一个校园小说网站的设计与实现

网络小说的重要物质载体是网络媒体.本文对网络小说的特点以及建设小说网站应该注意的问题进行说明,并设计和实现一个校园小说网站,为大学学生通过互联网技术进行小说阅读和发表提供一个平台.     

新颖的全局和声搜索算法在经济性调度中的应用

提出采用新颖的全局和声搜索算法来解决经济调度问题,并设计了一种新颖的处理系统约束的方法;介绍了经济调度问题数学模型、新颖的全局和声搜索算法实现过程及其应用方法。实验结果表明,采用新颖的全局和声搜索算法所获得的最优值要明显好于采用进化算法、粒子群算法所获得的最优值,新颖的全局和声搜索算法为解决经济性调度问题提供了一种新的解决方案。     

A novel distortion elimination method of power amplifier in wideband OFDM system

The nonlinear,memory and saturate distortion caused by power amplifier lead to low BER per-formance and low power efficiency in wideband OFDM system.To solve the related problems,a novel power amplifier based on combined amplification with multi power amplifiers is proposed in this paper.In addition,a novel simplified filter look-up table predistorter is proposed.The novel predistorter connects to a memoryless predistorter subsystem with an adaptive filter subsystem in cascade,and can eliminate the nonlinear memory distortion effectively.The simulation results indicate that the power efficiency and BER performance will be improved by using the novel amplifier structure together with the novel predistorter.     

新型前馈网络学习算法在语音识别中的应用

简要介绍一种全新的前馈神经网络学习算法--代数算法,以该理论提出者针对这一算法做出的多方面的理论证明为基础,将此算法与传统BP算法相比较,利用其在网络结构和获取全局最优点方面具有传统BP算法不可比拟的优点,在matlab7.0实验平台上,分别把传统BP算法和此种新型学习算法应用于语音识别,通过仿真实验说明了新型算法的有效性,语音识别的精度在一定程度上得到了改善.     

Computing solutions to medical problems via Sinc convolution

We illustrate some novel procedures of using Sinc methods to compute solutions to three types of medical problems. The first of these is a novel way to solve optimal control problems, the second is an original way to reconstruct images for X-ray tomography, and the third is a novel way to do ultrasonic tomography inversion. Each of these procedures uses Sinc convolution, which is a novel computational procedure for obtaining accurate approximations to indefinite convolutions     

基于图谱域移位的带限图信号重构算法

针对带限图信号的重构问题, 本文提出了基于图谱域移位的带限图信号重构模型, 该模型将图带限分量的恒等不变特性建模为最小二乘问题. 基于所提出的重构模型, 本文设计了基于谱移位的重构算法和基于残差谱移位的重构算法. 相比于其他重构算法, 两种新算法提升了迭代效率和重构精度. 此外, 本文算法还适用于分段带限图信号的重构问题, 并且具有良好的迭代效率和重构精度.通过实验仿真表明, 相比于目前其他的带限图信号重构算法, 新算法的迭代效率提升约70%和重构精度提升约60%.     

一种新的灰色预测模型及其建模机理

为了提高灰色模型的预测精度,并拓展其应用范围,针对具有近似非齐次指数律特征的数据序列,构建了一种新的灰色预测模型ＮＧＭ（１,１,k）．通过最小二乘法求出了新灰色模型参数的计算公式,以微分方程作为演绎推理工具,得到了该模型的时间响应序列函数,并对其建模精度进行了理论和实验分析．研究结果表明了所提出的灰色模型的有效性和适用性．     

一种OFDM系统定时偏移估计新算法

提出一种利用时域训练序列的定时估计算法。在分析Schimdl算法和Minn算法的缺点的同时,提出了算法,并在两种信道中进行了仿真,仿真结果表明算法在性能上有所提高。     

基于"新颖度"的关联挖掘算法

关联挖掘的目的是从大量数据中发现对用户有用、新颖、重要的关联规则.传统的关联挖掘算法会产生大量对用户而言显而易见的平凡规则,使那些真正对用户有用的新颖规则被淹没,而一些针对新颖性的改进算法往往又存在先验知识表达复杂且工作量极大的问题.在本文中,我们运用简单的分类树,引入"新颖度"的概念,对Apriori算法进行改进,得到了基于"新颖度"的关联挖掘算法,此算法既充分考虑了挖掘过程中得新颖性问题,又克服了先验知识表达过于复杂的困难.     

