【作者】 丁浩；

【导师】 郭金旭；

【作者基本信息】 武汉理工大学 ， 通信与信息系统， 2008， 硕士

【摘要】 数据挖掘可以称为数据库中的知识发现,它是从大量数据中发现并提取隐藏在其中的可信的、新颖的、有效的并能被人理解的模式的高级处理过程。数据挖掘解决了传统分析方法的不足,并能够对大规模数据的进行分析处理。数据挖掘从大量数据中提取出隐藏在数据之后的有用的信息,为人们的正确决策提供了很大的帮助。证券业作为我国较早应用信息技术的行业,已经建立了较完善的事务处理系统。多年的应用也使得证券行业积累了大量的数据,其中隐含着大量有价值的信息。如何利用这些数据,深层次地挖掘信息,为证券投资的决策服务,已成为证券行业研究的当务之急。(1)本文介绍了数据挖掘技术的发展现状和基本原理,讨论了数据挖掘技术在证券行业中的应用意义。(2)提出了一个集数据采集,转换和挖掘于一体的数据挖掘系统,并对其主要功能进行了比较详细的介绍。(3)聚类分析在证券投资方面的研究有很大的发掘空间。本文将聚类分析方法引入到证券投资分析中,对股票的行业因素、公司因素、收益性、成长性等基本层面进行考察,建立了较为全面的综合评价指标体系,衡量样本股票的“相似程度”。然后通过聚类分析模型来确定投资范围和投资价值。结果表明该方法能帮助投资者准确地了解和把握股票的总体特性,预测股票的发展潜力,并通过类的总体价格水平来预测股票价格的变动趋势,选择有利的投资时机。实证研究表明该方法对指导证券投资具有有效性和实用性。(4)数据挖掘的结果进行了分析,提出了下一步的具体工作。更多还原

【Abstract】 The data mining is also called the knowledge discovery in database, it discovers from large quaintly of data and find authentic, novel and effective model that can be comprehended by people. The data mining finds useful information from data of large quantity to support for reasonable decision-making. Mining data from large amounts of data to extract hidden in the data after the useful information for people to the correct policy decision to provide a lot of help.China’s securities industry as an earlier application of information technology industry has been a better transaction processing system. Years of application also makes the securities industry has accumulated large amounts of data, which implied a large number of valuable information. How to make use of these data, deep-level mining, information, decision-making for investment in securities services, and the securities industry research has become a pressing task.Firstly, this paper introduced data mining technology and the development of basic principles, discussed the data mining technology in the securities industry of significance.Secondly, the paper proposed a set of data collection, conversion and mining data collected in one of the mining systems, and its main function of a more detailed briefing.Thirdly, the primary methods of stock investment analysis are fundamental analysis, technical analysis and portfolio investment analysis. Based on analyzing the applicability and limitation of these methods, the paper is applying the cluster analysis in the stock investment, and pointing out it is a feasible and effective way to give advices to stock investment. Cluster analysis firstly looks into the industry factor, company factor, profitability and pullulating of each stock, so people can discover the investment value through stock’ immanent value which is decided by their fundamental characters. Together the similar stock, a synthetic evaluation index system is erected to measure the similarity of stock ail-roundly and objectively. Cluster analysis can help investors hold the whole character of stocks, make certain of investing scale and suppose the possible variation of stock price. So it will be easier for investors to choose investing occasion.Finally, the paper Mining data acquisition system for the overall debugging, and the results are analyzed, the next step of the concrete work.更多还原