A class of monotone kernelized classifiers on the basis of the Choquet integral

The key property of monotone classifiers is that increasing (decreasing) input values lead to increasing (decreasing) the output value. Preserving monotonicity for a classifier typically requires many constraints to be respected by modelling approaches such as artificial intelligence techniques. The type of constraints strongly depends on the modelling assumptions. Of course, for sophisticated models, such conditions might be very complex. In this study, we present a new family of kernels that we call it Choquet kernels. Henceforth, it allows for employing popular kernel-based methods, such as support vector machines. Instead of a naïve approach with exponential computational complexity, we propose an equivalent formulation with quadratic time in the number of attributes. Furthermore, because coefficients derived from kernel solutions are not necessarily monotone in the dual form, different approaches are proposed to monotonize coefficients. Finally, experiments illustrate beneficial properties of the Choquet kernels.     

采用编码输入的生成对抗网络故障检测方法及应用

针对传统基于生成对抗网络的故障检测方法中，生成器输入使用随机噪声，不包含训练集中有效信息造成模型检测效果不够理想的问题，提出一种采用编码输入的生成对抗网络故障检测策略。通过引入自编码器，基于最小化重构误差构建隐变量空间，将降维后的隐变量信息作为生成器输入以提升生成对抗网络的训练效果；进一步考虑故障检测方法中基于生成器的统计量计算成本高和对离群点敏感的问题，计算待测样本经编码后的隐变量到训练集隐变量空间中心点的曼哈顿距离，并作为新统计量进行故障检测。将所提故障检测方法用于田纳西伊斯曼过程及实际的磨煤机过程，本文方法较传统生成对抗网络故障检测在田纳西伊斯曼过程上报警率提升了13%，在磨煤机过程上各统计量报警率均得到了显著提升且本文所提统计量将传统方法中针对生成器的统计量大大降低了检测用时，从而验证了方法的有效性和性能。     

面向高级别代码克隆检测方法的设计与实现

代码克隆检测是软件工程中的基础研究，在软件分析和维护方面有着广泛应用。目前对于有文本差异的高级别（即学术界定义的级别3和级别4）克隆检测，现有方法存在检出率（回收率）不高的问题。基于程序依赖图PDG的检测方法是高级别克隆检测的一类重要方法，但这类方法依赖子图同构的精确图匹配算法，算法时间复杂度高且回收率较低。为此，提出了一种新的高级别代码克隆检测方法，使用基于 Weisfeiler-Lehman图核的非精确图匹配算法进行代码克隆检测。实验结果表明，与已有的代码克隆检测方法相比， 该方法可以检出更多的高级别克隆且计算时间较短。     

Kernel Search-Framework for Dynamic Controller Placement in Software-Defined Network

In software-defined networking (SDN) networks, unlike traditional networks, the control plane is located separately in a device or program. One of the most critical problems in these networks is a controller placement problem, which has a significant impact on the network’s overall performance. This paper attempts to provide a solution to this problem aiming to reduce the operational cost of the network and improve their survivability and load balancing. The researchers have proposed a suitable framework called kernel search introducing integer programming formulations to address the controller placement problem. It demonstrates through careful computational studies that the formulations can design networks with much less installation cost while accepting a general connected topology among controllers and user-defined survivability parameters. The researchers used the proposed framework on six different topologies then analyzed and compared with Iterated Local Search (ILS) and Expansion model for the controller placement problem (EMCPP) along with considering several evaluation criteria. The results show that the proposed framework outperforms the ILS and EMCPP. Thus, the proposed framework has a 38.53% and 38.02% improvement in reducing network implementation costs than EMCPP and ILS, respectively.     

Pursuing the kernel of truth in the psychotherapy of schizophrenia.

This article concerns the utility of pursuing the kernels of truth in the delusions of schizophrenic patients as a means of empathizing with their unique personal experience of events. Pursuing the kernel of truth implies listening for the real interpersonal experiences past and present that are represented within delusions, as opposed to focusing on derivatives of infantile fantasy. This orientation restores the rights of the schizophrenic person as a perceptive observer of events with a valid, although not commonly understood, point of view. A clinical case example is presented in which the delusion of being raped while others stand idly by both literally and metaphorically described a wide variety of interpersonal experiences which had shaped the patient's life. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Melting and solidification characteristics of confectionery fats: Anhydrous milk fat,cocoa butter and palm kernel stearin blends

Differential scanning calorimetry measurements of crystallization and melting characteristics of commercial samples of anhydrous milk fat (AMF), cocoa butter (CB) and hydrogenated palm kernel stearin (PKS) in ternary blends were studied. Results showed that stabilization at 26°C (either for 40 h or 7 d) did not greatly affect the melting thermogram trace of PKS. However, the effect of stabilization became prominent as CB was added into the system. Deviation of measured enthalpy from the corresponding values, calculated for thermodynamically ideal blends, showed clear interaction between all three fats. At 20°C, the strongest deviation occurred at about the AMF/CB/PKS (1∶1∶1) blend, whereas at 30°C the deviation moved toward the CB/MF (1∶1) blend. The presence of 25% AMF in PKS had little effect on its solidification capability, but solidification was adversely affected with inclusion of CB.     

基于嵌入式Linux的综合录井系统研究

针对国内外现存的油田钻井监控系统的现状，提出了一种适用于离散井群的综合录井系统，该系统以嵌入式操作系统Linux为核心。由于嵌入式Linux的出现使得构造一个大型机器的复杂控制系统变得简单和容易。同时结合综合录井系统的工艺特点，将嵌入式Linux操作系统应用到综合录井控制系统中的一些关键问题，重点是讨论构造其控制器的软、硬件设计方法。     

确定任意多边形的核的算法

本文提出确定任意简单多边形Ｌ的核的算法，该算法的时间复杂性是Ｏ（ｌｎ）次乘法，其中ｎ是多边形的Ｌ的顶点数，ｌ是多边形Ｌ中凹点的数目。     

A spline strip kernel particle method and its application to two‐dimensional elasticity problems

In this paper we present a novel spline strip kernel particle method (SSKPM) that has been developed for solving a class of two‐dimensional (2D) elasticity problems. This new approach combines the concepts of the mesh‐free methods and the spline strip method. For the interpolation of the assumed displacement field, we employed the kernel particle shape functions in the transverse direction, and the B₃‐spline function in the longitudinal direction. The formulation is validated on several beam and semi‐infinite plate problems. The numerical results of these test problems are then compared with the existing solutions obtained by the exact or numerical methods. From this study we conclude that the SSKPM is a potential alternative to the classical finite strip method (FSM). Copyright © 2003 John Wiley & Sons, Ltd.