基于深度学习的新型视频分析系统综述

摄像设备在生活中的普及,使得视频数据快速增长,这些数据中蕴含丰富的信息.早期,研究人员基于传统的计算机视觉技术开发视频分析系统,用于提取并分析视频数据.近年来,深度学习技术在人脸识别等领域取得了突破性进展,基于深度学习的新型视频分析系统不断涌现.从应用、技术、系统等角度,综述了新型视频分析系统的研究进展.首先,回顾了视频分析系统的发展历史,指出了新型视频分析系统与传统视频分析系统的区别;其次,分析了新型视频分析系统在计算和存储两方面所面临的挑战,从视频数据的组织分布和视频分析的应用需求两方面探讨了新型视频分析系统的影响因素;再次,将新型视频分析系统划分为针对计算优化的系统和针对存储优化的系统两大类,选取其中典型的代表并介绍其核心设计理念;最后,从多个维度对比和分析了新型视频分析系统,指出了这些系统当前存在的问题,并据此展望了新型视频分析系统未来的研究和发展方向.     

基于聚磁技术的新型脉冲涡流传感器设计

对裂纹缺陷长度和深度进行定量是脉冲涡流无损检测的一项重要内容.提出了一种基于聚磁技术的新型脉冲涡流传感器.采用大型电磁仿真软件ANSYS建立了传统脉冲涡流传感器和新型脉冲涡流传感器的仿真模型,然后对比分析了两者对裂纹缺陷长度定量的结果,仿真分析表明:新型传感器可以实现对裂纹长度的准确定量,同时还能够对裂纹深度进行定量....     

Global Asymptotic Periodic Synchronization for Delayed Complex-Valued BAM Neural Networks via Vector-Valued Inequality Techniques

In this paper, we are concerned with a class of delayed complex-valued BAM neural networks. In stead of using the priori estimate method of periodic solutions, by means of combining Mawhin’s continuation theorem of coincidence degree theory with novel LMI method and some analysis techniques, a novel LMI-based sufficient condition is obtained for the existence of periodic solutions of the delayed complex-valued BAM neural networks. Then by using novel LMI method, a novel sufficient condition on global asymptotic periodic synchronization of above complex-valued BAM neural networks is established.     

Analysis and application of a novel fast algorithm for 2-D ARMA model parameter estimation

In this paper, we analyze a novel algorithm for 2-D ARMA model parameter estimation in the presence of noise and then develop a fast and efficient blind image restoration algorithm. We show that the novel algorithm can minimize a quadratic convex optimization problem and has a lower computational complexity than the conventional algorithms. As a result, the novel algorithm involves no convergence and local minimum issue. Moreover, the proposed blind image restoration algorithm can overcome the local minimization problem. Computed results confirm that the novel algorithm can more quickly obtain more accurate estimates than the conventional algorithms in the presence of noise.     

小说人物性格的文学智能分析: 以《平凡的世界》为例

以往小说人物心理分析主要是对人物性格的定性分析,易受研究者个人主观经验影响;而相比于描述繁杂的性格而言,更加稳定系统的人格能够更好地描述并传达小说人物心理。该文采用基于数据挖掘的文学智能分析方法,通过中文心理分析系统对《平凡的世界》人物对话进行处理,得到人物的大五人格预测分数;进而,考察文艺学文献、小说剧情对预测分数的验证情况,以确定这种方法的有效性。结果表明: 年轻的孙少平和田晓霞开放性相对较强,而年长的孙少安和田润叶外向性较强;此外,孙少平和田润叶尽责性较强,孙少安和田晓霞宜人性较好,孙少安和孙少平情绪性较高。上述预测结果得到文献、剧情的支持,这说明文学智能分析小说人物人格是有效的,它为小说人物心理分析开辟出一条客观、体系化且智能化的道路。     

Does stress training generalize to novel settings?

Many high-stress task environments are complex and dynamic, and it is often difficult during training to anticipate the exact conditions that may be encountered in these settings. We conducted an empirical study to examine whether the positive effects of stress training that addressed one specific type of stressor and task would remain when trainees performed under a novel stressor or performed a novel task. Participants performed a laboratory task under stress conditions. Measures of task performance and self-reported stress were obtained at three performance trials: (a) prior to stress training, (b) after a stress training intervention targeted to that specific task environment, and (c) under novel stressor/task conditions. Results indicated that the beneficial effects of stress training were retained when participants performed under a novel stressor and performed a novel task. We discuss the implications of this study with regard to their application in the design of stress training and the transfer of learning to complex, dynamic task environments.     

基于双自编码器和Transformer网络的异常检测方法

基于自编码器（AE）的异常检测方法利用重构误差判断待测样本是正常数据还是异常数据。然而,上述方法在正常数据与异常数据上产生的重构误差非常接近,导致部分异常数据很容易被错分为正常数据。为解决上述问题,提出一种由两个并行的AE和一个Transformer网络组成的异常检测方法——DATN-ND。首先,Transformer网络利用输入样本的瓶颈特征生成伪异常数据的瓶颈特征,从而在训练集中增加异常数据信息;其次,双AE将带有异常数据信息的瓶颈特征尽可能地重构为正常数据,增加异常数据与正常数据的重构误差差别。与记忆增强自编码器（MemAE）相比,DATN-ND在MNIST、Fashion-MNIST、CIFAR-10数据集上ROC曲线下面积（AUC）分别提升6.8、12.0和2.5个百分点。实验结果表明,DATN-ND能够有效扩大正常数据和异常数据在重构误差上的差别。     

Nominal state relationship with support vector machines and its sliding mode control application

This paper derives a novel nominal state relationship (NSR) by using support vector machines (SVM). By using the NSR, a novel sliding surface which has the dynamics of a nominal system controlled by pole placement is proposed. The sliding mode control (SMC) with this novel sliding surface gives its robustness to pole placement control systems with uncertainties.     

Dynamics analysis of a new simple chaotic attractor

In this paper, a novel simple chaotic system is discussed. Some basic dynamical properties, such as Lyapunov exponents, Poincaré mapping, fractal dimension, bifurcation diagram, continuous spectrum and chaotic dynamical behaviors of the novel chaotic system are studied, either numerically or analytically. The obtained results show clearly that the system discussed in this Letter is a novel chaotic system and deserves a further detailed investigation.     

A novel set of Wiener indices

According to the definition of molecular connectivity and the definition of the Wiener index, a novel set of Wiener indices (mV(t)) were defined, named novel Wiener index. The potential usefulness of the novel Wiener index in QSAR/QSPR is evaluated by its correlation with a number of C(3)-C(8) alkanes and by a favorable comparison with models based on molecular connectivity index and Wiener index.     

新型复合结构电热微驱动器的研制

在弯曲梁电热微驱动器原理的基础上,通过引入热膨胀系数更大的聚合物,提出了一种新型复合结构电热微驱动器。采用了ANSYS有限元分析软件,对其基本结构形式进行了系统的仿真分析,得到了一组初步优化的器件结构参数。仿真结果表明:新型电热微驱动器有更低的能耗和更大的位移。采用非硅表面微加工技术,成功制备了新型复合结构电热微驱动器。