统计数据图形化方法及其应用

发布时间：2018-06-16 16:39

本文选题：数据可视化 + 数据图形化　；参考：《重庆大学》2015年硕士论文

【摘要】：大数据比较公认的概念是4V特点数据——变化速度快(Velocity)、数据量大(Volume)、价值密度低(Value)、数据类型多样化(Variety),这意味着快速且低成本的处理、巨大的数据量、多样化的来源。大数据给人们带来了机遇和挑战,但是数据给人的直观感受总是冰冷枯燥,让人望而生畏,百思不得其解。为了使数据生动有趣,让数据使用者一目了然,豁然开朗,需要我们采用一些特别的方式展示数据,来解释、分析及应用数据,而且使得其能有效传播,这就是数据可视化技术。数据可视化中运用图形化,主要目的在于对复杂数据信息的更直观解释。信息图形化主要步骤如下:获取、解析、过滤、挖掘、展示、总结。这个过程需要研究人员对多种专业技能熟练掌握。从获取数据开始,研究人员首先需要解决数据易读的问题。数据挖掘和展示时,则需要研究人员挖掘数据的本质特征、模式等。可知,描述统计是表述数据内在含义的一个大类。数据作为信息存在的重要形式,在人们的工作生活中所起到的作用越来越大,而计算机技术的发展则使人们越来越依赖各种计算机化的数据。一方面计算机的应用领域几乎遍及各行各业(科研、工程、管理、医学、电子商务、金融等),另一方面,计算机处理的数据量也呈几何级数增长,不仅数据采集能力和手段日趋多元化,存储设备技术也发展迅猛,为人们在大数据时代实现海量数据的充分应用创造了条件。面对大量且庞杂的数据信息,如何从中提取出有价值且便于观察的信息是目前最迫切的问题。显然要解决上诉的问题,仅仅采用统计数据分析方法容易引起数据的不易理解,因此,笔者将统计数据分析技术和数据图形化方法结合起来,力求实现数据分析的可读性,易读性。从统计数据分析方法和图形化技术的统计学基础入手,研究统计数据图形化的表示方法,并在此基础上进行统计数据分析及图形化技术在具体经济数据中的应用。本文的研究内容分为两部分:①常见数据图形化方法及其应用,这部分主要整理了常用的统计数据图形方法,如条形图、直方图、散点图、饼图、线图、面积堆积图等。介绍了这些图形的做法,功能及应用。在此基础上对图形的着色、大小、形状等图形属性以及标度注解各方面在计算机上进行综合运用。②多维数据图形化方法及应用,收集整理多维股票数据,在此基础上进行相关矩阵图、轮廓图、星图、脸谱图以及谱系图的图形展示,并对图形展示结果进行分析。
[Abstract]:The generally accepted concept of big data is 4V characteristic data--fast changing speed, large volume of data, low value density and diversified data types, which means fast and low cost processing, huge amount of data, and diversified sources. Big data brings people opportunities and challenges, but the intuitive feeling of data is always cold, boring and daunting. In order to make the data lively and interesting, to make the data user clear and clear, we need to show the data in some special way, to interpret, analyze and apply the data, and to make it spread effectively. This is data visualization technology. The use of graphics in data visualization is mainly aimed at the more intuitive interpretation of complex data information. The main steps of graphic information are as follows: access, analysis, filtering, mining, display, summary. This process requires researchers to be proficient in a variety of professional skills. Starting with getting data, researchers first need to solve the problem of readability. When mining and displaying data, researchers need to mine the essential characteristics and patterns of data. It can be seen that descriptive statistics is a large class that describes the intrinsic meaning of data. As an important form of information, data plays a more and more important role in people's work and life, and the development of computer technology makes people rely more and more on computerized data. On the one hand, the application of computers is almost universal in all walks of life (scientific research, engineering, management, medicine, electronic commerce, finance, etc.). On the other hand, the amount of data processed by computers also increases in a geometric series. Not only the ability and means of data acquisition are becoming more and more diverse, but also the technology of storage devices is developing rapidly, which creates conditions for people to realize the full application of massive data in the era of big data. In the face of a large amount of data information, how to extract valuable and easy to observe information is the most urgent problem. Obviously, to solve the problem of appeal, it is easy to understand the data by using the statistical data analysis method only. Therefore, the author combines the statistical data analysis technology with the data graphic method to realize the readability of the data analysis. Readability. Based on the statistical basis of statistical data analysis method and graphic technique, this paper studies the graphical representation method of statistical data, and on this basis carries on the statistical data analysis and the application of graphic technology in the concrete economic data. The research content of this paper is divided into two parts: 1 common data graphic method and its application. This part mainly arranges the commonly used statistical data graph method, such as bar chart, histogram, scattered plot, pie chart, graph, area stacking diagram and so on. The methods, functions and applications of these graphics are introduced. On the basis of this, the graphics attributes, size, shape, and scale annotation of the graphics are comprehensively applied to the computer with .2 dimensional data graphical method and application, and the multidimensional stock data are collected and sorted out. On this basis, the graph display of correlation matrix map, contour map, star map, face map and pedigree diagram is carried out, and the result of graph display is analyzed.
【学位授予单位】：重庆大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：C81

【相似文献】