First and Second Order SMO Algorithms for LS-SVM Classifiers

Least squares support vector machine (LS-SVM) classifiers have been traditionally trained with conjugate gradient algorithms. In this work, completing the study by Keerthi et al., we explore the applicability of the SMO algorithm for solving the LS-SVM problem, by comparing First Order and Second Order working set selections concentrating on the RBF kernel, which is the most usual choice in practice. It turns out that, considering all the range of possible values of the hyperparameters, Second Order working set selection is altogether more convenient than First Order. In any case, whichever the selection scheme is, the number of kernel operations performed by SMO appears to scale quadratically with the number of patterns. Moreover, asymptotic convergence to the optimum is proved and the rate of convergence is shown to be linear for both selections.     

Meshless Finite Difference Method with Higher Order Approximation—Applications in Mechanics

This work is devoted to some recent developments in the Higher Order Approximation introduced to the Meshless Finite Difference Method (MFDM), and its application to the solution of boundary value problems in mechanics. In the MFDM, approximation of the sought function is described in terms of nodes rather than by means of any imposed structure like elements, regular meshes etc. Therefore, the MFDM, using arbitrarily irregular clouds of nodes using the Moving Weighted Least Squares (MWLS) approximation falls into the category of the Meshless Methods (MM). The MFDM, dating to early seventies, is one of the oldest and possibly the most developed one. In this paper considered are some techniques which lead to improvement of the MFDM solution’s quality. The main objective of this paper is the presentation and overview of new ideas and the development of the Higher Order solution approach in the MFDM provided by correction terms, preceded by a brief information about the current state-of-the art of this method. The main concept of the Higher Order Approximation (HOA) used here, is based on consideration of additional terms in the local Taylor expansion of the sought function. It shall be demonstrated that such a move may essentially improve, in many ways, efficiency and solution quality of the Higher Order MFDM. The Higher Order correction terms may be applied in many aspects of the MFDM solution approach. Among them one may distinguish the a-posteriori error estimation as well as adaptive solution process with multigrid strategy. Moreover, in the present work considered are: computational implementation of the Higher Order MFDM algorithms, examination of the above mentioned aspects using 1D and 2D benchmark tests, as well as an application of the Higher Order MFDM solution approach to selected boundary value problems in mechanics.     

SYSTEMATIC SYNTHESIS OF PARALLEL ARCHITECTURES FOR THE REAL-TIME ESTIMATION OF HIGHER ORDER STATISTICAL MOMENTS

Abstract

The Higher Order Statistics, such as the Higher Order Moments, Cumulants and Polyspectra, have been recognized as important tools in modem time series analysis since they overcome well-known limitations of the autocorrelation/power spectrum second order methods. The systematic synthesis of parallel algorithms and architectures for the real-time estimation of moments up to any desirable maximal order k > 3 is presented. First, a design methodology is developed which can take into account the desirable characteristics of the targeted parallel architecture and used to construct an optimal locally recursive form of the algorithm amenable to efficient parallelization. The design methodology is then used to synthesize a family of algorithms and minimum latency, low granularity, processor array architectures that can compute all lags of Higher Order Moments, from the samples of the incoming data sequence in real-time.     

Aesthetic design of e-commerce web pages – Webpage Complexity,Order and preference

This study was conducted to understand the perceptual structure of e-commerce webpage visual aesthetics and to provide insight into how physical design features of web pages influence users’ aesthetic perception of and preference for web pages. Drawing on the environmental aesthetics, human–computer interaction, and psychology literature, we identify webpage visual Complexity and Order as two salient webpage aesthetic features and investigate the moderating role of the e-commerce customer’s motivational orientation on his/her preference for web aesthetic features. Two independent samples were recruited and presented with images of 24 homepages varying across six levels of Complexity and four levels of Order. The results of this study suggested significant influences of webpage Complexity and Order on customer’s preference for the web pages and a moderating effect of customer’s shopping motivational orientation on his/her preference for webpage Complexity. However, no moderating effect of motivational orientation was found on the preference for webpage Order.     

Reader Feedback

Abstract

Security Analysis and Enhancements of Computer Operating Systems, by R. P. Abbott and others. U. S. Department o f Commerce, National Bureau of Standards. (Order from National Technical Information Service, Spring field, VA, 22151. Order No. NBSIR-76-1041.) April 1976; 69 pages; $4.50.     

Cry Woolf

The Woolf Reforms, recently instituted to make civil litigation more user friendly, targeted, amongst other things, the investigator’s best friend, the Anton Piller Order (AP). The Order was targeted directly, by having its name changed to the Search Order, and indirectly, in that Woolf aimed to make claimants expend more effort in finding solutions before resorting to the “nuclear option” of the AP.     

Impacts of remote sensing on U.S. geography

This paper examines some impacts of remote sensing on geography. As geographers involved in this new technology, we review and place in context its development; and with a vision of its value to our discipline, we are concerned at its modest impact on major research areas in geography. Even acknowledging that, in a broader view, remote sensing's impact on geography has been greater than in other disciplines does not alleviate our concerns that few senior academic geographers have remote sensing research underway; that few economic geographers have examined its potential; that even fewer regional geographers with interests in less-developed countries have remote sensing research interests; and that although there has been growth in the number of remote sensing courses taught in geography, this has not been accompanied by publication in reviewed geographic journals. We ask the questions: What does remote sensing permit the geographer to do better and/or cheaper than he or she could do in the past? What may remote sensing enable future geographers to do which would be significant to professional and academic geography? We then find that the exploitation of the improved or unique information available to the geographer via the application of remote sensing techniques has barely begun. Yet, remote sensing is a reality within geography whose time has come. It is too powerful a tool to be ignored in terms of both its information potential and the logic implicit in the reasoning process employed to analyze the data. We predict it could change our perceptions, our methods of data analysis, our models, and our paradigms.     

α稳定分布噪声下的空间时频DOA估计

当信号中存在α稳定分布噪声时,传统空间时频多重信号分类(STF-MUSIC)算法的空间波达方向(DOA)估计性能会降低甚至失效。为此,利用分数低阶矩(FLOM)代替二阶协方差矩阵,定义分数低阶矩空间时频分布矩阵(FLOM-STFDM)。对FLOM-STFDM进行特征分解,得到适用于稳定分布噪声环境的空间时频TF-FLOM-MUSIC算法,分析该算法的信噪比及误差估计,并给出算法实现步骤。仿真结果表明,TF-FLOM-MUSIC算法可有效降低DOA估计的均方误差,提高估计的分辨率和平滑性。     

Headquarter-centered common sourcing management through order coordination and consolidation

It is common for subsidiaries of a group company to use the same types of components for producing similar products. Different subsidiary companies may well procure such components from the same suppliers. This paper studies two sourcing management models. One is the Subsidiary-Autonomous Sourcing Management (SD-ASM) where subsidiaries manage their inventories and place purchasing orders independent of each other. The other is the Headquarter-centered Common Sourcing Management (HQ-CSM) where purchasing orders of subsidiaries are processed centrally through some kind of headquarter coordination. In the SD-ASM model, each subsidiary places replenishment orders at a time interval corresponding to their economic order quantity (EOQ). In the HQ-CSM model, two purchasing order management policies are examined. One is the Order Coordination policy in which common replenishment epochs or time periods are proposed by the headquarter and the subsidiaries are encouraged to coordinate the timing of their orders based on the common replenishment epochs. The other is the Order Consolidation policy in which the subsidiaries combine the quantity of their orders and the headquarter places a combined order with the supplier. In the Order Coordination policy, classic RAND heuristic is used to find the best common replenishment epoch and the best replenishment timing of each subsidiary. In the Order Consolidation policy, the optimal order quantity of the combined order is obtained from a mathematical model. The combined order is then allocated to the subsidiaries according to a proportional allocation rule. A series of numerical studies is conducted to compare the costs of the SD-ASM and HQ-CSM policies. The results show that HQ-CSM outperforms SD-ASM in terms of cost and robustness against demand uncertainties. This achievement is largely due to the economies of process (synergistic ordering process), the economies of scale (large order quantity with price discount) and risk pooling effect (transshipments). The results also reveal that the Order Consolidation policy with a combined order always performs better than the Order Coordination policy with common replenishment epochs especially in face of high demand uncertainties and high service level in the global market.     

