Fractal measure theory for knowledge representation

A quantitative analysis of knowledge representation is attempted.Based on the fractal theory and information measuring principles and methods,a series of quantitative concepts for knowledge representation,e.g.information capacity,entropy,information processing channel (IPC),are proposed.From the viewpoint of information processing,the principles of maximum entropy and increasing entropy are discussed,and the types,characteristics and structures of information processing channel are also discussed.Finally,the quantitative theoretical analyses of several typical knowledge representation methods are given.     

BC-iDistance: an optimized high-dimensional index for KNN processing

To facilitate high-dimensional KNN queries, based on techniques of approximate vector presentation and one-dimensional transformation, an optimal index is proposed, namely Bit-Code based iDistance （ BC-iDistance）. To overcome the defect of much information loss for iDistance in one-dimensional transformation, the BC-iDistance adopts a novel representation of compressing a d-dimensional vector into a two-dimensional vector, and employs the concepts of bit code and one-dimensional distance to reflect the location and similarity of the data point relative to the corresponding reference point respectively. By employing the classical B ＋ tree, this representation realizes a two-level pruning process and facilitates the use of a single index structure to further speed up the processing. Experimental evaluations using synthetic data and real data demonstrate that the BC- iDistance outperforms the iDistance and sequential scan for KNN search in high-dimensional spaces.     

The Technology of Individual and Digital Pattern-making in the MTM System

The Garment MTM enables to realize the individual and quick manufacture to meet the development of the per- fectly popular and various garment industry,whose core technology is to produce the individual and digital pattern,that is,get- ting the individual pattern-making according to the body information.It is relatively correct and wise to get the individual pattern by modifying the basic pattern and the experts in this field do their researches on this way.One vital and unsolved problem is that how to find the very pattern which is the closest to the customers in the pattern-database which determines the validity of the final pattern.We conclude a technology methods from body information to digital pattern,design the rule of seeking yards and pattern- modification,build the B-P system (the system of the quick manufacture technology of body information-digital pattern),set up the style database,yarn database,pattern database,the rule databases of seeking yards and pattern-modification.It is confirmed that the B-P system is available,the setting-up of all the databases in the system is reasonable and the final pattern is correct.     

Enterprise modeling method of calculation-independent model level

Aimed at deficiencies in the development and implementation of Enterprise Service Architecture (ESA) software,an ESA software developing mode based on Model Driven Architecture (MDA) is put forward.This mode includes a calculation-independent model (CIM),a platform-independent model (PIM),a platform-specific model (PSM) and a code level.Based on this mode,the modeling architecture of CIM level is presented.CIM here includes a global model,a process model,an information model and an organization model.The modeling elements of global model,process meta-model,information meta-model and organization meta-model are defined in detail and the relationship between them is described.The reflecting relationship between these models is established as well.     

User’s Relevance of PIR System Based on Cloud Models

A new method to evaluate fuzzily user＇s relevance on the basis of cloud models has been proposed. All factors of personalized information retrieval system are taken into account in this method. So using this method for personalized information retrieval （PIR） system can efficiently judge multi-value relevance, such as quite relevant, comparatively relevant, commonly relevant, basically relevant and completely non-relevant, and realize a kind of transform of qualitative concepts and quantity and improve accuracy of relevance judgements in PIR system. Experimental data showed that the method is practical and valid. Evaluation results are more accurate and approach to the fact better.     

Hybrid optimization model of product concepts

Deficiencies of applying the simple genetic algorithm to generate concepts were specified. Based on analyzing conceptual design and the morphological matrix of an excavator, the hybrid optimization model of generating its concepts was proposed, viz. an improved adaptive genetic algorithm was applied to explore the excavator concepts in the searching space of conceptual design, and a neural network was used to evaluate the fitness of the population. The optimization of generating concepts was finished through the ＂evolution - evaluation＂ iteration. The results show that by using the hybrid optimization model, not only the fitness evaluation and constraint conditions are well processed, but also the search precision and convergence speed of the optimization process are greatly improved. An example is presented to demonstrate the advantages of the orooosed method and associated algorithms.     

A Semantic Model Faced on the Uniform Product Knowledge Representation

In order to realize the uniform knowledge representation including STEP and SGML,aimed at the defects of cur- rent methods,a new semantic model that is named XOEM OWL is put forward.And then the correspondent mapping between STEP Schema Graph and OWL Schema Graph are build as Cos(sc,oc),so we can get the semantic pattern matching degree for the semantic representation on the product information.At last the example is presented.     

Generalized multi-layered granulations and approximations based on neighborhood systems under incomplete information systems

A generalized multi-layered granulation structure used by neighborhood systems is proposed. With granulated views, the concepts of approximations under incomplete information systems are studied, which are represented by covering of the universe. With respect to different levels of granulations, a pair of lower and upper approximations is defined and an approximation structure is investigated, which lead to a more general approximation structure. The generalized multi-layered granulation structure provides a basis of the proposed framework of granular computing. Using this framework, the interesting and useful results about information granulation and approximation reasoning can be obtained. This paper presents some useful explorations about the incomplete information systems from information views.     

