多值动态不确定因果图的推理算法研究

利用领域知识求解多值动态不确定因果图的联合概率分布所涉及的领域因果图的概率分布表达式构造难度大，针对这一问题，从2类因果循环图出发，提出一种基于图分解的推理算法。该算法极大地简化了全局概率分布表达式的构造过程，有效地降低了领域因果图概率分布表达式构造的难度。提高了多值动态不确定因果图的推理效率。     

采用特征识别技术的MBD模型自动语义标注方法

为了解决设计重用过程中基于模型的产品数字化定义(MBD)模型的问题,采用特征识别技术,对MBD模型自动添加语义标注,从而提高关键字检索的准确性.首先,对MBD模型的构成原理与模型要素进行分析,并在融合关键信息的基础上建立零件模型的属性面邻接图(AAG),根据零件模型上加工特征,将特征划分为螺钉头部特征、螺钉功能特征和材料特征.其次,通过顶点属性结合邻接矩阵重构图的顶点序列,动态编码结合距离匹配,求出最大公共子图,得出MBD模型之间的相似度.最后,利用聚类法实现对MBD模型的自动语义标注.实验结果表明:文中方法可以实现MBD模型的自动语义标注,很大程度上提高语义标注的自动化程度.     

多维时间序列Granger因果图Markov性

文章利用Granger因果图表示多维时间变量序列间的因果关系,图中的顶点集由序列的各个分量组成,顶点间的有向边表示分量序列间的Granger因果关系,无向边表示分量间的同期因果关系.建立Granger因果图的p-分离准则,研究Granger因果图的Markov性.     

直线驱动铰链机构的键合图因果关系分类整理

采用键合图方法对直线驱动铰链机构进行动力学建模时,传统的虚拟弹性铰接点或拉格朗日乘子法会引入新的动态环节导致出现刚性问题,势流关系混乱难以规范化和模块化.提出了一种针对该机构的键合图因果关系分类整理方法,通过元件接口势流因果分析及元件间连接规则整理两个方面,分类为液压缸、主部件和连杆三种元件,并分别统一各铰接点势流因果关系接口为根点、外力顶点、力输出点和负载点,使得刚体元件接口的因果关系得以延续.在此基础上建立了元件间连接的规范方法以简化建模过程.最后反铲挖掘机的建模及仿真示例证明了本方法对于工程机械的常见工作机构均可适用.     

基于分层排列图结构的流媒体应用层组播系统

为增强基于应用层组播的流媒体广播性能,该文提出构建具有分层排列图结构的应用层组播系统.在该结构中,参与主机组合成多个排列图,每一个排列图中嵌入多个独立组播树;排列图间形成树状结构,对应组播树相互衔接并保证独立性.流媒体数据采用多描述编码(MDC),每一描述数据通过一个独立的组播树传输,使得单个主机故障至多影响一个数据流,以提高系统流媒体服务的稳定性.实验表明,在节点故障率为5%时,分层排列图结构的容错能力与传统随机多组播树结构相比提高20%以上.利用拓扑测量技术,分层排列图结构将参与主机根据相互距离进行聚类,以提高系统的传输效率.     

校注本《歧路灯》存在问题初探

《歧路灯》长期以抄本形式流传,1980年中州书画社出版的栾星校注本是三十多年来最为通行的版本。校注本系由多种残本合校、补缀而成,同时并对底本作了删改,文字面貌不但与现存主要底本差异明显,而且与《校勘记》也很不吻合,不利于《歧路灯》的传播和研究。     

Semi-supervised learning based probabilistic latent semantic analysis for automatic image annotation

In recent years, multimedia annotation problem has been attracting significant research attention in multimedia and computer vision areas, especially for automatic image annotation, whose purpose is to provide an efficient and effective searching environment for users to query their images more easily.In this paper, a semi-supervised learning based probabilistic latent semantic analysis ( PL-SA) model for automatic image annotation is presenred.Since it' s often hard to obtain or create la-beled images in large quantities while unlabeled ones are easier to collect, a transductive support vector machine ( TSVM) is exploited to enhance the quality of the training image data.Then, differ-ent image features with different magnitudes will result in different performance for automatic image annotation.To this end, a Gaussian normalization method is utilized to normalize different features extracted from effective image regions segmented by the normalized cuts algorithm so as to reserve the intrinsic content of images as complete as possible.Finally, a PLSA model with asymmetric mo-dalities is constructed based on the expectation maximization( EM) algorithm to predict a candidate set of annotations with confidence scores.Extensive experiments on the general-purpose Corel5k dataset demonstrate that the proposed model can significantly improve performance of traditional PL-SA for the task of automatic image annotation.     

