并行概念学习的探讨

为了解决机器学习的低效问题，将并行技术中引入概念学习。在深入探讨概念学习本质的基础上，分析了并行概念学习的可能性及其前景；研究了建立一个并行概念学习系统所需考虑的几个问题；探索性地给出了一种特殊的实例——概念描述语言；在此语言基础上提出了一种并行概念学习系统的实现方案，并详细讨论了系统的体系结构和学习算法。     

Notes数据库与关系型数据库之间的整合在办公自动化系统中的应用

本文在基于LotusDomino／Notes平台开发的办公自动化系统基础之上,介绍了非结构化数据的特点及非结构化数据库在办公自动化系统中的作用,对非结构化数据库与传统关系型数据库的存储机制和访问机制进行了详细的分析探讨,提出了Notes数据库与关系型数据库整合的技术方式,其目．的是实现企业信息系统中Notes数据库与关系型数据库信息交互。以利用Notes平台提供的良好性能共享信息数据。     

多层数据实现电子地图的算法

本文阐述了一种实现电子地图功能的数据结构及其算法，提出信息按层存放，电子地图由基本住图文件叠加可编辑的结构化信息，所有的信息存放在关系型数据库中，利用该算法开发的软件具有查找、删除、增加、煽辑等多项功能。     

智能CAD系统中的关系型图形数据结构

本文阐述了关系型图形数据结构在智能CAD系统中的应用 ,通过实例说明了如何建立关系型图形数据结构,并指出了其对提高图形设计效率、方便图形修改和支持参数化设计的重要作用.     

智能CAD系统中的关系型图形数据结构

本文阐述了关系型图形数据结构在智能CAD系统中的应用,通过实例说明了如何建立关系型图形数据结构,并指出了其对提高图形设计效率、方便图形修改和支持参数化设计的重要作用。     

一种概念学习分类方法的探讨

为了解决数据分类效率问题,引入了并行概念学习的方法,针对二叉树分类法的“累积误差”问题,提出了一种并行概念的描述语言.在此基础上,设计了基于二叉树的多类SVM并行分类系统.通过对系统性能的分析表明,在样本空间庞大、样本涉及属性广的数据分类时,该系统具有很好的分类效果.     

人事管理系统中的数据库设计

基于Windows平台的关系型数据库管理系统,以Microsoft SQL Server2000为软件平台,采用PowerBuilder为前端开发工具,完成了针对高校人事管理系统的数据库设计。利用结构化的分析方法进行系统分析,并结合数据库相关理论以高校的实际情况,进行系统数据库的概念结构和逻辑结构设计,最后讨论数据库的连接方法。     

基于关系数据库的关联规则的形式化开采

从研究关联规则概念入手，探讨了项目、项目集合和交易数据库等概念在关系数据库中的具体含义，提出了一种全面的、易于理解和使用的关联规则形式化描述方法，并给出了开采关系型数据库中关联规则的一般过程，这为关系型数据库中关联规则开采系统的设计奠定了基础，对系统设计者有一定的指导意义。     

基于RDFS的数据结构描述

资源描述框架(RDF)/资源描述框架模式(RDFS)是语义互联网架构中的技术。尽管XML/RDF数据库正迅速地发展,关系型数据库仍是目前企业级应用的首选。针对在应用中混合使用关系型数据库管理系统和RDF/RDFS的问题,该文提出了关系数据库表结构的RDFS定义方法以及在此基础上的数据获取方式,目的是在关系数据库与应用系统之间建立基于RDFS的抽象数据结构描述,实现数据结构的可配置性以及RDF数据的可访问性。     

一种基于MongoDB的企业内容管理系统实现

随着IT技术在企业实际业务中扮演着越来越重要的角色,企业的IT系统中所产生的数据量也越来越大,其中非结构化的数据又占有相当大的比例,传统的利用关系型数据库对非结构化数据的管理在面对海量数据时面临着性能不足的瓶颈.基于此,提出了使用非关系型数据库MongoDB作为后端数据库的内容管理解决方案,利用MongoDB的文档数据库且易于横向扩展的特性来解决基于关系型数据库的内容管理系统的性能瓶颈.     