基于锁集合的动态数据竞争检测方法

数据竞争使得共享存储程序难于调试．以前大部分针对共享存储程序的动态数据竞争检测工作都是通过维护发生序来实现．这种方法有一个重要缺点，即针对程序的一种输入，对程序的一次执行进行检测，不能检测出所有的可行数据竞争．文中利用存储一致性模型的框架模型，针对域一致性模型提出了增强发生序概念，并依此得出一种基于锁集合的动态数据竞争检测算法，克服了这个问题．在软件DSM系统JIAJIA上的实现获得了很好的性能，应用平均减速比为3．14．利用该方法，在TSP程序中找到了大量的读写数据竞争的情况．     

Activity-Based Costing and Management applied in a hybrid Decision Support System for order management

This article introduces a new Cost Management and Decision Support System (DSS) applicable to Order Management. This model is better fit and compatible with today's competitive, and constantly changing, business environment. The presented Profitable-To-Promise (PTP) approach is a novel modeling approach which integrates System Dynamics (SD) simulation with Mixed-Integer Programming (MIP). This Order Management model incorporates Activity-Based Costing and Management (ABC/M) as a link to merge the two models, MIP and SD. This combination is introduced as a hybrid Decision Support System. Such a system can evaluate the profitability of each Order Fulfillment policy and generate valuable cost information. Unlike existing optimization-based DSS models, the presented hybrid modeling approach can perform on-time cost analysis. This will lead to better business decisions based on the updated information.     

Mobile broadband internet access service is a commercial mobile service,and hence must be regulated as a common carrier service

ABSTRACT

The FCC’s Restoring Internet Freedom Order reclassified mobile broadband service from a commercial mobile radio service to a private mobile radio service, thereby eliminating common carrier regulation of the service. The reclassification is an incredible assertion: that the most important public mobile service of our time is classified under statute as a private mobile service. We analyze the relevant precedent from the 1940s through 2017. We find that the Order’s lack of consideration of the relevant precedent and technology undermines its interpretation of the statute on which it relies. The Order’s reversion to the 1994 definitions of public switched network and interconnected service ignores the growth of the public switched network to include the Internet. The Order’s conclusion that the public switched telephone network and the Internet are separate non-interconnected networks is factually wrong. Critically, when interpreting interconnection, the Order ignores the fact that in order for meaningful communication to occur, the users’ devices and subscribed services must be compatible. A proper interpretation of relevant statute and precedent leads to the opposite conclusion of the Order, that mobile broadband service is a commercial mobile service. This finding lays the foundation for reinstatement of net neutrality protections on mobile broadband service.     

基于SVM的地理试题自动分类

本文针对地理课程自动解题,采用SVM学习算法实现地理试题自动分类。通过采用TFIDF技术提取地理试题文本中的特征关键词,并选取LibSVM中的Linear核函数进行训练,以构建用于地理试题分类的预测模型。在所收集的地理试题集上的实验结果表明,在22个试题类别上的单类分类精度达到80%以上,整体分类精度也达到了87%。     

优序法在软件质量评价中的应用

优序法是一种用于评价的较好方法，目前已在管理成果评价等方面得到了应用，该方法同样达用于软件质量评价，该方法应用简单，即能处理定性问题，又能处理定量问题，文章给出了优序法在软件质量和评价中的应用。     

