Seeking Mrs. Right: Uncertainty Reduction in Online Surrogacy Ads

Surrogate mothers use online advertisements to reduce the uncertainty associated with the surrogacy journey. This paper identifies five key desirability traits used by surrogates in online advertisements. These perceived desirability traits include idealism, logistics, moral boundaries, willingness parameters, and personal disclosure. Our findings suggest surrogates construct communication messages using these perceived desirability traits to reduce uncertainty and encourage matching.     

Innovative R&D and optimal investment under uncertainty in high-tech industries: An implication for emerging economies

In order to continue on the path to convergence with advanced countries, emerging countries need to strengthen their research capabilities. Often they try to do this through the development of strategic industries. The objective of this paper is to contribute to the debate on the effectiveness of policies aimed at the development of strategic industries in emerging economies. The paper develops a three-phase model for product innovation with capital investment under uncertainty to study the investment decisions of a manufacturer in an industry facing a volatile market demand for new inventions. The findings demonstrate the importance of interactions between market structure, a firm's market power and the associated cost of adjustment. The paper draws out implications for emerging economies with regard to policies striving to develop strategic industries in sectors such as semiconductors and information technology.     

一种基于属性核的Rough集属性约简搜索算法

本文提出了一种新的可以快速寻找Rough集约简属性的算法,该算法利用计算属性核来提高计算效率.首先,从属性核出发,不断地在属性核上添加非核属性以构成新的属性集合,并判断新的属性集合是否为属性约简,然后在此基础上进行剪枝运算并在不构成约简的属性集合的基础上再生成新的属性集合并进行属性约简判断.该算法为一种搜索算法,由于采用了属性核算法和剪枝算法,提高了计算效率并能够找到给定信息系统的全部属性约简.通过测试,表明该算法是一种可行且有效的属性约简算法.     

非线性数据变换及其在离群聚类中的应用

为了提高高维数据集合离群数据挖掘效率,在分析了传统的离群数据挖掘算法优点和缺点的基础上,提出了一种离群点检测算法,首先将非线性问题转化为高维特征空间中的线性问题,然后利用非线性数据变换进行维数约减,对从高维采样数据中恢复得到低维数据集,通过本文提出的离群数据假设,并结合本文给出的离群聚类方法对所得数据对象投影分量是否是离群数据进行判别。仿真实验的结果表明了该方法能够有效地发现高维数据集中的离群点。与此同时,该算法具有参数估计简单、参数影响不大等优点,为离群点检测问题的机器学习提供了一条新的途径。     

基于RBF网络的粗糙集属性约简策略

粗糙集理论被广泛应用于人工智能、模式识别、数据挖掘和知识发现等领域。而对象的属性约简是是粗糙集理论中的重要问题之一。由于属性约简计算量较大,影响了的粗糙集的实际应用。本文用RBF神经网络高效和OLS对称性的特点,研究粗糙集属性的约简,解决了属性约简的难题,完成了算法的实现,取得了较好的效果。     

测不准原理与单缝衍射

本文中说明了如何正确理解海森伯测不准原理，指出不少量子力学［1］中利用单缝衍射推导测不准原理的错误，说明了如何利用单维衍射获得测不难关系的正确方法。     

Do You Know What It Means to Miss New Orleans? Internet Communication, Geographic Community, and Social Capital in Crisis

This study investigated Internet communication and use in a crisis situation, Hurricane Katrina, to explore the role of the Internet in supporting or diminishing geographically-based community during a crisis. The researchers conducted an online survey of Internet users (n=1192) from the dispersed metropolitan New Orleans area. The survey focused on amount, type, function, and importance of Internet use to creating and maintaining social capital, supporting geographically-based communities, activating social networks, reducing uncertainty, and achieving both expressive and instrumental communication goals. The results indicated that Internet users in a crisis situation went online to seek interactive fora specific to their neighborhoods and to activate weak ties in their social networks. They engaged in more uncertainty reduction behavior when experiencing higher degrees of damage. They turned to the Internet in place of other media as a result of disruptions caused by the crisis. Women valued online expressive communication more than men did. The findings suggest that social capital theorists would benefit from a communication perspective on the Internet. The study also led to the formation of suggestions for emergency preparedness agencies, shelter providers, crisis victims, and online news providers that can improve emergency response.     

一种不确定离散时间系统的变结构控制

讨论了利用指数趋近律设计不确定离散变结构控制系统的问题,给出了改进的离散趋近律,得出了新的控制律,并证明了系统在此控制律下可以渐近趋于原点的性质,降低了系统的抖振,同时结合时延技术对干扰进行在线估计,抵消掉系统中的慢变不确定性.最后仿真结果证明了该方法的有效性.     

Bootlegging and path dependency

This paper confirms the importance of path dependency in the accumulation of firm-specific technological competencies. It shows that firms are guided by the selective logic of path dependency in their innovation processes, even if management has no part in decisions to invest in a new business idea. The research focuses on the output of bootlegging, defined as research in which motivated corporate entrepreneurs pursue innovative activity, which they themselves define and secretly organise. Bootlegging emerges as an incremental continuous trial-and-error learning process in R&D and its uncertainty is no better or worse than that of ‘normal’ R&D. The fear of some managers that violation of strategic direction setting and total anarchy will ensue from bootlegging is unfounded.     

A two-step approach to investigate the effect of rating curve uncertainty in the Elbe decision support system

For river basin management, the reliability of the rating curves mainly depends on the accuracy and time period of the observed discharge and water level data. In the Elbe decision support system （DSS）, the rating curves are combined with the HEC-6 model to investigate the effects of river engineering measures on the Elbe River system. In such situations, the uncertainty originating from the HEC-6 model is of significant importance for the reliability of the rating curves and the corresponding DSS results. This paper proposes a two-step approach to analyze the uncertainty in the rating curves and propagate it into the Elbe DSS： analytic method and Latin Hypercube simulation. Via this approach the uncertainty and sensitivity of model outputs to input parameters are successfully investigated. The results show that the proposed approach is very efficient in investigating the effect of uncertainty and can play an important role in improving decision-making under uncertainty.