Total systems intervention and human inquiry: The search for a common ground

Total Systems Intervention (TSI) is an approach to intervening in problem situations which has much to offer where complex interacting issues need to be addressed by the complementary use of intervention methodologies. That such an approach has much in common with Action Research (AR) has been recognized, with much recent effort being devoted to the relationship between AR and Critical Systems Thinking (CST), the theoretical endeavor underpinning TSI. This paper further develops this line of debate and relates AR or Human Inquiry (HI) more directly to TSI, using an information systems intervention to enhance the study. The outcome is a demonstration of how TSI implicitly uses techniques informed from the field of Action Research, and how a more thorough synthesis of HI with TSI might serve to improve the overall intervention process.     

Systems and organizations: Distal and proximal thinking

The paper introduces the contributions to this special issue ofSystems Practice on Systems and Organizations: New Directions in terms ofdistal andproximal thinking. Distal thinking refers to ready-made concepts, to the finished effects and outcomes of thought and action; proximal thinking, to process and event, to the continuous and unfinished. The papers presented here deal with various aspects of the distal/proximal distinction in social systems and organizations, and especially draw out the implications of recent work in information technology, sociology of technology, accounting theory, and organization studies for a proximal conception of systems and organizations.     

Evaluating organizations: A systems and contingency approach

It has become increasingly difficult to keep pace with the amount of information being generated about how to evaluate organizations. If it were not enough that the situation is made difficult by the sheer mass of material on evaluation, clarity is further hindered by many of the publications on the subject failing to make explicit the principles and assumptions upon which they are based. This was the situation confronting the authors when they began a national project with the National Association of Councils for Voluntary Service on the evaluation of the performance of Councils for Voluntary Service. In an attempt to bring some order to the field, this paper adopts a systems and contingency approach to elucidate the nature and practical usefulness of the different methods of evaluation. It first seeks, using some tools of Checkland's soft systems methodology, to present a systematic analysis of the subject of evaluation. Then, in the light of the analysis, an attempt is made to formulate a simple classification of approaches to evaluation which serves to match the different forms of evaluation to the contexts in which they are most appropriate for use.     

Power in organizations: A contingent,relational view

The concept of power is central to social inquiry, yet it is highly varied and enigmatic. There is a wide-ranging and competing variety of subjectivist, objectivist, and relational conceptions of power. The first part of the paper examines this problem in an exegetical but critical manner. The aim is to develop a critical conception of this social phenomenon. The second part of the paper seeks to probe the nature of three types of systems approach in terms of their underwriting assumptions about power in society. The intent is twofold: to uncover the ideologically suppressed truths about the noxious forms of power and to develop a penetrative understanding of how the mystified “truths” in ideological conceptions of power contain hidden meanings that have the critical potential for educative enlightenment, empowerment, and emancipation for systems theory and practice.     

基于hMETIS与FP-Growth的协同创新社区领域知识发现方法

基于在线社区的协同创新作为一种汲取用户智慧的有效机制,有利于企业发掘其产品创新方向.然而大数据时代下,信息过载及失真为社区中的知识获取带来了诸多挑战,且已有研究主要聚焦于利用价值较低的高频知识的提取.因此,为了充分挖掘用户经验与创新知识,本文基于社区知识间的共现关系构建了知识超图模型,并利用超图划分算法hMETIS对知识超图进行了探索性分析,同时结合关联规则算法FP-Growth进一步研究了知识间的关联关系,以此获取不同领域的价值知识.最后,本文通过一个实例证明了该方法的有效性.本文的研究实现了对社区知识的深度挖掘与分析,提出的领域知识发现方法对企业社区管理和产品创新极具指导意义.     

计算实验金融在中国:研究现状及未来发展

通过对国内计算实验金融研究发表论文进行统计和分析后,发现目前我国计算实验金融研究领域学者队伍已经初步形成,但是,国内还缺乏有影响力的专业核心期刊.从发展动力来看,国外相关研究带动和国内基金资助是我国计算实验金融研究领域发展的2个主要推动因素.在对未来的研究方向分析中,发现对金融经济学、行为金融学基础性问题的研究将成为未来计算实验金融的主要研究方向,而建模方法也将由原来的个体简单适应行为刻画转变为具有复杂适应性和交互性的个体行为刻画.     

