分类数据

In GIS, data quality is one of the key aspects.The Recent studies of data quality in GIS most focus in the error of spatial data. The error study on attribute data is very weak. Most studies centralize in precision analysis of attribute data from remote sensing and introduction of new method. The studies of attribute information quality stored in RDB are very weak.

近年来，GIS数据质量的研究集中在位置数据的误差上，而对属性数据误差的研究较少，多年来的研究主要集中在遥感分类属性数据的精度分析和一些新方法的引入上，而关系数据库存储的属性信息质量方面的研究相对比较薄弱。

The main contents are listed as follows: Analyse and research of the data mining technique. the dissertation summarizes,analyses and studies the current status of the data mining technique in our native country and overseas widely and roundly research . The definiens and orientation of the data mining is reviewed in brief first.

对数据挖掘技术的国内外研究现状进行了广泛而全面地归纳、分析和研究，对数据挖掘的定义及定位进行了简要的回顾，在数据挖掘基本概念的基础上，对数据挖掘常使用的技术和研究的对象进行了详细地分类、归纳和总结。

It becomes a information exchange site,the exploration project,work area,research production,seismic datum,explainer datum,well datum can be organized into one system,which can share the research production in the exploration project;Through the datum expression of well decision,we can achieve the browsing of work area basemap,the seismic section"s remote display,the well position report"s display,the visual information of well"s wall.The system can lead to a right direction for those expert"s exploration decision .The system runs through all the process of well positon"s decision,deployment and management,which makes the exploration work becoming integrated.Now we can manage the well positon"s decision with computer,which improves work efficiency,condense whole exploration circle.

油田井位部署辅助决策系统以项目数据库的方式将分散在多个研究单位或个人手中的各类图档资料信息集中分类地保存起来，通过该系统使勘探项目、工区、研究成果、地震数据体、解释数据、井数据等有机地组织在一起，使勘探研究成果在勘探项目组内进行了充分共享；基于项目库，系统实现了井位部署决策所需的：工区底图浏览、地震剖面远程显示、井筒信息可视化，为论证专家提供了论证现场所需数据的展示；采用井位聚类分析和模糊数学综合评判决策方法，为井位论证决策提供了基本模型；利用井位部署流程子系统贯穿了井位论证、部署、管理业务的全过程中，使勘探业务有机的整合为一体，实现了井位论证工作的计算机管理。

Based on data cloud, which is measured from 3-coordinate measuring machine or so, an algorithm to search lopological structure from 3D unorganized points using Envelopment-box technology is proposed in this thesis. Based on this method, we searching neighbor points of sampling point. We also improve the Max-Min angle criteria to realize local triangulation, and then get the normal of sampling point from the triangulation. Using the normal and neighbor relationship, classification of scattered data points is realized according to the quadric surface.

本文以测量得到的曲面数据点点云为基础，提出了一种基于包围盒的自动寻找三维散乱数据点之间拓扑结构的方法，采用该方法寻找采样点的邻域结点，并对三角剖分中的典型优化准则—最小内角最大准则提出了改进，按照改进后的优化准则实现了采样点的局部三角划分，并进一步求解得到采样点的法向量，依据法向量及邻域拓扑关系在二次曲面的基础上实现了散乱数据曲面重构中的数据点的自动分类。

In order to exchange data between the disconnected databases in the condition of physical isolation,this paper classes that as bidirectional mirror exchange and unilateral superadd, and emphasizes on discussing the method of data consistency between mirror database , the types and resolvent of conflict caused by data update, and the method of relation consistency among tables in unilateral superadding.

为了实现在物理隔离条件下非连接数据库之间的数据交换，该文将数据交换分类为双向镜像和单向追加两类，重点论述了镜像数据库数据一致性的方法、数据更新冲突的类型和解决以及单向追加时表间链接关系一致性的维护方法。

In the light of field features in desertification areas, the data quality and band combination of different bands are assessed, the indicators, principles and methods for data assessment and band option are put forward to, and optional band combination for desertification monitoring is determined primarily. The data pre-treatment model for desertification monitoring is developed after studying the pre-treatment algorithm of restoration and rebuilding of hyperspectral resolution data. The spectral features and variation rules of different objects in experimental areas are understood and analyzed. The different classified methods are given and discussed on the basis of characteristics of desertification region. The remote sensing quantitative retrieval model for the appraisable factors of desertification monitoring are founded.

