基于粗糙集和神经网络的供应商决策选择研究

本文提出了一种基于粗集理论与神经网络相结合的供应商优化选择新方法.首先利用粗集理论对供应商原始指标决策表进行一致性属性约简,然后借助神经网络强大的学习能力对决策表进行自学习,完成属性的不一致约简,最后再由粗糙集对其进行值约简,得到最终的供应商优化选择决策规则,并给出了相应的算法.最后通过实例说明该方法的有效可行性.     

多标准决策表中发现概率规则的变精度粗糙集方法

用优势关系代替不可分辨关系,本文提出了一种可以从多标准决策表中获取概率规则的扩展变精度粗糙集模型,该模型能够处理多标准决策表中可能的不一相容性,获取由偏好对象组成的概率决策规则集,并导出事例决策系统的偏好模型。研究结果表明:基于优势关系,从多标准决策表中获取的最小概率规则集,使用的条件数量较少,且导出规则的数量较少、较强。     

基于粗糙集理论的格序选择方法

在粗糙集理论中,为了识别由偏好属性导致的不相容性,已提出基于优势关系的粗糙集理论,然而这一理论在实际应用中往往导致多个决策规则的出现,如何辨别由偏好多属性决策表中获取的多个决策规则从而选取最优方案,尚未见到有关的研究文献。本文针对这一问题,利用格序理论,给出属性约简的贴近度,比较不同约简所得决策规则贴近于原知识库的程度,从而得到一种选择最优决策方案的方法。     

杂合灰色聚类与变精度粗糙模糊集的概率决策方法及应用

在基于粗糙集理论或扩展粗糙集理论的不确定性决策中,仅由数据集无法获得决策表,同时由于决策属性呈现模糊性,以致于难以获取概率决策规则。针对这一类决策问题,基于灰色定权聚类与优势粗糙集变精度方法,本文构建了一种的杂合决策方法。该方法首先利用中心点三角白化权函数的灰色定权聚类方法将包含不同量纲指标的知识表示系统生成多属性决策表,而后运用变精度粗糙模糊集进行决策分析,生成概率决策规则,最后以区域关键技术的选择为例,表明该模型的可行性与实用性。     

基于粗集理论的一种归纳学习算法

归纳学习是粗集理论的主要应用领域.本文基于粗糙集理论,针对相容决策表,提出一种新的归纳学习算法.与以往的方法相比,这种算法比较简单,能够全面地获取规则且没有冗余.此外,还讨论了不相容决策表的学习,给出了具有较高可信度的规则挖掘过程.     

一种基于VPRS的IT项目投标风险要素评价方法

变精度粗糙集模型允许有一定程度的错误分类率存在,利用这种特性可以有效消除投标风险信息系统中的数据噪声,提高风险要素评价的精度.结合层次分析法中判断矩阵方法提出一种基于变精度粗糙集的评价IT项目投标风险要素重要度的算法.首先,由专家对各要素风险程度打分,建立决策表;其次,利用变精度粗糙集属性依赖度求得属性间的相对重要性来构造判断矩阵;最后,确定各要素的综合重要度,并排序.     

粗糙集在教学质量评价中的应用

粗糙集不依赖于数据集之外的附加信息,是处理含有噪声、不精确、不完整数据的有力工具,是一种新的数据挖掘技术.首先,本文介绍了粗糙集理论和决策表约简算法,然后采用粗糙集数据挖掘技术对多指标教学质量进行评价,挖掘出数据背后隐含的规则.     

一种基于粗集的实例分解归纳学习方法

粗集是知识约简的一种有效工具,但决策表的最小约简已被证明是NP难题.对于复杂的决策表,由于必要的条件属性数目可能较多,直接用粗分析比较困难.为此本文基于粗集理论,给出了一种决策表的分解方法.实例表明决策表的分解可以减少知识约简的难度.这对于大型的决策表,有潜在的应用价值.     

基于蚁群算法的成本敏感线性集成多分类器的客户流失研究

针对电信客户流失预测问题的复杂性,融合自组织神经网络良好的连续属性值离散化优势、粗糙集理论出色的属性约简功能和蚁群优化算法全局的随机搜索特点,在模型集成技术和成本敏感学习理论的基础上,提出了一种新的基于蚁群算法的成本敏感线性集成多分类器的电信客户流失预测模型。构建该集成模型可分为4个阶段:(1)连续属性值的离散处理:利用自组织神经网络对连续属性值进行非监督离散化处理;(2)原始属性集的约简处理:使用粗糙集理论按属性重要性原则对离散属性进行约简;(3)子分类器的建立:分别使用NaiveBayes、Logistic回归、多层感知器和决策树等4种差异性很大的分类技术在约简属性集上建立4个对应的客户流失预测子分类器;(4)子分类器的集成:基于成本敏感学习理论,构建了4种不同的线性集成模型,采用蚁群优化算法求解集成模型的最优线性组合权重系数。将该模型应用于某电信客户流失预测,其实验结果表明该集成方法是可行且有效的。     

基于粗糙集理论的财务危机预警研究

本文针对财务危机预警中财务比率数量较多时计算量较大的问题,提出了基于粗糙集理论的财务危机预警模型,将财务比率分别作为条件属性,将企业所处的状态作为决策属性,从而构成一个财务预警决策系统.通过对决策表的属性约简和规则约简,得到决策系统的最小决策规则,从而实现财务危机预警.测试结果表明,该方法的预警结果正确、有效.     

