Technical Note: Selecting a Classification Method by Cross-Validation

If we lack relevant problem-specific knowledge, cross-validation methods may be used to select a classification method empirically. We examine this idea here to show in what senses cross-validation does and does not solve the selection problem. As illustrated empirically, cross-validation may lead to higher average performance than application of any single classification strategy, and it also cuts the risk of poor performance. On the other hand, cross-validation is no more or less a form of bias than simpler strategies, and applying it appropriately ultimately depends in the same way on prior knowledge. In fact, cross-validation may be seen as a way of applying partial information about the applicability of alternative classification strategies.     

Germanium islands embedded in strained silicon quantum wells grown on patterned substrates

Germanium islands were embedded in strained silicon quantum wells in order to provide an improved electron confinement in vicinity of the islands. Growth was performed on relaxed SiGe layers. Patterned substrates were used, favouring lattice relaxation as well permitting the fabrication of small Ge islands at deposition temperatures above 500 °C. Photoluminescence analysis reveals a strongly reduced dislocation related signal. The low temperature spectra are dominated by intense signals from the germanium islands. The origin of these signals were investigated by removing the islands by etching, analysing reference samples without a silicon quantum well, varying the germanium deposition and the growth temperature.     

Flexible microelectronics becoming a reality with Suftla transfer technology

Abstract— Suftla is a technology that is used to transfer polycrystalline silicon (polysilicon) thin‐film‐transistor (TFT) circuits from an original glass substrate to a plastic sheet. The electronic devices in the next generation will be thin, lightweight, and will handle huge amounts of data, yet consume less energy. Suftla technology, together with high‐performance polysilicon TFTs, meets all these requirements because we have developed a variety of smart flexible electronic devices, such as thin paperback‐sized displays and microprocessors. Suftla will usher in a new era of life‐enhancing flexible microelectronics.     

我国的金矿物分类及其主要特征

我国的金矿物约49种(包括变种和未定名矿物).经济矿物主要是自然金和银金矿,少数矿床有金银矿、碲金矿、针碲金银矿、碲金银矿和黑铋金矿等.上述金矿物可分为四大类和十五种类型.文中对各种矿物的主要特点作了简要介绍.     

货运调车场减速顶噪声及治理的探讨

铁路货物运输枢纽的调车场中，驼峰调车是主要的调车手段，为提高生产效率和确保安全，沿驼峰下滑道布置的减速器（减速闸，减速顶）的噪声是调车场噪声的主要声源之一，本文通过对昆明东站铁路降噪工程的观测试验项目，探讨了减速顶噪声的特性及治理措施。     

A stability classification method and its application to pipelined solution of linear recurrences

A new method of classification for numerical stability of parallel algorithms is proposed based on the theoretical foundation of forward error analysis. It partitions the algorithms according to their asymptotic stability—a measure introduced to relate the limiting behavior of the stability to the size of the problem. Using this method, the stability aspect of the pipelined solution technique for first-order and second-order linear recurrences—the core of a tridiagonal linear equation solver—is studied. In particular, it shows that the pipelined solution method of the first-order linear recurrences has the same degree of stability as the commonly used sequential evaluation algorithms. The stability problems of sequential and pipelined solution methods of the second-order linear recurrences are also studied.     

航天产品多源可靠性信息的收集与整理--建立可靠性信息数据库

航天产品可靠性信息的来源多样，可靠性信息的时间性、随机性、有价性、时效性和可追溯性等特征，决定了可靠性数据收集应满足真实性、连续性、完整性等基本要求。同时，对多源信息的收集应遵循先进行需求分析，再确定数据收集点并制定数据收集表格的程度进行。给出了信息的适用性准则和信息分类的三维模型。最后结合具体型号航天产品的可靠性工作的要求，给出了可靠性信息数据库系统的整体方案设计。     

Defect formation in epitaxial oxide dielectric layers due to substrate surface relief

Defects were characterized in epitaxial (001) CeO₂ films deposited and planarizedin situ on patterned (001) LaAlO₃ substrates by ion beam assisted deposition (IBAD). A hill and valley structure with steps running parallel to the [100] LaAlO₃ axis was produced on the surface of the substrate by photolithography and ion beam etching prior to film deposition. A conformai epitaxial CeO₂ layer of ∼ 100 nm thickness was deposited on the heated substrate by e-beam evaporation. Lattice-matching between the e-beam film and the substrate was of the type: (001) CeO₂∥(001) LaAlO₃ and [110] CeO₂∥[100] LaAlO₃. Evaporative deposition of additional film onto the conformai layer was accompanied by bombardment with a 500 eV argon/oxygen ion beam to promotein situ planarization. Extreme lattice misfit for the orientation (001) CeO₂∥(001)LaAlO₃ and [001] CeO₂∥[001] LaAlO₃ caused formation of dislocations in the e-beam CeO₂ film in the vicinity of individual ledges in the substrate surface. Coherence of the CeO₂ film was locally lost in the step regions of the hill and valley structure. The large patterned steps, which are composed of numerous adjacent ledges in the LaAlO₃ surface, caused nucleation of CeO₂ with a tilt misalignment of up to ∼5‡ about the substrate [100]. Nucleation and growth of nonepitaxial CeO₂ crystallites was observed along the step regions of the film during the IBAD portion of deposition. Defect formation in the e-beam ceria layer due to substrate surface relief indicates that “lattice engineering≓ of multilayer epitaxial structures may not be possible when nonplanar surfaces are created during device fabrication. The IBAD CeO₂ layer was more defective than the conformai layer deposited without the impinging ion beam, even in the portions of the film where epitaxy was maintained throughout both layers.     

On-Line Shape recognition with incremental training using binary synaptic weights algorithm

Recognition of hand drawn shapes is beneficial in drawing packages and automated sketch entry in handheld computers. In this paper, we propose a new approach to on-line geometric shape recognition with incremental training function, which utilizes a heuristic function to reduce noise and a neural network for classification and on-line training. Instead of recognizing segments of a drawing and then performing syntactical analysis to match with a predefined shape, which is weak in terms of generalization and dealing with noise, we examine the shape as a whole. The main concept of the recognition method is derived from the fact that internal angles are very important in the perceived shape of outlines. Our application's aim is to recognize elliptic, rectangular, and triangular shapes in a way similar to human cognition of these shapes. Human beings recognize such basic shapes regardless of the variations in size, noise on the shape border, rotation and in the case of triangles, regardless of the type of the triangle. The key concept is that the neural network learns the relationships between the internal angles of a shape and its classification, therefore only a few training samples which represent the class of the shape are sufficient. Fast meremental training, which is performed on-line, is accomplished by the use of the Binary Synaptic Weights algorithm, a one pass, feedforward neural network training algorithm. Incremental training offers the advantage of adjusting the recognition capability of the system to the user's drawings. the results are very successful, such that the neural network correctly classified shapes that did not have any resemblance to the shapes in the initial training set.     

基于Sql Server BI平台和决策树技术的CET-4成绩分析

文章首先对Sql Server 2008商业智能平台及决策树技术相关理论作了介绍,然后对挖掘数据源数据进行了一系列预处理。利用Sql Server 2008商业智能平台下的决策树技术并采用数据挖掘扩展语言即DMX语言创建了CET-4成绩分析决策树模型。分类矩阵和挖掘提升图分别对该模型的评估结果表明了模型具有较高的可靠性和分类准确度,同时模型对应的一些规则可作为英语教学管理和改革的重要参考依据。