业务模型驱动的应急物流本体知识表示及应用

为解决应急物流智能协作面临的异构信息融合与知识管理难题,以有效知识应用为目标,利用业务模型与本体模型对组成构件上的相关性进行映射,并根据系统控制与实体行为约束所需的业务推理规则,设计了3种模块之间的关系与四阶段转换步骤,形成了业务模型驱动的本体表示方法.以该方法为基础,结合应急物流领域需求,建立了由概念集合、概念属性、语义约束、推理规则等要素组成的应急物流本体模型.通过应急配送路径选择案例,自动推理得到可替代邻近点以完成配送路径求解.结果表明,应急物流本体模型及其应用实现了异构信息的本体融合,消除了以关键词匹配为基础的常规检索失效现象,满足了最小路径算法的执行条件,为应急物流决策模型及算法应用提供了知识层面的支持.      

Web语义的分类和缓存

针对Web信息查询缺乏对查询概念的相关语义这一问题，引入一种本体的形式化模型，通过引入本体内聚度等概念建立本体中概念间的语义关联，从而便于对常识知识的处理，并作为语义扩展及缓存的依据，由此建立起查询的上下文关联及联想，给查询带来较好的适应性。     

METADATA EXPANDED SEMANTICALLY BASED RESOURCE SEARCH IN EDUCATION GRID

With the rapid increase of educational resources, how to search for necessary educational resource quickly is one of most important issues. Educational resources have the characters of distribution and heterogeneity, which are the same as the characters of Grid resources. Therefore, the technology of Grid resources search was adopted to implement the educational resources search. Motivated by the insufficiency of currently resources search methods based on metadata, a method of extracting semantic relations between words constituting metadata is proposed. We mainly focus on acquiring synonymy, hyponymy, hypernymy and parataxis relations. In our schema, we extract texts related to metadata that will be expanded from text spatial through text extraction templates. Next, metadata will be obtained through metadata extraction templates. Finally, we compute semantic similarity to eliminate false relations and construct a semantic expansion knowledge base. The proposed method in this paper has been applied on the education grid.     

生物医学信息系统及Ontology构建

介绍了生物医学信息系统的设计、实现及其功能，并讨论了ontology及知识库构建的相关问题．基于现有的浪潮天梭TS10000集群系统，采用数据处理的新技术和新方法构建的这一用户化的、可扩展的综合系统，支持用神经影像方法来对脑结构、功能和疾病进行研究；为新的知识发现提供大量的文献数据集及神经影像ontology和基础知识库．     

TOWARDS AN ONTOLOGY-BASED MULTI-AGENT MEDICAL INFORMATION SYSTEM BASED ON THE WEB

Introduction　　 Medical information systems cover a numberof medical domains to satisfy different medical ap-plications and characterized by the rich functionali-ties and valuable medical case data[1,2 ] .But infor-mation systems are heterogeneous and lack of effi-cient knowledge organization mechanisms.So theredundant information is existing,and real time in-formation acquisition and updating is difficult andmedical information systems are hard to interoper-ate and implementcomplex medical a…     

特定领域中语义校对系统的开发

介绍应用于报刊书籍政治性错误校对的一个语义校对软件系统的设计原理和实现过程．本系统先把文本切分成词语，找出敏感词，合并成术语，部分术语通过短语本身或左右片段即可做出正误判断；另一部分术语则用整个句子语义与知识库进行模糊匹配得到它的错误隶属度．     

基于时序融合的自动驾驶多任务感知算法

采用连续图像帧作为输入,挖掘连续图像帧之间的时序关联信息,构建一种融合时序信息的多任务联合驾驶环境视觉感知算法,通过多任务监督联合优化,实现交通参与目标的快速检测,同时获取可通行区域信息;采用ResNet50作为骨干网络,在骨干网络中构建级联特征融合模块,捕捉不同图像帧之间的非局部远程依赖关系,将高分辨率图像通过卷积下采样处理,加速不同图像帧的特征提取过程,平衡算法的精度和速度;在不同的图像帧中,为了消除由于物体运动产生的空间位移对特征融合的影响,且考虑不同图像帧的非局部关联信息,构建时序特征融合模块分别对不同图像帧对应的特征图进行时序对齐与匹配,形成融合全局特征;基于共享参数的骨干网络,利用生成关键点热图的方法对道路中的行人、车辆和交通信号灯的位置进行检测,并利用语义分割子网络为自动驾驶汽车提供道路可行驶区域信息。研究结果表明:提出的感知算法以多帧图像代替单一帧图像作为输入,利用了多帧图像的序列特性,级联特征融合模块通过下采样使得计算复杂度降低为原来的1/16,与CornerNet、ICNet等其他主流模型相比,算法检测精确率平均提升了6%,分割性能平均提升了5%,并保持了每秒12帧图像的处理速度,在检测与分割速度和精度上具有明显优势。      