阮籍诗歌笺注述评

关于阮籍诗歌的笺注,自晋宋之交的颜延之注以来,一直受到诗论家、学者的重视,在诗歌注释史上占有重要地位。通过梳理,阮诗笺注脉络如下：最早的阮诗笺注是颜注和沈注。自唐李善注和五臣注,阮诗笺注以《文选》所录17首《咏怀》为主体。至清朝,阮诗笺注突破17首格局,阮诗笺注深入化。进入近代,阮诗笺注综合化、系统化。     

标注工具中OWL本体编辑功能的实现研究

基于本体的语义标注工具使用已有本体在Web页面中插入语义元数据信息,从而使Web页的内容机器可识别,是将现有Web提升为语义Web的有效方法之一.大多数标注工具仅支持使用已存在的本体词汇来标注Web页,不具备标注过程中新建本体或在已有本体中添加新词汇从而补充标注词汇的本体编辑功能.针对语义标注中的本体编辑功能的特点及存在问题进行研究,并从实现的角度对OWL本体编辑进行讨论.     

连续语流中重音标注和大规模重音语料库建设

规范的重音标注将对重音研究和语音合成技术产生积极影响。该文通过对大规模语料进行语流中词重音和句重音标注,总结出了重音标注的规则。重音标注包括标注训练、正式标注、语料库建立3个步骤。通过标注训练明确重音概念和标注方法,词重音采取词内对比法,所谓"重"主要指突显度和饱满度;句重音采取呼吸句群单位标注法,注重标注的层级性,不跨句群进行重音知觉对比。重音标注完成后,重音语料库也随之建立起来。这个大规模的重音语料库具备词重音和句重音信息,适用于语音合成领域,能较好地提高合成语音的自然度。     

Semantic image annotation based on GMM and random walk model

Automatic image annotation has been an active topic of research in computer vision and patternrecognition for decades.A two stage automatic image annotation method based on Gaussian mixturemodel (GMM) and random walk model (abbreviated as GMM-RW) is presented.To start with,GMM fitted by the rival penalized expectation maximization (RPEM) algorithm is employed to estimatethe posterior probabilities of each annotation keyword.Subsequently, a random walk processover the constructed label similarity graph is implemented to further mine the potential correlations ofthe candidate annotations so as to capture the refining results, which plays a crucial role in semanticbased image retrieval.The contributions exhibited in this work are multifold.First, GMM is exploitedto capture the initial semantic annotations, especially the RPEM algorithm is utilized to train themodel that can determine the number of components in GMM automatically.Second, a label similaritygraph is constructed by a weighted linear combination of label similarity and visual similarity ofimages associated with the corresponding labels, which is able to avoid the phenomena of polysemyand synonym efficiently during the image annotation process.Third, the random walk is implementedover the constructed label graph to further refine the candidate set of annotations generated byGMM.Conducted experiments on the standard Corel5k demonstrate that GMM-RW is significantlymore effective than several state-of-the-arts regarding their effectiveness and efficiency in the task of automatic image annotation.     

VTC-KSVD:一种融合视觉特征与标签一致性的多标签图像标注方法

提出一种融合视觉特征及标签一致性的多标签图像标注方法VTC-KSVD.首先通过K均值奇异值分解（KSVD）法建立图像的标签一致性模型TC-KSVD,然后将多视图特征融合在该模型中.该方法既利用了训练样本的类标与编码系数的判别式模型,又利用了训练样本的标签与编码系数的关系,增加了字典的判别性,提高了标注性能.在Corel5K数据集上的实验结果表明,融合了多视图视觉特征与标签一致性的VTC-KSVD方法可以较为准确地找到视觉特征与语义特征均相似的图像近邻,能明显提升多标签图像的标注性能,并能有效缓解训练数据有限而引起的稀疏性问题.      