利用堆数据结构实现邻域重叠社团结构挖掘

基于当前复杂网络中社团划分算法普遍存在算法复杂度过高以及重叠节点挖掘不准确的局限性,提出了一种高效、快速、准确的社团划分算法。基于贪婪算法,建立最大模块度矩阵,并采用堆数据结构,划分非邻域重叠社团。通过分析局部网络的连边情况,计算邻域社团的划分密度,以准确挖掘社团间的重叠节点。新算法经过仿真分析和实证研究表明,算法复杂度降到近线性。     

判据搜索算法及其在DNA序列模式发现中的应用

模式发现是计算生物学一个重要的研究方向,但目前的大部分算法还不能保证获得最优的模式。将模式发现问题转化成层次图的路径搜索问题,推导了针对三个序列片段相似性关系的判据,以其作为剪枝规则提出并实现了一种深度优先的穷举搜索算法:判据搜索算法(CriterionSearchAlgorithm,CRISA)。理论分析表明,对于绝大多数模式发现问题,CRISA具有多项式的计算时间复杂度和线性的空间复杂度。对仿真的和实际的DNA序列数据的测试表明,CRISA能够快速而完全地识别出序列中所有的模式,并且获得了优于其它算法的总体评价。     

动态网络中机器人信息的感知与搜索策略

为建立一个机器人信息动态交互平台,提出机器人自主感知模型,动态建立机器人路径,并基于遗传算法优化多机器人路径。机器人自主感知模型通过抽象各种网络实体,对网络信息进行分类搜索,自主感知实时的网络传感信息,从而动态规划机器人路径。遗传算法优化用于解决多机器人的路径交叉问题,避免机器人工作空间的资源竞争。仿真和实验表明,机器人自主感知模型具有很强的扩展能力和合理的构架,突破了以往移动机器人仅仅使用自身传感器或网络固定传感器的局限,能够完成超过机器人视野范围的任务。     

Optimal reserve price in static and dynamic sponsored search auctions

Sponsored search advertising is a significant revenue source for search engines. To ameliorate revenues, search engines often set fixed or variable reserve price to in influence advertisers＇ bidding. This paper studies the optimal reserve price for a generalized second-price auction （GSP） under both static and dynamic settings. We show that if advertisers＇ per-click value has an increasing generalized failure rate, the search engine＇s revenue rate is quasi-concave and hence there exists an optimal reserve price under both settings. Different from a static GSP auction where the optimal reserve price is proved to be constant, in a dynamic setting the optimal reserve price is dependent on not only advertisers＇ per-click values, but also the number of ad links sold. A search engine should gradually raise reserve price as more qualified advertisers arrive, and maintain the same threshold after all first-page positions are occupied.     

原油价格预测：近30年研究回顾和未来展望

原油价格预测是能源经济和资产定价领域的重要研究方向,其研究热度正逐年上升。文系统梳理了30多年来原油价格预测领域的相关文献,从以下4个方面进行了现有文献的综述与讨论：(1)原油价格预测对象;(2)原油价格的影响因素、预测变量和理论基础;(3)原油价格预测方法与模型;(4)原油价格预测的实践意义。最后,结合当前热点和未来发展趋势给出值得进一步研究的方向：一是使用气候风险等新的预测信息;二是充分利用多源异构数据;三是提高机器学习等预测模型的可解释性。     

调频连续波SAR慢速动目标参数估计与成像

由于连续波合成孔径雷达（synthetic aperture radar, SAR）特有的工作方式,带来了不同的回波信号形式。首先分析了调频连续波（frequency modulated continuous wave, FMCW）SAR动目标回波信号模型,然后在补偿连续波SAR与脉冲式SAR不一致性的基础上,提出了一种基于Chirp Z变换校正动目标回波包络距离弯曲,Radon变换校正剩余距离走动,改进的离散Chirp傅里叶变换（modified discrete Chirp Fourier transform, M DCFT）搜索参数的新方法,动目标仿真数据处理验证了该方法参数估计的准确性和有效性。另外根据估计的参数对动目标成像取得了良好的聚焦效果。     

阵列测向系统中应用四阶累积量算法抑制杂散噪声与干扰

阵列测向中空间相关噪声、杂散的多径反射噪声以及空间杂散干扰是影响测向性能的较大因素。高阶累积量具有抑制这样的杂散噪声与干扰的性能。本文理论分析了这一性能的基本原理,给出了实验结果。     

Guidelines for a constructivist approach: Steps toward the adaptation of ideas from family therapy for use in organizations

Constructivist family therapy offers a model for the application of Maturana's theories to practice. This paper describes key concepts of a constructivist approach and draws on family therapy to provide guidelines for applying them in an organizational setting. It offers a view of the organization as a network of conversations in which change occurs through the coconstruction of new conversations which widen or change the rational domain in which a problem occurs.     