基于头脑风暴原则的主观证据融合决策方法

为了解决证据源是主观的或者客观证据源因无法获取有效信息而不得不寻求主观经验、知识、直觉帮助的决策问题,采用基于互补判断的相对估计方式构建了能够提取决策专家主观推断信息的柔性知识矩阵,给出了由柔性知识矩阵向BPA函数转化的主观证据提取模型,在此基础上将头脑风暴方法中的延迟评判、独立思考、以量求质、结合改善四项原则引入到主观证据的提取过程之中,构建了可以从识别框架中辨识出可能为最优方案的融合决策方法,最后分别应用数值对比分析和案例模拟分析验证了提出方法的科学有效性和应用可行性。     

Knowledge sharing for innovation performance improvement in micro/SMEs: an insight from the creative sector

Abstract

As economies become more reliant on innovative, knowledge-intensive firms, understanding the interaction between knowledge and improving innovation performance is increasingly important. Although most UK businesses are micro, small or medium-sized enterprises (micro/SMEs), knowledge management research has tended to focus on large companies Knowledge sharing can be critical for innovation performance, especially for smaller players with limited resources. Our study presents an insight from micro/SMEs operating in the highly knowledge-intensive and innovative games/entertainment software development sector. Using a mixed method approach, we investigate knowledge sharing and its contribution to firm innovation performance improvements. Our findings suggest that micro/SMEs are at the forefront of the creative sector precisely because of their smaller size. Our study reveals evidence of knowledge donation but limited evidence of knowledge collection in the knowledge sharing process. We develop a model highlighting the importance of industry context, individual knowledge and organizational size in knowledge sharing for innovation performance.     

Evidence-based human resource management practices in three EU developing member states: Can managers tell truth from fallacy?

Good management practice is assumed to be the product of a good knowledge base and its application, not least in Human Resource Management (HRM). The aim of this study is to assess the extent to which managers adhere to practices of HRM that are more likely to be upheld by research evidence as opposed to beliefs for which research evidence is highly lacking. In addition, it evaluates practitioners’ explanations about adopting HR practices. This study was conducted in three European Union (EU) developing countries (Poland, Croatia and Malta). A mixed-methods approach is adopted, utilising a web-based questionnaire targeting a purposive sample of 300 practitioners occupying managerial positions and directly involved in people management (Study 1), followed by 20 in-depth interviews with similar participants (Study 2). Our study reveals that managers are always fully cognisant of the main body of research evidence related to specific HR practices. Practitioners are more likely to access required knowledge for applications through popular sources rather than more reliable ones due to time constraints, inaccessibility and inability to evaluate evidence. While they appreciate the role of theory, the leap from theory to application is not easy or straightforward, commenting that large gaps between research and practice prevail. These overall trends may imply that practitioners base their decisions on personal experience rather than on evidence-based knowledge or expertise acquired through evidence-grounded applications. Results suggest that future research should focus on closing the gap by evaluating how academics impart the knowledge and how practitioners apply that knowledge.     

Analytical Paradigms: The Epistemological Distances between Scientists, Policy Makers, and the Public

The effective use of evidence and its resultant knowledge is increasingly recognized as critical in risk analysis. This, in turn, has led to a growing concern over issues of epistemology in risk communication, and, in particular, interest in how knowledge is constructed and employed by the key players in risk--scientists, policy makers, and the public. This article uses a critical theoretical approach to explore how evidence is recognized and validated, and how limits are placed on knowledge by scientists, policy makers, and the public. It brings together developments in the sociology of science, policy and policy development, public understandings of science, and risk communication and analysis to explicate the differing forms of rationality employed by each group. The work concludes that each group employs different, although equally legitimate, forms of rationality when evaluating evidence and generating knowledge around risky environment and health issues. Scientists, policy makers, and the public employ scientific, political, and social rationality, respectively. These differing forms of rationality reflect underlying epistemological distances from which can develop considerable misunderstandings and misinterpretations.     

