基于数据挖掘算法的金融数据采集与分析研究

发布时间：2018-07-03 18:43

本文选题：网络爬虫 + K-means算法　；参考：《天津大学》2016年硕士论文

【摘要】：股票市场作为金融市场中重要的一部分,在我国经济中起着非常重要的作用,成为经济的“晴雨表”,有效的预测股票走势在金融投资领域占有非常重要的地位。然而要分析股票市场,就需要及时的获取到股市的相关数据,但随着经济的不断发展,股票数据的规模越来越大,已经不是人力所能获取的。尽管网上有着很多数据集的提供网站,但是多数都需要付出大量的金钱,同时所得到的数据多为不全数据,无法及时得到最新的数据。近年来兴起的数据挖掘技术很好的提供了获取海量数据并从中分析出有价值、隐藏着的规律的技术。本文主要是运用网络爬虫技术对雅虎金融中的股票数据进行采集,利用贝叶斯推断的基本原理及其推广对数据进行分类和提取,从股票海量的数据中提取出有用的数据进行分析。同时建立一套数据提取系统,基于对网络爬虫的深入理解,对爬虫框架进行二次开发,实现快速抓取数据。随后初步探讨了利用K-means算法和改进的K-medoids对我国A股市场的交易数据进行聚类分析。
[Abstract]:As an important part of the financial market, the stock market plays a very important role in the economy of our country and becomes the "barometer" of the economy. The effective prediction of the stock trend occupies a very important position in the field of financial investment. However, in order to analyze the stock market, we need to get the relevant data of stock market in time, but with the development of economy, the scale of stock data is getting larger and larger, which can not be obtained by manpower. Although there are many web sites providing data sets on the Internet, most of them need to pay a lot of money. At the same time, most of the data they get are incomplete, so they can't get the latest data in time. In recent years, data mining technology has provided a good way to obtain massive data and analyze the valuable and hidden rules. This paper mainly uses the technology of web crawler to collect the stock data in Yahoo Finance, and classifies and extracts the data by using the basic principle of Bayesian inference and its generalization. Extract useful data from massive stock data for analysis. At the same time, a data extraction system is established. Based on the deep understanding of the web crawler, the crawler framework is redeveloped to realize fast data capture. Then it discusses the clustering analysis of the trading data of A-share market using K-means algorithm and improved K-medoids.
【学位授予单位】：天津大学
【学位级别】：硕士
【学位授予年份】：2016
【分类号】：TP311.13;TP274.2

【相似文献】