财经资讯的多维分类聚合技术在股市中的应用研究

发布时间：2018-04-13 01:28

本文选题：财经资讯 + 多维　；参考：《北京交通大学》2014年硕士论文

【摘要】：随着互联网技术的发展,资讯量迅速膨胀,海量的财经资讯里既包含着有价值的信息,又有大量无用和干扰信息。利用相关技术实时采集、处理和多维分类聚合资讯,提取有用信息来指导人们的投资决策是未来的发展趋势。本文采用理论和实证相结合的方法通过多维分类聚合技术,系统地研究了互联网财经资讯对中国股票市场价格走势的影响,通过对于股市价格走势的判断来指导帮助投资者在股市投资中做出正确的决策。文章提出了以下思想：资讯的异动和市场的人气会决定流动性好的股票的买卖机会。通过对资讯的分类聚合和数量化统计,研究了媒体关注度和股吧人气指标,可以发现市场关注热点,轻松跟踪热点板块和热门股票。首先,文章分析了互联网财经资讯的市场现状,阐述了目前财经资讯市场的背景、投资者获取资讯的方式,以及存在的缺陷。然后,详细介绍了财经资讯多维分类聚合的实现过程,对财经资讯的采集、分析、结构化和分类处理的过程都做了进一步的分析,并根据道氏理论、空间信息理论以及资讯文本分类的理论知识,利用互联网资讯多维分类聚合技术,对大盘、行业、板块和个股资讯进行了多角度解读；通过分析博弈市场下的资讯,给出相应的股价趋势判断,使得投资者有能力把握投资机会。最后,文章运用财经资讯多维分类聚合技术从股市中选取了三支股票,并研究了资讯趋势、人气热度和资讯多空信息对股票买卖决策的影响,根据多维分类聚合结果得到相应上市公司短期股价趋势的发展变化方向,对买入卖出点进行分析。实证分析表明,资讯传播趋势和媒体、股吧多空分析会发现并预测股价的反转和走势,通过对互联网财经资讯的多维分类聚合分析,可以指导投资者在股市中的投资行为。
[Abstract]:With the development of Internet technology, the amount of information expands rapidly. The massive financial information contains not only valuable information, but also a large amount of useless and interfering information.It is the trend of future development to collect, process and aggregate information in real time with relevant technology, and extract useful information to guide people's investment decision.In this paper, the influence of Internet financial information on the price trend of Chinese stock market is systematically studied by means of multi-dimensional classification and aggregation, which is based on the combination of theory and practice.To guide investors to make the right decision in stock market investment by judging the stock market price trend.This paper puts forward the following idea: the change of information and market sentiment will determine the trading opportunities of liquid stocks.By classifying and aggregating the information and quantifying statistics, the paper studies the media attention and the index of stock popularity. It can find the hot spot of market, track the hot spot and hot stock easily.First of all, the paper analyzes the current situation of the Internet financial information market, expounds the background of the current financial information market, the way for investors to obtain information, and the shortcomings.Then, the realization process of multi-dimensional classification aggregation of financial information is introduced in detail. The process of collecting, analyzing, structuring and classifying financial information is further analyzed, and according to Dow theory,The spatial information theory and the theoretical knowledge of information text classification, using the technology of multi-dimensional classification and aggregation of Internet information, have made a multi-angle interpretation of the market, industry, sector and stock information; through the analysis of the information in the game market,Give the corresponding stock price trend judgment, make the investor have the ability to grasp the investment opportunity.Finally, the paper selects three stocks from the stock market by using the multi-dimensional classification and aggregation technology of financial information, and studies the influence of information trends, popularity heat and information multi-space information on stock trading decisions.According to the multi-dimensional classification and aggregation results, the trends of short-term stock price trends of listed companies are obtained, and the buying and selling points are analyzed.The empirical analysis shows that the trend of information dissemination and the analysis of media and stock bar are able to find and predict the reversal and trend of stock price. Through the multi-dimensional clustering analysis of financial and financial information on the Internet, investors can be guided to invest in the stock market.
【学位授予单位】：北京交通大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：F832.51

【参考文献】