关于三阶统计量定义的研究

高阶统计量是信号处理中的一种重要方法,已经得到较为广泛的运用。本文在三阶统计量的传统定义上提出了一种新的三阶统计量定义,提出新定义较之旧定义的优越性,并通过实验验证该结论。     

模型的快速降阶技术

为快速进行模型的降阶,结合平衡截断（Balanced Truncation,BT）方法和特征正交分解（Proper Orthogonal Decomposition,POD）方法提出一种模型降阶方法.该方法采用频域POD快照矩阵低阶逼近系统的可控、可观Gram矩阵;通过奇异值分解（Singular Value Decomposition,SVD）提取BT＋POD模态,对低能量模态截断形成降阶子空间,并将其映射到全阶系统,从而形成基于状态空间的降阶模型（Reduced Order Model,ROM）;该模型就成为全阶模型（Full Order Model,FOM）的ROM.通过对阶数n=406的LTI SISO系统和阶数n=9的2区间电力系统进行的验证表明,在保留BT方法输入输出平衡特性的基础上,该方法效率高于BT方法.     

GAME

哈利波特与凤凰社;变形金刚;忍者龙剑传∑;科林麦克雷 尘埃;霸王.     

A decision support system for real-time order management in a heterogeneous production environment

In today's competitive market, many companies are morphing from the traditional new build, single brand, and silo environments to facilities accommodating diverse business missions. The later are called heterogeneous production environments in which the different business channels share their final production stage (shipping) to enable competitive advantages. In these production environments, at the operational level, the critical success factors are customer satisfaction, on-time delivery, product complexities, supply allocation, and resource utilization. At the strategic level, the success factors are revenue, customer urgency, and sales impact. This study proposes an End-to-End Customer Order Management System (E2E COMS) focusing on effective utilization of individual and shared resources to support real-time order management and mitigate risk of managing diverse missions. The proposed system consists of three integrated tools: Order Prioritization Tool (OPT) to assess and prioritize customer orders for each business channel, Order Fulfillment Progress Projection Tool (OFPPT) to predict the expected remaining order completion time considering inventory and resource capacity constraints, and risk mitigation tool to assess the risk of missing an order shipment due to shipping constraints. A real-time dashboard is developed to visualize the prioritized customer orders, expected time to arrive at the shipping area, shipping instructions, and two-dimensional risk assessment charts. The proposed system can effectively be used for shipping capacity management as well as prompt decision making.     

地理因子对卫星成像质量影响预估方法研究

重点研究地理因子对卫星成像质量的影响,预估影响程度,为卫星规划调度提供辅助决策信息。本文分别分析光照、地形起伏、云层3个地理因子对可见光传感器成像质量的影响,给出考虑这些因素的可见光传感器辐射接收计算公式,并结合ArcGIS分析功能,建立了地理因子对可见光传感器成像质量影响预估模型。同时,考虑地形起伏对SAR成像的影响,给出了利用DEM数据求几何畸变系数的方法,利用ArcGIS建立地理因子对SAR成像质量影响预估模型。两个模型都给出了成像质量影响程度预估值计算方法,最后,本文基于ArcGIS软件实现其原型系统。     

支持向量回归在不确定面向订单装配环境下交货期预测中的应用

针对面向订单装配(ATO)生产环境,如何根据订单信息、生产系统特性快速地估算出准确、可靠的交货期问题,在分析不确定性要素对交货期影响机制的基础上,构建了订单交货期预测模型。模型参数包括三个部分：订单上线时间、装配周期和异常拖期。订单上线时间基于零部件、生产能力的可用性,订单装配周期和异常拖期采用基于实际生产历史数据的支持向量回归(SVR)方法进行预测。案例研究表明该模型预测结果与实际交货期接近,可以用于指导订单交货期协商。