基于超网络演化模型的社区知识发现与分析

知识作为最重要的无形资源在学界受到了广泛关注,尤其是利用知识网络对社区知识进行表达与挖掘.然而这些同质,静态的网络往往难以充分地描述社区中的知识资源要素及变化.为解决异质性表达问题,本文针对大众协同创新社区中知识的增长和消退机制,构建了一种超网络动态演化模型.同时,应用该模型的拓扑特性指标对社区中的知识分布特性进行分析,提出了基于动态超图的新的知识分类,并以此识别高价值的社区知识.最后,本文对该模型进行了实例验证,对实例中不同类型的知识点进行分析,并提出相应的管理策略.本文的研究对于发现社区中的创新知识,实现社区知识的精确管理提供了方法指导,对于提高社区管理水平和启发企业创新趋势具有重要意义.     

DSS信息组织技术的现状与未来

DSS是一种为管理服务的信息系统。在现代社会, 决策任务、决策环境及决策过程是复杂多变的, 因此, 在这种信息系统中, 信息组织的结构、形态及加工处理方法, 都要与复杂多变的决策任务、决策环境及决策过程相适应, 才能为用户提供有效的决策支持。本文在回顾DSS二十年的发展历程中, 已经取得的信息组织技术成就基础上, 阐述了需要继续研究的各种技术方法, 并对进入21世纪未来十年DSS信息组织技术研究可能取得的突破作了进一步展望。     

A TABU SEARCH APPROACH TOWARDS CONGESTION AND TOTAL FLOW MINIMIZATION IN OPTICAL NETWORKS

This paper considers rearrangeable multihop lightwave networks whereby each network node is equipped with a number p of transmitters and receivers, and a spectrum of wavelengths is accessible by, and shared among, all nodes by using the Wavelength Division Multiplexing (WDM). Depending on input traffic flow, nodal transmitters and receivers can be re-tuned to create virtual connectivity best suited with respect to a given optimization criterion. We present an efficient heuristic algorithm that combines two criteria for optimization: throughput maximization, as well as total flow minimization. Throughput maximization criterion is equivalent to congestion minimization, while minimizing total flow under the assumption of having links with equal lengths implies minimization of the average number of hops. Taking into account lengths of the links (i.e. link costs proportional with distances), the total flow minimization becomes equivalent to the total delay minimization. Tabu search is implemented as a two-ph     

Development and implementation of a technique for norms-adaptable agents in open multi-agent communities

In open normative multi-agent communities, an agent is not usually and explicitly given the norms of the host agents. Thus, when it is not able to adapt the communities’s norms, it is totally deprived of accessing resources and services from the host. Such circumstance severely affects its performance resulting in failure to achieve its goal. Consequently, this study attempts to overcome this deficiency by proposing a technique that enables an agent to detect the host’s potential norms via self-enforcement and update its norms even in the absence of sanctions from a third-party. The authors called this technique as the potential norms detection technique (PNDT). The PNDT consists of five components: Agent’s belief base; observation process; potential norms mining algorithm (PNMA); verification process; and updating process. The authors demonstrate the operation of the PNMA algorithm by testing it on a typical scenario and analyzing the results on several perspectives. The tests’ results show that the PNDT performs satisfactorily albeit the success rate depends on the environment variables settings.     

编队协同对空目标探测及搜索力的最优分配

为探索编队空战中快速有效地搜索空中目标的理论和方法,将搜索论引入到编队对空中目标协同探测研究中.将雷达发现目标概率密度作为对空目标连续探测程度的量化指标,通过首次发现目标的概率,导出了无先验信息条件下编队协同搜索的概率计算式.考虑到雷达对目标探测的不确定性,提出了利用3机协同搜索确定空中目标位置和速度方向的几何方法.根据目标位置划分搜索空域,建立了预警机指挥下的编队搜索力最优分配模型.通过示例验算表明,所提方法是有效的,并且用Lingo软件求解较大规模目标搜索力分配问题可满足实时性要求.