针对荒漠化地区的地物特征，对高光谱数据不同波段的数据质量、波段组合进行定性和定量的评价，提出用于数据评价、波段选择的指标、原则和方法；初步确定一个适用于荒漠化监测的优化波段组合，建立适合于荒漠化监测的数据预处理模式；初步分析、掌握研究区内各类地物的光谱特性及变异规律；针对荒漠化地区特点对多种分类方法进行探讨分析；建立荒漠化监测主要评价因子的定量反演模型。

Under the guidance of DPRIF integrating strategy, we set up an integreted SOFM-SVM model. Then we analyzed and optimized the model from the aspects of operating mechanism, data interface and function expandedness. The PCA method was introduced to reduce dimensionality and extract features, then to strengthen the clustering explanation; defined a CMI index, to ascertain the most effective or the best clustering number; A new Anti-NO algorithm was proposed to recognize and to filter the suspecious data in the sample; The medium result of SVM model was used to extract the borderline datas between two classified groups. This research compensated for the achievements of data recognition including pattern data, noises data and borderline data.

在DPRIF整合策略指导下构建了一个整合的SOFM-SVM模型，对该模型从运作机制、数据接口、功能扩展几个方面进行分析和改进：引入PCA方法进行数据降维和特征提取，用以加强聚类解释；结合统计聚类中的聚类误差概念定义一个聚类数有效性指标，利用SOFM算法中间结果进行指标求解，以筛选出有效或最佳聚类数；提出一种新的噪声识别算法用于对样本中的异常数据进行甄别和过滤；利用SVM模型的中间结果提取分类边界数据；进一步充实了包括模式类、噪声集和边界在内的模式识别成果。

Based on the basic concepts of data mining , this dissertation compares and analyses the differences of data mining and other methods such as KDD and OLAP , classifies and summarizes the objects of data mining , the findable patterns and the common techniques in detail .

在介绍数据挖掘基本概念的基础上，对数据挖掘与传统分析方法，数据库中的知识发现和联机分析处理做了深入地分析和比较，对数据挖掘的对象，可发现的模式进行了详细地分类，归纳和总结，对数据挖掘常使用的技术做了介绍和分析。

This paper first introduces the basic principle and concept of data fusion, summarizes levels and model structure of data fusion, summarizes common algorithms and sorts them, separately expounds data fusion algorithms, neural network and evidence theory, and lucubrates the structure design of neural network. And then, the advantages and disadvantages of neural network and the evidence theory are discussed. In order to colligate the two algorithms strongpoint and increase the fault diagnosis accuracy, fault diagnosis method using data fusion integrated Ann and evidence theory is presented and diagnosis approach of this method is expatiated on.

本文首先对数据融合的基本原理、基本概念进行了阐述，总结了数据融合的层次和模型结构形式，并对常用的融合算法进行了总结分类；分别对数据融合算法—神经网络方法和证据理论方法—进行了详细的阐述，同时对神经网络的结构设计进行了深入研究；然后对神经网络和证据理论方法优缺点进行了分析，在此基础上为了实现两种算法优势互补，提高诊断的准确率，提出了神经网络和证据理论集成的数据融合故障诊断法，并详细阐述了这种方法的诊断原理。

Formly, HDR casts classification problems and regression problems into a unified regression problem. Clustering is performed in both output space and input space at each internal node, termed "doubly clustered" and discriminants in the input space are automatically derived from the clusters in the input space. A hierarchical probability distribution model is applied to the resulting discriminating subspace at each internal node.

在图像的分类和检索中，以前的层次判别回归方法把分类和回归统一成回归问题，在每个内部节点进行输入和输出空间的双聚类，输入空间的分类子空间可以自动从聚类中获得，再用层次概率分布模型计算判别子空间，这样就构成了输入空间由粗到精的概率分布，可以准确而且快速地实现高维数据的检索。

Neither the killing of Mr Zarqawi nor any breakthrough on the political front will stop the insurgency and the fratricidal murders in their tracks.

在对危险的南部地区访问时，他斥责什叶派民兵领导人对中央集权的挑衅行为。

In fact,I've got him on the satellite mobile right now.

实际上 我们已接通卫星可视电话了