黄节《谢康乐诗注》的笺注动机和特色

黄节先生曾致力于汉魏六朝诗注,成就斐然。《谢康乐诗注》便是其中的经典注释之一种。黄节先生之所以以诗为业,写诗注诗,一是为教化人心、匡正世风;二为保存国粹、复兴古学,重振"国家特别之精神";三是心系康乐诗散亡、不易为人理解之忧而纠谬释误;四是与谢灵运精神相通、情感契合所致。《谢康乐诗注》继承李善《文选》注的传统,扬长避短;以史论诗、以诗注诗;案语严谨折衷、集众家所长。对于地名和声韵的注释,也颇具特点。以上笺注特色均为今人读解谢诗带来了很大帮助,具有极高的学术价值。     

Exploring Attack Graphs for Security Risk Assessment: A Probabilistic Approach

The attack graph methodology can be used to identify the potential attack paths that an attack can propagate. A risk assessment model based on Bayesian attack graph is presented in this paper. Firstly, attack graphs are generated by the MULVAL (Multi-host, Multistage Vulnerability Analysis) tool according to sufficient information of vulnerabilities, network configurations and host connectivity on networks. Secondly, the probabilistic attack graph is established according to the causal relationships among sophisticated multi-stage attacks by using Bayesian Networks. The probability of successful exploits is calculated by combining index of the Common Vulnerability Scoring System, and the static security risk is assessed by applying local conditional probability distribution tables of the attribute nodes. Finally, the overall security risk in a small network scenario is assessed. Experimental results demonstrate our work can deduce attack intention and potential attack paths effectively, and provide effective guidance on how to choose the optimal security hardening strategy.     

基于视觉内容与语义相关的图像标注模型

针对当前标注系统的不足,设计了一种高效的标注模型,其标注步骤包括标注和标注改善,标注算法采用加权的正反例标志向量法,标注改善采用NGD方法。实验表明,标注效率远优于经典的标注模型,标注质量优于大多数标注模型。     

一类带有纽结分支的边—不交链图

为进一步探讨边—不交链图的分类,研究其中至少有一个链的分支是非平凡纽结的情形,给出了带有纽结分支的边—不交链图的定义。在给出内在纽结图H0的基础上,利用其与边组成的图形成完全图K7,并采用该方法构造出一类带有纽结分支的边—不交链图H（43）。分析点扩张对H（43）的作用,得到了点扩张的图变换不保持带有纽结分支的边—不交链图的性质这一结论。     

图谱法在反倾销预警机制中的应用

反倾销预警机制是通过监测进口产品的价格、销量、份额等因素，对国内企业或产业的同类产品带来的影响进行跟踪，并适时发出预警，以便能为反倾销工作提前做好各种准备。图谱法主要是利用逻辑图形(即进口产品价格变化图、国内同类产品价格变化图以及市场份额变化图)的方法进行预警，它是以数据分布和逻辑判断为基础的动态直观的预警方法。     

藏语语音合成中语料数据标注规则的设计

基于隐马尔可夫模型（HMM）的藏语语音合成系统中,语料数据的标注是训练声学模型的基础和关键。文章结合藏语独有的语音特点对语料数据设计套标注规则,并以Praat为标注环境实现语料的标注。     

基于排序学习的文本概念标注方法研究

提出一种基于排序学习的方法CRM (concept ranking model), 来实现文档的维基百科概念自动标注。首先人工对一定规模的文档进行概念标注, 建立训练集合, 然后利用排序学习算法在多项特征上得到对概念排序的模型, 利用这个概念的排序模型对任意文档进行概念标注。实验表明, 相对于传统的文档概念标注方法, 此方法在各类指标上都有相当大的提高, 标注结果更加接近人类的概念标注。     

几类图的Kirchhoff指标

求出了基于圈或路的多重星图和多重完全图的Laplace特征多项式,并利用图的Kirchhoff指标与其补图的Laplace特征多项式之间的关系,得到了基于圈或路的多重星图和多重完全图的相关图的Kirchhoff指标的计算公式