基于搜索空间划分的并行概念生成算法

概念格作为形式概念分析理论中的核心数据结构,在机器学习、数据挖掘和知识发现、信息检索等领域得到了广泛的应用。概念格的构造在其应用过程中是一个主要问题。本文提出了一种基于搜索空间划分的并行概念生成算法,它对整个闭包搜索空间进行划分,并引入一种有效的测试方法,只搜索那些能生成正规闭包的子搜索空间,从而有效提高搜索效率;同时,在计算闭包过程中保存一些必要的中间结果,用来提高闭包运算速度;由于所有子搜索空间相对独立,因此很容易得到一个井行的概念生成算法。     

Using Hellinger distance in a nearest neighbour classifier for relational databases

Nearest neighbour algorithms classify a previously unseen input case by finding similar cases to make predictions about the unknown features of the input case. The usefulness of the nearest neighbour algorithms has been demonstrated in many real-world domains. Unfortunately, most of the similarity measures discussed in the current nearest neighbour learning literature handle only limited data types, thus limiting their applicability to relational database applications.In this paper, we propose an enhanced nearest neighbour learning algorithm that is applicable to relational databases. The proposed method allows one to define similarity on a wide spectrum of attribute types. It automatically assigns to each attribute a weight of its importance with respect to the target attribute. The method has been implemented as a computer program and its effectiveness has been tested on four publicly available machine learning databases. Its performance is compared to another well-known machine learning method, C4.5. Our experimentation with the system demonstrates that the classification accuracy of the proposed system was superior to that of C4.5 in most cases.     

基于数据库的层次概念知识体系的一种获取方法

以关系数据理论为基础,提出一种基于数据库的大量概念学习和概念体系构建的新方法,给出了基于数据库的概念定义和一种简化的获取方法。实例计算表明了该方法的有效性     

基于搜索空间划分的概念生成算法

概念格作为形式概念分析理论中的核心数据结构,在机器学习、数据挖掘和知识发现、信息检索等领域得到了广泛的应用.概念格的构造在其应用过程中是一个主要问题.提出了一种基于搜索空间划分的概念生成算法SSPCG(search space partition based concepts generation),它将属性集合的幂集看作初始闭包搜索空间,迭代地将每个搜索空间划分为一些子搜索空间,并引入了子搜索空间的有效性判断,只搜索那些能生成正规闭包的子搜索空间,有效地提高了搜索效率;同时,在计算闭包过程中保存一些必要的中间结果,用来提高闭包运算速度.由于所有子搜索空间是独立的,所以该算法可以很容易地扩展为并行算法.在随机生成的数据集和真实数据集上进行的实验测试表明,本算法的时间性能要优于Ganter提出的NextClosure算法.     

一种基于FCA的面向关系数据库的本体学习方法

从已有的数据模型中进行语义提取,经过一定的规则映射生成本体的过程称为本体学习。关系数据库模型是当前数据的存取与组织的主要模型,从中学习得到本体,一直是本体工程领域研究的热点之一。利用手工定义的E-R模型到本体的映射规则来完成本体的构建,是国内外大部分学者采用的方法。但这样获得的本体概念层次关系主观依赖性强,不利于本体的实际应用。为了能更加客观地获取数据之间的概念层次关系与语义信息,提出了一种基于FCA(形式概念分析)从关系数据库进行本体学习的方法。该方法既保持了关系数据表中原有的数据语义关系,又发挥了FCA自动提取语义信息的特点,提高了最终本体生成的质量,有利于在具体的领域应用中使用本体。最后结合材料服役安全数据库的数据信息,演示了运用所提出的方法学习得到领域本体的过程。     

基于FP-T的多层关联规则并发挖掘

现有的数据挖掘方法大致有两类：有候选项集和无候选项集，有候选项集的挖掘以Apriori算法为代表，其特点是产生大量的候选项集，重复多次扫描数据库，挖掘效率低，不适合大型数据库的挖掘。无候选项集的挖掘以FP-T方法为代表，但它不能同时挖掘多概念层的关联规则，对具有超大项ID的大型数据库，无法生成“树”结构，使用也受到限制。该文将FP-T原理引入多层关联规则的并发挖掘，通过构建一个特殊节点链的指针表，可实现超大规模数据库的并发、多层挖掘。对实现物流系统信息自动化及其它数据挖掘应用领域都具有极其重要的指导意义。     

Modelling Parallel Oracle for Performance Prediction

The problem of predicting the performance of a parallel relational DBMS for a set of queries applied to a particular data set on a shared nothing parallel architecture without transferring the application to a parallel system is a challenging one. An analytical approach has been developed to assist with this task and has been applied to the ICL GoldRush machine, a parallel machine with a shared-nothing architecture. This paper describes how the Oracle Parallel Server and the Parallel Query Option are modelled by the method and compares the predictions of the model against actual measurements obtained.     