Knowledge Management in Strategic Alliances: A Review of Empirical Evidence

Knowledge‐related and organizational learning processes in alliances have received much attention throughout the last 25 years. The field has generated a rapidly growing body of empirical evidence on how knowledge is managed in alliances. However, the sphere is highly complex, fragmented, incoherent, and heterogeneous in terms of the theoretical approaches applied. This paper presents an integrative and organizing framework for the empirical literature on knowledge management in strategic alliances. It illustrates how the knowledge management outcomes of knowledge creation, transfer and application are determined by four distinct sets of factors: knowledge characteristics, partner characteristics, partner interaction, and active knowledge management. Based on this framework, this review analyses and integrates empirical evidence in order to identify where findings converge and where results conflict. So far, research has focused strongly on singular interrelations between these four sets of factors and the transfer of knowledge. Conversely, the questions of how knowledge is created, retained, retrieved and applied and how the interplay of the different factors affects knowledge management in strategic alliances remain widely unexplored. The review concludes with a summary of the current state of the art in empirical research and discusses some promising avenues for future investigation.     

Using our brains to develop better policy

Current governmental practices often use a method called weight of evidence (WoE) to integrate and weigh different sources of information in the process of reaching a decision. Recent advances in cognitive neuroscience have identified WoE-like processes in the brain, and we believe that these advances have the potential to improve current decision-making practices. In this article, we describe five specific areas where knowledge emerging from cognitive neuroscience may be applied to the challenges confronting decisionmakers who manage risks: (1) quantifying evidence, (2) comparing the value of different sources of evidence, (3) reaching a decision, (4) illuminating the role of subjectivity, and (5) adapting to new information. We believe that the brain is an appropriate model for structuring decision-making processes because the brain's network is designed for complex, flexible decision making, and because policy decisions that must ultimately depend on human judgment will be best served by methods that complement human abilities. Future discoveries in cognitive neuroscience will likely bring further applications to decision practice.     

Explaining the Diffusion of Knowledge Management: The Role of Fashion

This paper provides evidence on the emergence and diffusion of the discourse of knowledge management. A literature review of the knowledge management and learning organization literatures demonstrates the lack of learning from one discourse to another and major differences in the concerns and issues that they address. At the same time, evidence on the level of interest in each discourse shows a tendency towards a normal curve distribution. Analytically, these findings suggest that the widespread diffusion of knowledge management might be explained in terms of the management fashion model. However, further consideration of the professionally-differentiated appropriation of knowledge management concepts by the information systems and human resource communities suggests that the fashion model provides only a partial explanation for the observed diffusion of knowledge management.     

知识共享与组织绩效:知识驱动的人力资源管理实践作用研究

本文基于知识观.提出并验证了知识驱动的人力资源管理实践的三维结构及知识相关的组织绩效的两维结构,探讨了知识驱动的人力资源管理实践对知识共享和组织绩效的影响作用.实证研究表明:第一,注重胜任力和学习潜力的人员选拔、多样性的培训与开发对知识共享和知识探索性绩效具有显著的正向影响,多样性的培训与开发对知识拓展性绩效具有显著的正向影响;第二,知识共享具有中介作用;第三,知识驱动的人力资源管理实践与知识共享对知识探索性绩效的影响要强于对知识拓展性绩效的影响.本研究结果从现代知识管理的独特视角,对人力资源管理实践影响组织绩效的"黑箱"(Black Box)进行了一次新的有益探索,揭示了知识驱动的人力资源管理实践的作用机理.     

The soft side of knowledge transfer partnerships between universities and small to medium enterprises: an exploratory study to understand process improvement

Abstract

The study explores the soft side of knowledge transfer partnerships between universities and small to medium enterprises (SMEs), a topic which is often neglected in the knowledge management literature. The aim of this paper is to uncover the issues which emerge during formation of a partnership between heterogeneous organizations and universities. In addition, the study unfolds the criticalities of typical process improvement capability that supports the knowledge transfer partnerships between universities and SMEs. Using multiple cases, this study unravels the dominant elements that influence knowledge transfer process development, governance, implications and responsibilities. The major contribution of this study is the development of a framework based on empirical evidence using three knowledge transfer partnerships (KTPs) which illustrates the way in which soft factors in knowledge transfer partnership phases may have an impact on success or failure of university–industry collaborations for innovation.     

基于加权知识网络的个人知识存量表示与度量方法

对个人知识存量的表示与度量方法进行了研究。在对个人知识存量结构分析的基础上,提出了个人知识存量的加权知识网络模型,该模型将个人知识体系表示为一个层级有向网络,并通过对边进行加权和对节点赋值来实现个人知识存量的度量,其中,节点的属性值可反映个人对不同知识点的掌握程度,边则反映了知识点之间的构成关系。还对知识点的获取和度量方法进行了探讨。最后结合一个实例进行了验证。