基于GIS的决策支持及其应用

从决策支持的角度，通过简要回顾信息系统的发展过程，指出传统决策支持系统（DSS）的不足之处，由此引入了将当前迅猛发展的地理信息系统（GIS）应用于决策这一主题，并给出新的信息系统层次图，在此基础上进一步介绍几种非常流行的在GIS决策支持应用中使用的分析方法和工具，最后给出了GIS用于支持决策的实例。     

基于业务数据的高校图书馆决策支持系统设计

目前图书馆缺乏有效的工具来统计、分析业务数据,以便为图书馆的决策提供有效支持,为此设计一个基于业务数据的决策支持系统,并在此基础上构建数据仓储模型,实现图书馆决策的智能化.     

面向主题的高校智能决策支持系统研究

在分析高校决策支持现实状况的基础上,确立了人、财物、活动的三大主题,进而综合运用数据挖掘、商业智能等工具提出了适用于高校决策支持服务的系统架构,最后给出了系统的工作路线图和部分结果展示.     

贝叶斯分类器的应用

贝叶斯决策理论是统计模式识别中的一个基本方法。依据贝叶斯决策理论设计的分类器具有最优的性能,即所实现的分类错误率或风险在所有可能的分类器中是最小的,因此经常被用来衡量其他分类器设计方法的优劣。贝叶斯决策是一个很有效的分类工具,但它仍然存在着一定的错误率和风险,因此还需进一步的改善和完善。     

C4.5决策树法在网络课程知识点个性化设计中的应用

决策树是归纳学习和数据挖掘的重要方法,通常用来形成分类器和预测模型。对网络课程知识点个性化设计中的大量数据,运用数据挖掘算法中的决策树C4.5算法对所给数据进行处理,选取决策属性,构造决策树,提取分类规则,获取每一个知识点与不同类型的学生之间的关系。通过实验仿真发现,C4.5决策树算法取得了较为理想的分类预测效果。     

基于互补判断矩阵的模糊多属性群决策方法

针对属性值为梯形模糊数,属性权重未知,决策者对方案有梯形模糊数互补判断矩阵的模糊多属性群决策问题,提出一种群决策方法.给出了由综合属性值得到方案间相互比较的梯形模糊数互补判断矩阵的转化方法.借助梯形模糊数间距离的顶点法表示,建立模糊规划模型,最小化两类梯形模糊数互补判断矩阵间的偏差.求解模型,得到属性的精确权重.进而根据综合属性值对方案排序.最后给出一个数值例子验证所提方法的有效性.     

Capturing and assessing evidence of student teacher inquiry: A case study

In this paper, we present five principles underlying teacher inquiry and report a case study during which student teachers in a US teacher education programme used evidence-based decision support, an inquiry method designed to help teachers analyse and adapt their own teaching through the use of a video analysis tool. This case study examined interplay among tools designed to guide teaching practice, student teachers’ own self-guided inquiry, and feedback from cooperating teachers as they made instructional decisions. Preservice teachers initially accepted the guidance provided by teaching analysis tools, but abandoned the tools in favour of informal self-assessments and feedback from cooperating teachers when they assumed teaching responsibilities in their own classrooms. We discuss the role of external support and video evidence in guiding preservice teacher inquiry.     

Selecting software tools for IS/IT curricula

The evaluation and selection of software tools for use in an IS or IT curriculum is difficult not only because actual industry software tools are often used but also because there is no formal approach to guide the process. How does one choose between SQL Server and MySQL, or Dreamweaver and Expression Studio? IS and IT educators must periodically go through the process of assessing the most suitable tools for their courses. Given how common such decisions are and how frequently they must be made, it is surprising to find that there is a lack of literature that deals with comparative studies of software tools for higher education. This paper proposes a set of criteria for the selection of software tools for IS and IT programs, explains how multi-criteria decision analysis can be used to weight those criteria, and details an approach for the application of those weighted criteria. The proposed approach is structured and replicable, and allows for a more thorough evaluation of the available options and a more easily supportable selection.     

Measurement Specialists: Testing the Faith–A Reply to Mehrens

The centrality of the validity issue in any discussion of test use is the focus of George's reply to Bill. While acknowledging that one can distinguish between a decision and the tools used to assist in making the decision, he argues that in most high-stakes tests the decision is inherently linked to the test. Hence, the validity of the decision is directly related to the validity of the test.     

Risk attitude assessment using roulette and a supply chain simulation

Many topics that are taught in supply chain management (SCM) programs include elements of variability and risk. With quantitative tools in particular, it is assumed that the decision makers using the tools are risk neutral, which is typically not the case in real-life decision situations. The topic of an individual's attitude toward risk has received limited attention in SCM literature and is not a high-profile topic in SCM textbooks. The teaching exercise presented in this paper uses two games—roulette and a SCM simulation—to provide students with an assessment of their own attitude toward risk, and in doing so, facilitates a classroom discussion of risk and risk attitude. These games were developed by the authors and can be played either within Excel or a web application.     

