DRG组数、费用消耗指数及时间消耗指数、低风险组死亡率变化。结果DRG培训前的1年中由该科室转诊至三甲医院患者共137人,培训后1年转诊患者81人。通过K-均值聚类法将218人转诊的患者的RW值最终分为4个区间,其聚类中心值分别为0.71、1.33、2.36(F=209.07,P<0.05)。比较DRG培训前后转诊患者在每一区间内的占比,发现培训后0.71<RW值≤1.33区间内转诊患者有所降低,RW值>1.33区间内转诊患者占比较前升高(χ^(2)分别=10.17、22.80,P均<0.05),RW值≤0.71区间比例在培训前后无明显差异(χ^(2)=3.82,P>0.05)。同时,培训后转诊患者的费用消耗指数及时间消耗指数均较培训前有所上升,低风险死亡率无变化。结论医联体模式下提高全科医师DRG认知水平可对全科医师转诊行为产生积极的影响作用。

普外科护士分层管理中目标管理的实施效果研究

目的探讨采用目标管理方式完成普外科护士分层管理后获得的应用效果。方法选取2020年2月—2022年4月淄博一四八医院的20名普外科护士作为研究对象,以投掷硬币法分为两组,每组10名。其中施以常规护士分层管理的设为参照组,施以目标管理的设为研究组。对比两组普外科护士相关能力评分、护士对管理模式总满意度、管理质量评分。结果两组操作技能、临床护理思维、沟通能力以及理论知识评分比较,差异有统计学意义(P<0.05)。两组管理模式总满意度比较,差异无统计学意义(P>0.05)。两组目标制定、管理环节、管理技术、管理态度、管理细节、管理效果、管理人员同护理人员配合评分比较,差异有统计学意义(P<0.05)。结论医院对普外科护士在实施分层管理工作期间,目标管理工作的有效开展,可显著提高护士的操作技能、临床护理思维、沟通能力以及理论知识掌握能力,有效提升管理质量。

冷热电联产CCHP综合评价模型的研究

为了科学评价冷热电联产系统,本文提出了一种考虑工程多方面属性的综合评价方法—混合灰色关联多层次综合评价法。该方法综合了几种常用评价方法的优点,建立评价冷热电联产系统的综合评判模型,采用层次分析和灰色关联相结合的方法进行数据分析,并用模糊隶属关系将定性指标定量化,从而提高了优选方案的准确性。

基于词形规则模板的术语层次关系抽取方法

术语层次关系抽取是领域概念关系体系构建的重要基础。针对目前术语关系抽取中手工实现的问题，提出了基于词形规则模板匹配的术语层次关系抽取方法，实现从科技论文文本中抽取类属关系（IS-A）和整体部分关系（PART—OF）关系。利用复合术语的head和modifier特征，比较两个术语之间存在的边缘共用词汇，构造模板来确定它们之间的IS—A和PART-OF关系；提出泛化度指标，用于测量两个术语在概念层次树上的相对位置；提出相关度概念，用于测量两个术语之间在语义上的相关性。对不存在共用词汇和不匹配模板的术语采用泛化度差值和相关度来判断它们之间是否存在层次关系。实验从信息资源管理领域的论文文本中提取层次关系术语对1306对，准确率达到92．5％，证明提出的方法是有效的。

基于改进的凝聚性和分离性的层次聚类算法

由于传统的数据聚类算法都是在单一表上进行,因此如何在多表中进行聚类是现在聚类分析的一个新方向.提出了一种基于改进的凝聚性和分离性的层次聚类算法——ICCSH(a hierarchical clustering algorithmbased oni mproved cluster cohesion and separation) ,该算法首先通过ID传播把关系数据库中的各个表联系起来,再通过计算共享最近邻的相似度和改进的凝聚性算法将数据对象聚类为大量相对较小的子聚类,然后通过计算改进的簇间分离性合并子类来找到真正的结果簇.实验表明,该算法不仅运行时间相对较短,具有较强的可伸缩性,还可以得到较高精确的聚类结果.

关系数据库层次树查询机制浅析

大多数关系数据库应用系统的结构体系和查询要求都呈现很强的层次特征。在进行数据库查询过程中,用一般的查询机制来查询具有层次结构的数据是较为繁琐的。目前Oracle9i及以上版本和SQLServer2005都引入了层次树查询机制,层次树查询是一种针对关系数据库中层次结构数据查询的树型解决方案,可较好地解决此类复杂的层次查询问题。以Ora-cle9i数据库为例,结合某公司的部门组织结构,对层次树查询机制进行了实例分析与研究。