Online Mind-Map as Interface of Electronic Resource Integration and Sharing

The paper aims to research on online mind map as interface of learning resource integration and sharing.We developed a web application that provides a interface to allow users to create a knowledge mind map.With the online mind mapping tool,the electronic learning resource will be discovered and shared in the form of subject-oriented package,it is more flexibly than the other E-Learning systems,and the interaction and cooperation of the users are strengthened to some extent.     

基于单片机的直流电机控制实验装置设计

该设计采用单片机构成直流电动机PWM调速系统,其调速过程是由单片机输出控制信号即PWM波形来控制电动机,驱动电路输出电压来驱动电动机工作在不同的运行状态。其运行状态经由反馈电路进行数据采集,并将采集信息反馈给单片机进行数据分析处理,并将当前的电动机运行状态经显示电路显示出来。     

轮机英语中介词的语义动化

语义范畴的转换,尤其是由静态到动态的转换,是英语介词与汉语介词的重要区别之一。工科专业英语对简洁度和准确度的特别要求使得轮机英语中大量存在语义动化的介词。这些介词在语义指涉、句法功能、翻译方法、所接宾语等方面存在一定的特点,了解这些特点有助于学生更加准确理解并灵活运用轮机英语。     

基于无监督学习的部分-整体关系获取

针对面向中文自由文本的部分-整体关系抽取问题,提出一种基于无监督学习的方法. 首先提出子模式提取算法,从领域文本集中获取概念对和概念对所在上下文模式,利用概念对和概念对上下文模式建立分布式语义模型;然后采用协同聚类算法将具有相同语义关系的概念对聚合成簇,通过训练L1正则化逻辑回归模型提取簇的特征并得到代表每个簇语义关系的概念对上下文模式;最后根据模式识别表达部分-整体关系的簇,从而获取部分-整体关系概念对. 实验结果表明,该方法取得较好的性能,F度量达到68.97%,优于传统聚类方法（55.77%）和模式匹配方法（61.95%）.      

Modular design of complex product based on the triple fusion of function/feature/knowledge

The modular design technology is of importance increasingly, as product structure is more and more complex. Modular design systems face challenging problems as the design information tends to be dynamic, redundant, and very large. This paper describes a novel approach for handling them. In this approach, a partition is firstly performed for the complex structural components by mapping functions to the structures layer by layer. Based on this partition, a comprehensive design matrix is then developed to identify the key design mode which is driven by a special function. The design process is also programmed by analyzing the coupled information on both the functional and structural hierarchies. Then, the integrated knowledge model based on object-oriented method and hybrid inference method is constructed. In this model, knowledge can be organized at hierarchical classification and expressed with different forms. Finally, the methodology developed has been applied to a real application in automobile cylinder block design and the results are presented.     

Web数据库中最小失败子查询问题研究

随着网络的普及,越来越多的网络用户通过Web数据库提供的查询接口进行数据查询。但是用户往往会遇到发出查询条件而没有返回结果的现象即空查询结果问题。造成这种情况的原因可能是一方面用户查询意图模糊,不能够准确地描述查询要求。另一方面,用户明确查询需求但Web数据库却找不到任何符合查询条件的结果。针对这种情况可以分析发现找到造成查询失败的子查询条件是解决问题的关键。本文的主要研究分两个步骤:首先,将用户发出的查询条件分解为单个的原子查询条件;然后找到造成查询失败的最小失败子查询(Minimal Failing Sbuquery,MFS)集合。     

基于APTS大数据的城市公交出行多维分析模型和方法

智能公交系统(Advanced Public Transportation Systems,APTS)数据具有海量、类型多样等大数据的典型特征,对其进行分析和挖掘可能获得丰富的公交出行特征和规律.构建基于APTS大数据的公交出行多维分析框架,在计算乘客出行时空信息(上车、下车和换乘)的基础上,建立包含4个维度(乘客、时间、空间和行为)的公交出行数据模型,系统提出基于5种联机分析处理方法的公交出行分析内容.应用APTS大数据对模型和方法进行了实验和验证,研究结果表明,该方法能够便捷地分析不同维度、不同粒度的公交出行信息,不仅能够应用于公交乘客出行行为的研究,还能够为城市公交系统的规划和管理提供决策支持.     

