一种海量数据的分析技术 ——符号数据分析及应用
A New Data Mining Method Based on Huge Data and Its Application
-
摘要: 传统的统计分析方法在处理海量数据方面存在很大的局限性。为了解决这一问题,符号数据分析(symbolic data analysis, SDA)方法被提出并得到了迅速的发展。SDA方法对传统的数据概念做了本质性的扩张,运用‘数据打包’的理念,对海量的原始数据在不破坏其原有内在逻辑关系的前提下,可以进行变量和样本点空间的双重降维,并将传统的统计分析技术扩展到符号数据分析体系中。符号数据的方法体系是知识发现和数据管理领域的最新研究方向之一,目前在国内鲜有相关的研究资料。文章详细阐述了符号数据因素分析技术的原理和概念,并以中国股票市场为案例研究背景,结果表明,SDA因素分析技术对综合简化大规模多维数据系统是十分有效的。Abstract: A kind of advanced complicated data analysis technology ——Symbolic Data Analysis (SDA)has been used in this paper with style indices of Chinese international trust and investment company (CITIC) of 2000 as research objects. The research result indicates that the style characteristic is distinct in Chinese stock market, which, as well as the factor of scale, has influenced Chinese stock market dramatically. The computational analysis has showed that, by the method of Symbolic Data Analysis, the research result is very consistent with the realistic characteristics of Chinese stock market, which proves that it is effectual to simplify the multidimensional dynamic data system.
下载: