智能设备的原理与开发技术

智能设备是基于计算机网络设备的自助应用系统中的重要组成部分，对改善服务质量、提升企业形象、拓展业务有重要意义。结合开发经历，总结讨论了智能设备的现状、组成原理、应用模式、软件开发、安全及整合等问题。     

An extended Lagrangian support vector machine for classifications

Lagrangian support vector machine (LSVM) cannot solve large problems for nonlinear kernel classifiers. In order to extend the LSVM to solve very large problems, an extended Lagrangian support vector machine (ELSVM) for classifications based on LSVM and SVMlight is presented in this paper. Our idea for the ELSVM is to divide a large quadratic programming problem into a series of subproblems with small size and to solve them via LSVM. Since the LSVM can solve small and medium problems for nonlinear kernel classifiers, the proposed ELSVM can be used to handle large problems very efficiently. Numerical experiments on different types of problems are performed to demonstrate the high efficiency of the ELSVM.     

Induction of hybrid decision tree based on post-discretization strategy

By redefining test selection measure, we propose in this paper a new algorithm, Flexible NBTree, which induces a hybrid of decision tree and Naive Bayes. Flexible NBTree mitigates the negative effect of information loss on test selection by applying postdiscretization strategy: at each internal node in the tree, we first select the test which is the most useful for improving classification accuracy, then apply discretization of continuous tests. The finial decision tree nodes contain univariate splits as regular decision trees, but the leaves contain Naive Bayesian classifiers. To evaluate the performance of Flexible NBTree, we compare it with NBTree and C4.5, both applying pre-discretization of continuous attributes. Experimental results on a variety of natural domains indicate that the classification accuracy of Flexible NBTree is substantially improved.     

Design and Research of Class 900 t Railway Box-girder Bridge Erecting Machine

In this article,the current railway box-girder bridge erecting machines at bome and abroad are briefly introduced and analyzed,the research ＆ design situation of class 900t rao wau box-girder bridge erecting machines is described,and also the principle for determin-ing the overall plan and a series of issues much concerning the design of key components of class 900t raihway box-girder bridge erecting machines are deseribed.     

Research and Application of All-Position Automatic Welding Machine

The all-position automatic welding machine system is the special welding system for pipeline girth automatic welding on site, which has been widely used in the long distance pipeline construction projects due to the advantages of automatic control for welding parameters at all position, moving speed of bugs and operating. This automatic pipeline welding system has been successfully used in several main pipeline projects in China, and has been approved by the constructors with the benefits of higher quality passing rate, higher welding efficiency and lower labor intensity.     

改进的核极限学习机定位算法

针对神经网络无线定位方法,存在训练耗时长,定位结果易受噪声干扰的问题,提出了一种改进的核极限学习机无线定位算法。采取在同一位置进行多次测量的方法得到训练数据;把同一位置测得的数据划分为一个样本子空间并提取样本子空间的特征,以样本子空间的特征代替原来的训练数据;利用矩阵近似及矩阵扩展的相关理论改进核极限学习机算法;将处理过的训练数据利用改进的核极限学习机进行训练,得到定位预测模型。仿真结果表明,在相同数据集下,改进的核极限学习机训练用时短、定位速度快;在相同噪声干扰情况下,此算法定位预测误差小。经验证,该算法不但能提高网络的训练速度、定位速度,还能有效地降低噪声的干扰,提高定位精度。     

基于伪数据的机器翻译质量估计模型的训练

为向基于深度学习的机器翻译质量估计模型提供高效的训练数据, 提出面向目标数据集的伪数据构造方法, 采用基于伪数据预训练与模型精调相结合的两阶段模型训练方法对模型进行训练, 并针对不同伪数据规模设计实验。结果表明, 在构造得到的伪数据下, 利用两阶段训练方法训练得到的机器翻译质量估计模型给出的得分与人工评分的相关性有显著的提升。     

融合倾斜仪数据的盾构姿态严密解算模型

推导了倾斜仪与盾构的严密标定模型;将倾斜仪采集的双轴角度数据视为平面点云,推导了倾斜仪角度数据的稳健估计方法;在前述模型方法的基础上,利用最小二乘配置方法建立了倾斜仪与棱镜数据融合的严密联合解算模型.最后,通过试验对上述模型方法进行了分析验证.结果证明,提出的盾构姿态解算模型正确可行.     

内蒙古草原核心保护功能区的划定与分析——基于支持向量机方法

草原生态保护建设关系全面建设小康社会奋斗目标的实现,而草原保护红线又是草原生态保护建设最基本的有效途径。从介绍内蒙古草原三大生态功能和“一线三区”入手,选取内蒙古三个典型草原区域的数据作为训练样本,经过一系列的数据预处理和以支持向量机为主要方法预测出三个草原核心保护功能区的分布结果。结果证明从国家文件理论指导到运用支持向量机方法实际操作后可以取得理想的效果,但是还有待基于这三个典型草原研究结果推广至全内蒙古。     

基于文本分类的农业种植信息集成推荐方法研究

目前网络上存在着海量的农业信息,但是对于广大农民来说信息得不到有效的利用,迫切需要对信息进行集成推荐.针对网络上的农业种植方面的文本信息进行了深入研究,该系统首先利用爬虫技术自动地爬取海量农业种植信息,经清洗整理后构建数据集语料库.其次利用机器学习中KNN方法找到每个样本的k近邻对文章进行聚类,通过TF-IDF方法提取出关键词并构造词频矩阵,然后从文本中构建特征向量,进而对相似文档进行分类,最后将加权值经排序后的结果推荐给用户.该系统实现了对农业文本进行准确的自动分类以及自动提取出文章摘要,并对相似文章进行推荐展示的效果.