控制“自我”意识的脑部区域发现之意义提示

人类“自我”意识产生的物理基础是什么?意识的本质究竟是什么?近年来，脑科学的迅猛发展在一定程度上回答了这些问题。本通过对科学家发现了控制“自我”意识的大脑区域这一医学新成果的分析，考察了历史上有关意识问题的多种理论的局限性，揭示了该成果蕴含的哲学意义是对辨证唯物主义关于“人脑是产生意识的物理基础”内容的补充、丰富和完善。     

Incrementally Exploiting Sentential Association for Email Classification

Introduction Email classification can be cast into text classifi-cation (TC). Many methods have been proposed forTC, which can be extended to Email domain. Popularmethods include NaiveBayes, decision trees, supportvector machine (SVM) and association rule…     

一种票务服务领域的本体模型设计

现有的Web服务缺少明显的形式语义支撑,而Web服务发现又因缺乏对服务质量(QoS)的描述而难以为用户选择最佳服务,为此,本文提出了一种基于QoS的Web服务发现本体模型,采用分类和反馈机制,在服务发现过程中添加服务质量本体信息,并针对票务服务领域进行了实例设计和语义描述。     

Automatic semantic analysis of software requirements through machine learning and ontology approach

Nowadays, software requirements are still mainly analyzed manually, which has many drawbacks (such as a large amount of labor consumption, inefficiency, and even inaccuracy of the results). The problems are even worse in domain analysis scenarios because a large number of requirements from many users need to be analyzed. In this sense, automatic analysis of software requirements can bring benefits to software companies. For this purpose, we proposed an approach to automatically analyze software requirement specifications (SRSs) and extract the semantic information. In this approach, a machine learning and ontology based semantic role labeling (SRL) method was used. First of all, some common verbs were calculated from SRS documents in the E-commerce domain, and then semantic frames were designed for those verbs. Based on the frames, sentences from SRSs were selected and labeled manually, and the labeled sentences were used as training examples in the machine learning stage. Besides the training examples labeled with semantic roles, external ontology knowledge was used to relieve the data sparsity problem and obtain reliable results. Based on the SemCor and WordNet corpus, the senses of nouns and verbs were identified in a sequential manner through the K-nearest neighbor approach. Then the senses of the verbs were used to identify the frame types. After that, we trained the SRL labeling classifier with the maximum entropy method, in which we added some new features based on word sense, such as the hypernyms and hyponyms of the word senses in the ontology. Experimental results show that this new approach for automatic functional requirements analysis is effective.     

一种关联多本体的科技奖励检索方法

现有的科技奖励检索都是基于关键词的匹配,忽略了对用户查询的语义理解。针对传统信息检索存在的问题以及结合当前面向实际应用的基于本体的语义检索的特点,提出一种关联多本体的科技奖励检索方法。通过对科技奖励项目的领域本体构建,对概念相似度计算方法的改进及关联多本体概念,使扩展词语更能表达用户检索意图。实验结果表明,该方法对比传统查询检索和单独本体扩展检索获得了更好的召回率和准确率。     

双注意力引导的跨层优化交通场景语义分割

针对交通场景目标分割边缘不平滑以及小目标难以准确分割等问题，本文提出一种双注意力引导的跨层优化交通场景语义分割算法。首先，构建多分支特征提取编码网络，并利用串行非比例式空洞卷积实现空间上下文信息提取，进而改善小目标信息的丢失；其次，构建基于空间对齐的跨层特征融合解码网络，实现语义信息和细节信息的融合，增强不同尺度目标的表达能力；最后，提出通道和空间注意力机制，建模全局通道相关性和长距离位置相关性，增强网络对关键特征的学习能力。交通场景数据集Cityscapes和CamVid上的实验结果表明，所提特征提取编码网络、跨层特征融合解码网络以及注意力机制模块是有效的；所提语义分割算法获得了77.79%和78.66%的平均交并比，能够平滑目标分割边缘，尤其对细长条形目标具有鲁棒性。