Metaform: updatable form screens and their application to the use of office metaphors in query language instruction

A database interface language and system, called Metaform, which automatically generates multi-relational form screen interfaces for use by non-computer professionals has been developed. A form screen is a subset of the relational database, with a particular relation or combination of relations being represented. Through form screens, users can simultaneously query and update several relations in the database without having to know about its underlying structure. An overview of the Metaform system is presented and several examples of the use of the Metaform query language and update operators are described.

A series of 'usability' studies were conducted on a prototype of the Metaform system to examine the claims that the form concept aids computer-naive users in building complex database queries. These studies adopted the form screen concept to present six office paper work analogies to users to help them to understand the database retrieval concepts. The analogies of a file cabinet, a file folder, a stack of forms, a single form, a table of information on a form and a field of information were used in a two-staged training module.

At the end of each training sequence, users answered questions with the prototype and with paper and pencil which tapped their understanding of the database retrievals they were learning to perform. The results from these questionnaires were mixed. Users performed successful relational queries for simple retrievals and for those using existential quantifiers. They had difficulty with queries involving multiple steps and intermediate stages. Although users understood and used the analogies, they ran into difficulties with the ambiguities in the English statements of the queries, thus suggesting a need for another level of metaphors and/or problem representation tools not associated with the machine but with the user's comprehension of database retrieval problems.     

Metaform: updatable form screens and their application to the use of office metaphors in query language instruction

Abstract

A database interface language and system, called Metaform, which automatically generates multi-relational form screen interfaces for use by non-computer professionals has been developed. A form screen is a subset of the relational database, with a particular relation or combination of relations being represented. Through form screens, users can simultaneously query and update several relations in the database without having to know about its underlying structure. An overview of the Metaform system is presented and several examples of the use of the Metaform query language and update operators are described.

A series of ‘usability’ studies were conducted on a prototype of the Metaform system to examine the claims that the form concept aids computer-naive users in building complex database queries. These studies adopted the form screen concept to present six office paper work analogies to users to help them to understand the database retrieval concepts. The analogies of a file cabinet, a file folder, a stack of forms, a single form, a table of information on a form and a field of information were used in a two-staged training module.

At the end of each training sequence, users answered questions with the prototype and with paper and pencil which tapped their understanding of the database retrievals they were learning to perform. The results from these questionnaires were mixed. Users performed successful relational queries for simple retrievals and for those using existential quantifiers. They had difficulty with queries involving multiple steps and intermediate stages. Although users understood and used the analogies, they ran into difficulties with the ambiguities in the English statements of the queries, thus suggesting a need for another level of metaphors and/or problem representation tools not associated with the machine but with the user's comprehension of database retrieval problems.     

A relational database machine with large semiconductor disk and hardware relational algebra processor

This paper describes the basic concepts, design and implementation decisions, standpoints and significance of the database machine Delta in the scope of Japan’s Fifth Generation Computer Project. Delta is planned to be operational in 1985 for researchers’ use as a backend database machine for logic programming software development. Delta is basically a relational database machine system. It combines hardware facilities for efficient relational database operations, which are typically represented by relational algebra, and software which deals with hardware control and actual database management requirements. Notable features include attribute-based internal schema in accordance with the characteristics found in the relation access from logic programming environment. This is also useful for the hardware relational algebra manipulation algorithm based on merge-sorting of attributes by hardware and a large capacity Semiconductor Disk for fast access to databases. Various implementation decisions of database management requirements are made in this novel system configuration, which will be meaningful to give an example for constructing a hardware and software combination of a relational database machine. Delta is in the stage between detailed design and implementation.