Bayesian texture segmentation based on wavelet domain hidden markov tree and the SMAP rule

According to the sequential maximum a posteriori probability (SMAP) rule, this paper proposes a novel multi-scale Bayesian texture segmentation algorithm based on the wavelet domain Hidden Markov Tree (HMT) model. In the proposed scheme, interscale label transition probability is directly defined and resoled by an EM algorithm. In order to smooth out the variations in the homogeneous regions, intrascale context information is considered. A Gaussian mixture model (GMM) in the redundant wavelet domain is also exploited to formulate the pixel-level statistical features of texture pattern so as to avoid the influence of the variance of pixel brightness. The performance of the proposed method is compared with the state-of-the-art HMTSeg method and evaluated by the experiment results.     

Linear feature detection based on ridgelet

Linear feature detection is a foremost problem in image processing and pattern recognition. An image is composed of a number of points, lines and planes. Effective detections of these com-ponents play an important role in simplifying feature model in pattern recognition and pattern match. Wavelets[1—3] perform very well for objects with point singularities in one and two dimen-sions and are shown to be optimal basis for representing discontinuous functions in one dimension and functions with …     

本体论在网络入侵检测技术中的应用

为解决当前入侵检测系统中的检测器协作问题和检测器内知识表达问题，将知识工程中的本体论技术应用于分布式入侵检测系统中，使用入侵检测本体在系统内形成了统一的全局概念视图。按照入侵检测体系的基本框架构建了入侵检测本体，该本体模型对入侵检测三要素(信息资产、攻击者、攻击行为)进行了分层的概念和属性描述。检测器按照作用域继承本体模型内的不同概念类形成自身的领域知识，同时在统一的知识论框架的引导下完成协作过程。最后，通过对基于本体模型的检测系统的实例证实了该方法的可行性。     

模式特征的几何代数多向量表示方法

模式表示是模式识别的一个基本问题。传统统计模式识别理论中模式特征一般表示为一个数值向量,并被视作”维欧式空间中的一个点。这种表示方法只利用了一阶特征,容易丢失模式特征间的关联信息和高阶结构。本文首先阐述了几何代数的公理化定义和一些基本概念,然后将传统的模式特征向量表示推广为几何代数空间的多向量表示,接着讨论了该表示的两种特例,最后阐述了基于该多向量表示进行特征提取和分类的一般思想和需要进一步研究的问题。     

本体模型内概念映射翻译之研究

在多应用主体的系统中，由于各应用主体工作领域和知识表达、处理方式的不同在协作过程中易造成语义歧义等问题。本体论为应用主体间的知识级协作提供了一条可靠的途径。探讨了与本体相关的应用主体和本体推理机的结构组成以及协作过程中的术语翻译转换流程。通过对本体领域相关性、概念术语相关性等重要概念的定义描述。给出了本体推理机在不同领域本体之间进行概念术语的翻译转换方法。     

可视化搜索引擎在信息素养培养中的应用

本文论述了信息素养与可视化搜索引擎的概念、信息素养培养的现状、可视化搜索引擎信息素养培养中的应用优势,通过分析与比较,展示了可视化搜索引擎的优越性及美好的发展前途。     

基于PROV的ETL起源信息统一表达机制

在异构环境下,目前数据起源研究主要基于OPM模型来表示数据在ETL中的来源过程,存在着起源概念不统一、词汇使用混乱以及无法提供标准化访问等问题。基于W3C的PROV模型,提出了ETL起源信息的统一表达机制。该机制首先对ETL过程的起源概念及其关系进行了统一描述。然后,针对ETL过程特殊的语义表达需求,建立了多粒度的ETL起源词汇表。最终,建立在RDF之上的标准化查询机制提高了起源信息的可访问性。     

二维形状的一种抽象模型

本文提出二维形状抽象描述的概念,以适应CAD过程中设计对象从抽象到具体的发展,不同性质的约束条件的逐步满足。一个二维多边形的抽象模型被提出。并以建筑CAD中的楼层平面规划为例,说明一个多边形网络如何在设计中从抽象到具体的发展,并逐步满足各种约束。     

软件体系结构在电子商务推荐系统中的实现

以某网上商城作为研究对象,运用软件体系结构中Web表现模式中的MVC模式实现一款智能电子商务推荐系统.顾客在购买商品时,系统可以主动推荐其可能感兴趣的商品信息.     

面向对象的机械设计专家系统工具的研究

介绍了面向对象的机械设计专家系统开发工具ＯＭＥＳＴ的系统结构,描述了面向对象的知识表示方法,着重阐述了适用于机械设计的规则—框架—过程知识表示法,以及如何将多种知识表示融合于知识对象中,讨论了面向对象的推理机制等     

基于Rete算法的几何自动推理系统

智能几何教学软件都使用基于前推法的几何推理系统作为推理引擎。推理引擎的主要工作是做推理规则和几何信息的匹配计算。推理过程中产生的大量冗余匹配严重地影响了推理效率。为了提高推理引擎的推理效率, 作者首次将Rete模式匹配算法整合到推理引擎中, 构造了一种高效的几何自动推理引擎, 称为几何自动推理网。几何自动推理网通过消除推理过程中的冗余匹配达到了提高系统推理效率的目的。使用Lisp语言编写了几何自动推理网程序iGeo, 并做了50多个非平凡几何定理的推理实验, 实验结果表明iGeo具有更高的推理效率。