基于异质属性融合的危重疾病二阶段预测模型

随着电子健康记录(Electronic Health Record,EHR)的出现与广泛应用,基于EHR数据的预测模型可以起到早期检测和干预疾病的作用。异质属性在EHR数据中普遍存在,但是难以做到深度利用,因此可通过对数据样本进行异质属性融合的方法,为后续模型训练提供信息丰富的数据表征基础。本文设计一种高效的二阶段预测模型,用于解决重疾预测中存在的时效与成本等问题。该模型的第一阶段对病例样本进行粗粒度预测,将危重程度低的病例进行疾病初筛,起到提前分流病人的作用;第二阶段模型则基于第一阶段的粗滤结果,对潜在的危重病例进行更细粒度的预测。通过实验验证,经过异质属性融合处理后,在选择前6个时间点构造非时序模型时,二阶段模型可以较好地兼具疾病初筛以及疾病预测的效果。

基于PPMI的异质属性网络嵌入

属性网络嵌入旨在映射网络中的节点和链接关系到低维空间,同时保留其固有的结构和属性特征.异质属性网络中多种类型的节点和链接关系给网络嵌入学习提供了丰富的辅助信息,同时也带来了新的挑战.提出异质属性网络嵌入模型(heterogeneous attribute network embedding based on the PPMI,HANEP),旨在将网络中多种类型的节点和(或)多种类型的链接关系映射到低维、紧凑的空间,同时保护节点的属性特征和不同类型对象之间的异质链接承载的复杂、多样且丰富的语义信息.HANEP模型首先基于样本属性的相似性构建属性图、依据元路径抽取异质属性网络的拓扑结构,然后通过随机冲浪获得属性和拓扑概率共现(probabilistic co-occurrence,PCO)矩阵,并计算其正点对互信息(positive point-wise mutual information,PPMI),进而采用多个自编码器(auto-encoder,AE)捕捉节点属性和异质链接的本质信息.元路径可以捕捉异质网络中多种类型节点间的链接关系,构建属性图可以清晰描述节点属性的非线性流行结构,属性和拓扑的局部成对约束和图表示有助于整合节点属性和网络拓扑的一致性和互补性关系,PPMI表示可以捕捉属性和拓扑的高阶近邻信息及潜在的复杂非线性关系.在3个真实数据集上的实验结果验证了HANEP算法的有效性.

“异托邦”视野下艺术群落的社会异质属性

法国社会学家福柯的空间哲学体系体现了他对人类社会中异质空间的重新思考和独特认识。以福柯的理论进行观照,城市落败地区出现的艺术群落是一种介于现实与理想间的异托邦空间,其发展与变迁的过程具有特殊的内在逻辑。作为一种异质性空间,这些艺术群落具有话语权上的反抗意义、时空上的并置能力、精神上的补偿效用,系统上的开闭特质以及社会上的规训功能等社会属性。

融合多特征的属性异质网络嵌入方法

网络嵌入旨在用低维、实值的向量表示非结构化网络中的节点,使节点嵌入尽可能地保留原始网络中的结构特征与属性特征。然而,当前研究主要集中于嵌入网络结构,对异质信息网络中具有丰富语义的关系属性和节点属性考虑得较少,可能导致节点嵌入语义缺失,从而影响下游应用的预测效果。针对该问题,设计了一种融合多特征的属性异质网络嵌入(Attributed Heterogeneous Network Embedding with Multiple Features,MFAHNE)方法。该方法通过序列采样、结构特征嵌入、属性特征嵌入、特征融合等步骤将网络中的关系属性、节点属性、结构语义等特征融合至最终节点嵌入。实验结果表明,该方法能兼顾结构特征与属性特征,实现两种特征信息的相互补充,优于传统的网络嵌入方法。

企业自播电商模式下企业与主播的双边匹配博弈研究

提高企业与主播的匹配度是增强双方满意度和效益的关键一步。以第三方直播平台为双边平台建立匹配博弈模型,运用Borda分值测算匹配度,考虑异质属性、满意度、稳定性等相关条件对匹配过程的影响建立效益最大化双边匹配模型,分析两企业均衡产量及主播的最优直播努力水平,探讨企业异质性系数和主播直播努力水平对匹配效益的影响。研究发现:以效益最大化为目标函数的双边匹配博弈模型更加符合企业实际情况;企业异质性系数对双方效益存在正向影响;在较高的直播努力水平下,企业和主播的收益随直播努力水平的升高而降低;最优直播努力水平与企业异质性系数存在相关性,且随企业异质性系数的增加而提高;市场规模在主播直播努力程度对销量和效益的影响过程中存在调节作用。因此,直播平台应综合考虑市场规模、主播直播努力水平及企业异质性系数对匹配过程的影响,与差异化企业进行合作。主播要根据市场规模及企业异质性系数及时调整努力水平。企业要制定差异化产品策略,提高产品核心竞争力以维持市场份额,培训个性化主播,赋予产品内涵,使其产品信息具备鲜明特色。

