聚集数据多元线性模型参数的多元聚集综合岭估计的相对效率

对于聚集数据的多元线性模型,提出了参数的多元聚集综合岭估计的概念,给出了多元聚集综合岭估计相对于最小二乘估计及最佳线性无偏估计的两种相对效率,并得到了这两种相对效率的上界.应用Monte Carlo模拟,验证了有关结论是合理的.     

Calculations of Energy-Loss Function for 26 Materials

We present a fitting calculation of energy-loss function for 26 bulk materials, including 18 pure elements (Ag, Al, Au, C, Co, Cs, Cu, Er, Fe, Ge, Mg, Mo, Nb, Ni, Pd, Pt, Si, Te) and 8 compounds (AgCl, Al₂O₃, AlAs, CdS, SiO₂, ZnS, ZnSe, ZnTe) for application to surface electron spectroscopy analysis. The experimental energy-loss function, which is derived from measured optical data, is fitted into a finite sum of formula based on the Drude-Lindhard dielectric model. By checking the oscillator strength-sum and perfectscreening-sum rules, we have validated the high accuracy of the fitting results. Furthermore, based on the fitted parameters, the simulated reflection electron energy-loss spectroscopy (REELS) spectrum shows a good agreement with experiment. The calculated fitting parameters of energy loss function are stored in an open and online database at http://micro.ustc.edu.cn/ELF/ELF.html.     

基于Tucker3分解的三路数据聚类方法

从两路数据聚类分析到三路数据聚类分析实质上是由平面分析到立体分析的过程。三路数据聚类方法研究的核心之一是如何把传统的两路截面数据聚类技术向三路数据聚类扩展的问题。本文基于Tucker模型的思路,提出一种先对三路数据执行矩阵分解,而后进行聚类分析的三路数据聚类方法。这种方法不但能够通过核心矩阵反映三路数据三个模式信息联系的强度大小,而且还可以在一个分解框架下对三路数据的三个模式同时进行聚类分析。实证分析结果表明,本文提出的聚类方法不但灵活、易于理解,同时也有着良好的判别性和实用性。     

����EM�㷨��II�ͻ��ɾʧ���ݵIJ�������

??In this paper, we estimate parameters of gamma life distribution and normal life distribution by EM algorithm based on Type-II hybrid censored data. The covariance matrices are derived as well. Some numerical examples are also presented for illustration.     

基于主题模型的半监督网络文本情感分类研究

针对网络评论文本的情感分类问题中存在的数据的不平衡性、无标记性和不规范性问题,提出一种基于主题的闽值调整的半监督学习模型,通过从非结构化文本中提取主题特征,对少量标注情感的文本训练分类器并优化指标调整闽值,达到识别用户评论的情感倾向的目的。仿真研究证明阈值调整的半监督模型对数据非平衡性和无标记性具有较强的适应能力。在实证研究中,对酒店评论文本数据构建的文本情感分类器显示该模型可以有效预测少数类评论样本的情感极性,证实了基于主题模型的闽值调整半监督网络评论文本情感分类模型在实际问题中的适用性与可行性。     

一般分布区间型符号数据的SOM聚类方法

区间型符号数据是一种重要的符号数据类型,现有文献往往假设区间内的点数据服从均匀分布,导致其应用的局限性。本文基于一般分布的假设,给出了一般分布区间型符号数据的扩展的Hausdorff距离度量,基于此提出了一般分布的区间型符号数据的SOM聚类算法。随机模拟试验的结果表明,基于本文提出的基于扩展的Hausdorff距离度量的SOM聚类算法的有效性优于基于传统Hausdorff距离度量的SOM聚类算法和基于μσ距离度量的SOM聚类算法。最后将文中方法应用于气象数据的聚类分析,示例文中方法的应用步骤与可操作性,并进一步评价文中方法在解决实际问题中的有效性。     

基于安德雷德通式的旋转黏度计校准方法研究与应用

基于安德雷德通式提出两种旋转黏度计校准方法:安德雷德通式运用法和标准黏度液定值数据拟合法。经过实验验证,安德雷德通式运用法得到旋转黏度计计算黏度示值同实测黏度示值的相对误差在±1%之内;标准黏度液定值数据拟合法得到的旋转黏度计计算修正系数同实测修正系数的相对误差为0.2%~0.4%。安德雷德通式运用法和标准黏度液定值数据拟合法在实际旋转黏度计校准工作中可行,可以提高校准效率。     

Purine 3′:5′‐cyclic nucleotides with the nucleobase in a syn orientation: cAMP,cGMP and cIMP

Purine 3′:5′‐cyclic nucleotides are very well known for their role as the secondary messengers in hormone action and cellular signal transduction. Nonetheless, their solid‐state conformational details still require investigation. Five crystals containing purine 3′:5′‐cyclic nucleotides have been obtained and structurally characterized, namely adenosine 3′:5′‐cyclic phosphate dihydrate, C₁₀H₁₂N₅O₆P·2H₂O or cAMP·2H₂O, (I), adenosine 3′:5′‐cyclic phosphate 0.3‐hydrate, C₁₀H₁₂N₅O₆P·0.3H₂O or cAMP·0.3H₂O, (II), guanosine 3′:5′‐cyclic phosphate pentahydrate, C₁₀H₁₂N₅O₇P·5H₂O or cGMP·5H₂O, (III), sodium guanosine 3′:5′‐cyclic phosphate tetrahydrate, Na⁺·C₁₀H₁₁N₅O₇P⁻·4H₂O or Na(cGMP)·4H₂O, (IV), and sodium inosine 3′:5′‐cyclic phosphate tetrahydrate, Na⁺·C₁₀H₁₀N₄O₇P⁻·4H₂O or Na(cIMP)·4H₂O, (V). Most of the cyclic nucleotide zwitterions/anions [two from four cAMP present in total in (I) and (II), cGMP in (III), cGMP⁻ in (IV) and cIMP⁻ in (V)] are syn conformers about the N‐glycosidic bond, and this nucleobase arrangement is accompanied by C_rib—H…N_pur hydrogen bonds (rib = ribose and pur = purine). The base orientation is tuned by the ribose pucker. An analysis of data obtained from the Cambridge Structural Database made in the context of syn–anti conformational preferences has revealed that among the syn conformers of various purine nucleotides, cyclic nucleotides and dinucleotides predominate significantly. The interactions stabilizing the syn conformation have been indicated. The inter‐nucleotide contacts in (I)–(V) have been systematized in terms of the chemical groups involved. All five structures display three‐dimensional hydrogen‐bonded networks.     

C语言在落球法测液体粘度实验中的应用

在大数据时代,为了提高物理实验数据的处理效率,减少数据处理过程中的人为误差,本文以落球法测液体的粘滞阻力系数实验为载体,运用高效率、结构化、模块化的编程语言,采取面向过程的编程技术,编写了适用于落球法测液体粘度实验的C语言程序,并根据相关理论计算了了粘滞阻力系数的相对不确定度,从而进一步地阐述了C语言程序设计在物理实验数据处理中的可行性、优越性、可推广性以及适用性,提出了Excel与C语言结合、高效率处理物理实验数据的方法。