数据挖掘

By analyzing transplant and reuse of data mining platform,an opening development platform for data mining was designed and set up,which was based on CWM and used PMML to describe data mining Model.

在对现有数据挖掘平台移植和复用等方面的需求分析的基础上，提出并建立了以公共数据元模型为基础，用预言模型标记语言描述挖掘模型的数据挖掘开放式开发平台，详细论述了该平台逻辑框架、组成模块的功能接口以及平台的执行过程等。

The article first discusses the related concepts and basic method of data mining. The task of data mining is to find mode from database. The mode can be many , we can dispart two according to function: predictive mode and descriptive mode.

文章首先论述了数据挖掘的相关概念以及数据挖掘的基本方法，数据挖掘的任务是从数据集中发现模式，模式可以有很多种，按功能可分为两大类：预测型模式和描述型模式。

Multi-dimension data visualization technology has been widely used in the world, and many common multi-dimension data visualization methods have been proposed, such as technology based on geometry, technology oriented ironic, technology based on layer, dense pels technology, and so on.

在本文中，我们将首先对数据挖掘技术、多维数据可视化技术、可视化数据挖掘技术进行介绍，然后我们会通过实例来介绍多维数据可视化技术在数据挖掘中的应用。

This paper is based on data mining technology,use weka as a library data mining software tool,use weka's j48 tree algorithm and data association analysis library data.mining useful data which user needs from the mass library data and get reasonable results.ultimately to improve work efficiency and scientific management.

该文以数据挖掘技术为基础，利用weka软件作为图书馆数据挖掘工具，通过weka里的j48树算法和数据关联等算法，对图书馆的馆藏数据进行相应的分析，从海量数据中挖掘出用户需要的有用数据，并得到合理的统计结果。最终达到提高工作效率，能够科学管理的目的。

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。

The major achievement of this paper is: Based on characteristics of the traffic data distribution, execute pattern recognition operations on traffic condition on two dimensions by clustering, then use BP neural network to describe and forecast traffic flow aiming at each pattern. Making use of classic flow-occupancy inverse "V" model, implement polynomial fitting using least-squares algorithm and statistics method on flow curves to detect outliers which are proved to be not accord with practice through the actual implement, then use the moving average model to recorrect the outliers and absent. Make correlation analysis on muti-direction flow queues of the intersection and ones of upriver intersections, choose flow queue with high correlation as assistant one to improve the error tolerance of the prediction system, at the same time we can use the method to give an estimation of flow in intersection with out sensors. We design and implement an SOA(Service-Oriented Architecture)-based UTDD(urban traffic data mining development) with high expansibility and performance, which implement unified management and call of the data-mining application though defining a XML-based description of data-mining process and a common interface to call data-mining process, finally we build traffic flow prediction application model on UTDD.

根据交通流量数据分布的特征，提出基于k-means的二次聚类方法，对交通流量在流量大小和时间上进行模式划分，进而对各个交通流模式进行基于BP神经网络的描述和预测，从而提高模型对流量预测的精度； 2）根据流量/时间占有率倒&V&字形曲线分布模型，提出基于最小二乘法的三次多项式曲线拟合和统计方法的异常检测方法，实际应用表明该方法能够有效识别异常数据，然后根据移动平均算法对异常数据进行修正； 3）基于序列相关性分析，分别对预测方向的交通流量数据序列、上游路口相关序列以及预测路口其它各个方向上的交通流量序列进行分析，选择相似性流量序列，作为辅助序列提供其他没有检测器路口的流量估计； 4）设计和实现了基于SOA（Service-Oriented Achitecture）的高性能、可扩展的智能交通数据挖掘系统UTDD，该系统通过定义基于XML的数据挖掘过程描述和通用的过程模型接口，实现数据挖掘应用的统一管理和调用，最后在UTDD上建立了基于路口流量预测的应用模型。

The dissertation focused on the technologies, algorithm and application of WBSDM, and the main studies and new contributions are described as follows: 1. Technologies of spatial data mining and web-based data mining are summarized and then the concept of WBSDM is put forward and its main functions and meaning of research are disserted in detail. 2. The framework of web-based spatial data mining is presented. And its background, definition, features, functions, system structure, implementation strategy, key technologies and so on are illustrated systematically.

本项研究旨在总结空间数据挖掘技术和基于Web的数据挖掘技术及WBSDM的应用前景；系统地提出基于Web的空间数据挖掘框架；设计基于XML/J2EE的WBSDM平台模型；将MAS(Multi-AgentSystem)技术引入WBSDM并提出应用策略；研究基于粗糙集的空间知识库刻画模型，提出并实现高维数据的特征提取和数据浓缩算法；提出并实现基于知识库的知识发现模型；构造一个WBSDM的原型实验系统，对本文的相关研究进行验证。

The paper summarizes theories and practices concerning data mining at the beginning. According to actual conditions of data forecasting, the paper particularizes two basic tasks in data mining, namely, exploring data analysis and forecasting modeling for classification, and employs them in load forecasting practically. Then a special short-term load-forecasting arithmetic is forwarded based on the former work, which can efficiently take into account weather effects on the load, and hence improve the accuracy of load forecasting.

本文概述了数据挖掘技术的有关内容；并针对负荷预测的实际情况，详细介绍了数据挖掘中的两个基本任务：探索性数据分析和用于分类的预测建模，并将其运用于负荷预测之中；根据所得结果提出了一种基于数据挖掘的短期负荷预测算法，该算法能够有效的考虑气象因素对负荷的影响，从而提高了负荷预测的精度。

This article not to the data mining theory and the modeling method and so on make excessively many elaborations, also does not have the logarithm to make the meticulous discussion according to the warehouse construction aspect, but is puts with emphasis on in the data mining model choice and the design, in overseas had the research in the foundation, in the union corporate business analysis actual need, proposed the AVON company's customer value, the customer retention and the customer thin classify three data minings model; And executed has implemented the questionnaire survey and the method collection research data to various exclusive agencies customer which unified to member's telephone investigation, has carried on the confirmation and the appraisal in SQL Server 2,000 Analysis Service to the model.

本文没有对的数据挖掘理论及建模方法等作过多的阐述，也没有对数据仓库建设方面做过细的探讨，而是将重点放在数据挖掘模型的选择与设计上，在国外的已有研究的基础上，结合公司业务分析上的实际需要，提出了雅芳公司的客户价值、客户保持和客户细分等三个数据挖掘模型；并施行了对各专卖店的顾客实行问卷调查和对会员的电话调查相结合的方法收集研究数据，在 SQL Server 2000 Analysis Service中对模型进行了验证与评价。

The method is realized via the following steps: first to apply Isomap or LLE to get the embeddings of the original data set in the low dimensional space; then to obtain support vectors, which are the most significant and intrinsic data for the final classification result, by using support vector machine on these low dimensional embedding data; subsequently to get support vectors in the original high dimensional space based on the corresponding labels of the obtained low dimensional support vectors; finally to apply support vector machine again on these high dimensional support vectors to gain the final classification discriminant function.

数据挖掘是数据库知识发现中最重要的步骤之一，其目标是从获取的数据中高效准确地挖掘出我们所需要的信息。在实际应用中，数据往往呈现海量、高维、非线性等特性，这些特性给数据挖掘带来了很多问题，例如海量特性导致的计算效率低下问题、高维特性带来的维数灾难问题和非线性特性引起的线性模型失效问题等。幸运的是，实际中高维数据的属性之间往往存在一定的规律性和相关性，即实际数据经常存在着外在与内在两个维数。

On the other hand, the more important thing is because the urban housing is a kind of heterogeneity products.

另一方面，更重要的是由于城市住房是一种异质性产品。

Climate histogram is the fall that collects place measure calm value, cent serves as cross axle for a few equal interval, the area that the frequency that the value appears according to place is accumulated and becomes will be determined inside each interval, discharge the graph that rise with post, also be called histogram.

气候直方图是将所收集的降水量测定值，分为几个相等的区间作为横轴，并将各区间内所测定值依所出现的次数累积而成的面积，用柱子排起来的图形，也叫做柱状图。