异质投资者股价影响研究综述

文章从信念异质与属性异质两个角度梳理了投资者异质行为对股价影响的现有研究成果。其中,对异质信念的综述围绕预期收益分歧度、收益率波动性及成交量三个代理变量展开,大多得到前两者与股票超额收益率负相关、成交量与投资收益正相关的实证结论;行为异质主要将股市交易者划分为理性投资者与正反馈交易者两类,大多研究得出正反馈交易者对股价、股市的扰动作用。通过综述得到理论与现实两方面启示:一是后续研究可尝试将价值投资者纳入行为框架,深入考察其股价作用;二是对个人交易者应加强教育,促进股价平稳、股市健康。

The algorithm of 3D multi-scale volumetric curvature and its application

To fully extract and mine the multi-scale features of reservoirs and geologic structures in time/depth and space dimensions, a new 3D multi-scale volumetric curvature （MSVC） methodology is presented in this paper. We also propose a fast algorithm for computing 3D volumetric curvature. In comparison to conventional volumetric curvature attributes, its main improvements and key algorithms introduce multi-frequency components expansion in time-frequency domain and the corresponding multi-scale adaptive differential operator in the wavenumber domain, into the volumetric curvature calculation. This methodology can simultaneously depict seismic multi-scale features in both time and space. Additionally, we use data fusion of volumetric curvatures at various scales to take full advantage of the geologic features and anomalies extracted by curvature measurements at different scales. The 3D MSVC can highlight geologic anomalies and reduce noise at the same time. Thus, it improves the interpretation efficiency of curvature attributes analysis. The 3D MSVC is applied to both land and marine 3D seismic data. The results demonstrate that it can indicate the spatial distribution of reservoirs, detect faults and fracture zones, and identify their multi-scale properties.

基于异质信息的被执行人的评估决策问题研究

针对异质信息下被执行人财产隐匿行为甄别评估问题,考虑到被执行人评价指标关联性和执行人员的心理行为,提出一种基于复杂网络和前景理论的异质多属性决策方法.首先,对被执行人进行定量与定性分析,建立被执行人多维评价指标体系,进而利用复杂网络确定评价指标的权重;然后,考虑到被执行人评价指标的异质性,评价指标信息采用精确数、语言变量、概率语言等异质信息表示,在此基础上考虑决策者心理行为,利用前景理论计算各个被执行人的前景值,确定重点查控对象;最后,通过具体案例验证所提出方法的有效性和可行性,切实推动解决“执行难”问题,并为解决具有异质信息的实际管理决策问题提供新思路和新方法.

3D modeling of geological anomalies based on segmentation of multiattribute fusion

3D modeling of geological bodies based on 3D seismic data is used to define the shape and volume of the bodies, which then can be directly applied to reservoir prediction, reserve estimation, and exploration. However, multiattributes are not effectively used in 3D modeling. To solve this problem, we propose a novel method for building of 3D model of geological anomalies based on the segmentation of multiattribute fusion. First, we divide the seismic attributes into edge- and region-based seismic attributes. Then, the segmentation model incorporating the edge- and region-based models is constructed within the levelset- based framework. Finally, the marching cubes algorithm is adopted to extract the zero level set based on the segmentation results and build the 3D model of the geological anomaly. Combining the edge-and region-based attributes to build the segmentation model, we satisfy the independence requirement and avoid the problem of insufficient data of single seismic attribute in capturing the boundaries of geological anomalies. We apply the proposed method to seismic data from the Sichuan Basin in southwestern China and obtain 3D models of caves and channels. Compared with 3D models obtained based on single seismic attributes, the results are better agreement with reality.