一种改进的属性约简方法

属性约简是粗糙集理论的核心问题之一,寻找最优约简现已证明是NP—hard问题。本文在差别矩阵的基础上,采用贪心算法的策略,提出了一种启发式算法来寻求最优约简。该算法不仅能在有核的决策表中通过核进行属性扩充求得最优约简,而且在无核的决策袁中也能根据属性在差别矩阵中出现的频率求得最优约简。经实例验证该算法是有效的。     

基于SWOT分析的东南亚语言学科毕业生出国就业前景研究——以广西民族大学为例

通过SWOT工具分析了目前东南亚语言学科毕业生出国就业具有的内部优势和劣势、面临繁荣外部机遇与挑战。根据分析结果,绘制SWOT矩阵图,制定各种相关战略,帮助毕业生整合有效资源,转变就业观念,寻求就业决策的最优化。     

新约简模型中两个重要性质

Mean decision power （MDP） is an important criterion of a new reduction model, and relative decision power （RDP） and amount of rules （AR） are key parameters of MDP. This paper presents two important properties： relationship between RDP and AR, and relationship between MDP rule set of parent decision table and MDP rule set of child decision table. These properties can help better understanding of the new reduction model and are useful tools by which one can rapidly derive an MDP rule set.     

基于服务质量的多属性群决策服务选择算法

为解决基于服务质量的服务选择,提出了一个多属性群决策算法.该算法通过使用Borda社会选择函数集成了多种具有不同原理用于计算不断变化的服务质量方法的结果,从而增强了服务选择的可靠性.实验结果表明该方法具有较好的可扩展性,从而能适用于大规模的分布式服务计算环境;同时相比于随机和轮转服务选择策略,所提出的基于群决策的服务选择策略具有更优的性能,它能够有效地优化服务选择.该算法不仅仅是增加了一种用于获取不断变化的web服务质量的方法,而是建立一个具有多层知识的综合信息体,从而保障了服务选择的正确性.     

基于粗糙集的属性约简算法

属性约简是粗糙集理论研究的关键问题之一,现已证明寻找一个决策表的最优约简是N P-hard问题。本文首先介绍可辨识矩阵属性约简的基本算法并对求核算法进行了改进。在此基础上给出决策表中属性重要性的两种度量,并以此为启发式信息,提出了一种属性约简的启发式算法。最后,实验结果表明,该算法在大多数情况下能有效地获得决策表的最优约简。     

数控机床计算机控制系统动态特性的数字仿真

对数控机床计算机控制系统进行建模,用SIMU语言进行仿真研究,分析系统参数的合理性,得到整个系统的最佳动态特性参数,为数控机床改进、设计提供依据．     

数据仓库和OLAP在配变通信系统中的应用

对数据仓库和OLAP在配变通信系统中的应用进行了研究,对系统中数据仓库和联机分析处理的开发过程进行了分析和探讨,论述了星型模型的设计,以达到使系统不仅是一个管理工具还是一个决策支持工具的目的。     

Collaboration in ePortfolios with Web 2.0 tools in initial teacher training / La colaboración en eportafolios con herramientas de la Web 2.0 en la formación docente inicial

Collaboration is a specific process in the construction of portfolios. In the case of electronic portfolios, Web 2.0 tools make collaboration a key process. The process known as networking refers to the online connection and collaboration with others, which the social web contributes to the literature of electronic portfolios. This research describes and analyses the pilot implementation of an ePortfolio in the Early Childhood degree programme at the Ibiza Campus of the University of the Balearic Islands. The paper presents a qualitative research study which aims to understand collaborative students’ activity from multiple points of view. In view of this, the data collection techniques used are diverse: analysis of work done by students, Likert scale and group discussion. Atlas.ti software has been used to obtain a qualitative analysis of the students’ discourse. In general, the results show positive attitudes, with a gradual increase noticed in the students’ use of the tools and, finally, with collaboration being expressed in peer work, especially in relation to technology. The analysis carried out enables us to observe tensions between open tools, collaboration and assessment. The research has important implications for practice as it allows us to deduce how to improve the collaboration process in ePortfolios thanks to the introduction of Web 2.0 tools.     

Evaluating the performance of calculus classes using operational research tools

This paper compares the efficiency of calculus classes. Two kinds of classes are evaluated: the traditional ones and others that use computational methods in teaching. This experiment was performed at Fluminense Federal University, Niterói, Brazil, from 1998 to 2000. The emphasis of this paper is on the quantitative evaluation using two operational research tools: multicriteria decision aid methods (mainly using the MACBETH approach) and data envelopment analysis. The evaluating variables are the level at which students enter the university and the performance of the students after studying calculus.     

傅立叶级数+机器学习法双联预测小行星分类

利用傅立叶级数对散点小行星光变数据进行曲线拟合,以获得小行星表示参数,选取傅立叶级数为6的拟合方式对单体或双体小行星光变曲线进行分类,并用机器学习算法中的SVM和决策树建立预测模型。检验结果表明,SVM模型对单体和双体小行星的预测正确率达到95%,相较于决策树正确率提高了10%,为从小行星实际观测数据直接推测双体小行星潜在相关应用提供了参考。