Leukolysin/MMP25/MT6-MMP: a novel matrix metalloproteinase specifically expressed in the leukocyte lineage

A novel matrix metalloproteinase (MMP) was identified from leukocytes and found to be specifically expressedby peripheral blood leukocytes among 29 different tissuesexamined. Named leukolysin, it encodes for 562 residueswith a conserved MMP structure, i.e., pre-, pro-, catalytic , hinge- and hemopexin-like domains, but also a RXK/RRmotif, known for its role in MMP zymogen activation, anda C-terminal hydrophobic segment. Overall, leukolysindisplays the strongest homology to the newly identifiedMT-MMP subgroup with 45% and 39% identities to MT4and MT1-MMPs vs 30% and 31.5% to MMP1 and 3 respectively. Unlike MT4-MMP whose proteolytic activityremains undefined, a C-terminally truncated leukolysin isexpressed as a strong gelatinolytic species at 28 kDa whichis derived from a cell-associated 34 kDa proenzyme, presumably by furin or proprotein convertase mediated removal of the propeptide (-6 kDa). By green fluorescentprotein (GFP) tagging, the intracellular proenzyme is localized to granules throughout the cell, suggesting thatactivation occur immediately prior to secretion. Taken together, leukolysin may be part of the proteolytic arsenaldeployed by leukocytes during inflammatory responses.

Analysis of Price and Industry Dynamics of Sustainable and Specialty Coffee in Costa Rica

Volatility of commodity prices has affected dramatically the coffee industry in recent years, particularly small holder farmers. Differentiation of coffee through certification, such as sustainahility and quality attributes, has been proposed as a strategy for protection of the farmers against volatility in the international prices. This research paper evaluated three different models to explore the effectiveness of the differentiation strategies in protecting the farmer against price volatility in recent years, focusing on the case of Costa Rica. Evidence showed important differences in the price dynamics over time when comparing three groups of coffee.

Properties of color-flavor locked strange quark matter and strange stars in a new quark mass scaling

Considering the effect of one-gluon-exchange interaction between quarks,the color-flavor locked strange quark matter and strange stars are investigated in a new quark mass density-dependent model.It is found that the color-flavor locked strange quark matter can be more stable if the one-gluon-exchange effect is included.The lower density behavior of the sound velocity in this model is different from the previous results.Moreover,the new equation of state leads to a heavier acceptable maximum mass,supporting the recent observation of a compact star mass as large as about 2 times the solar mass.

Structural heterogeneity and deformation rheology in metallic glasses

The atomic structure and associated deformation behavior of metallic glasses(MGs)have been long standing issues.Although recent computational/experimental results indicate that the structure of MGs is heterogeneous at the nano scale,the fundamental knowledge of the atomic basis for such structural heterogeneity and its impact on the overall properties of MGs is still lacking.We reviewed recent research on unraveling the structure heterogeneity in MGs,with emphases on the use of dynamic atomic force microscopy,the characterization of glass anelasticity by nanoindentation,and the establishment of numerous correlations with structural heterogeneity.

Melamine-assisted synthesis of ultrafine Mo_(2)C/Mo_(2)N@N-doped carbon nanofibers for enhanced alkaline hydrogen evolution reaction activity

Noble metal-free electrocatalysts with high activity are highly desirable for the large-scale application of hydrogen evolution reaction(HER). Mo_(2) C-based nanomaterials have been proved as a promising alternative to noble metal-based electrocatalysts owing to the Pt-resembled d-band density and optimal intermediates-adsorption properties.However, the aggregation and excessive growth of crystals often occur during their high-temperature synthesis procedure, leading to low catalytic utilization. In this study, the ultrafine Mo_(2) C/Mo_(2) N heterostructure with large surface and interface confined in the N-doped carbon nanofibers(NCNFs) was obtained by a melamine-assisted method. The synergistic effect of Mo_(2) C/Mo_(2) N heterostructure and plenty active sites exposed on the surface of ultrafine nanocrystals improves the electrocatalytic activity. Meanwhile, the N-CNFs ensure fast charge transfer and high structural stability during reactions. Moreover, the in-situ synthesis method strengthens the interfacial coupling interactions between Mo_(2)C/Mo_(2) N heterostructure and N-CNFs, further enhancing the electronic conductivity and electrocatalytic activity. Owing to these advantages, Mo_(2)C/Mo_(2) N@N-CNFs exhibit excellent HER performance with a low overpotential of 75 mV at a current density of 10 mV cm^(-2) in alkaline solution, superior to the single-phased Mo_(2)C counterpart and recently reported Mo_(2)C/Mo_(2) N-based catalysts. This study highlights a new effective strategy to design efficient electrocatalysts via integrating heterostructure, nanostructure and carbon modification.