计算机与现代化

• 人工智能 • 上一篇    下一篇

基于C4.5决策树的股票数据挖掘

  

  1. (四川大学计算机学院,四川 成都 610065)
  • 收稿日期:2015-04-03 出版日期:2015-10-10 发布日期:2015-10-10
  • 作者简介:王领(1990-),男,四川巴中人,四川大学计算机学院硕士研究生,研究方向:数据挖掘,机器学习; 胡杨(1987-),男,硕士研究生,研究方向:计算机网络与信息系统。

Stock Data Mining Based on C4.5 Decision Tree

  1. (College of Computer Science, Sichuan University, Chengdu 610065, China)
  • Received:2015-04-03 Online:2015-10-10 Published:2015-10-10

摘要: 由于目前利用数据挖掘算法对股票分析和预测存在数据量及技术指标等方面的问题,本文基于对股市数据的分析,适当选取某些指标作为决策属性,利用C4.5决策树分类算法进行分类预测。主要对股票技术指标进行介绍和优化,对C4.5算法的效率进行改进。改进后的算法结合优化的技术指标不仅能够提高数据挖掘的执行效率,同时也能在股票预测方面得到更高的收益。

关键词: 数据挖掘, 决策树, 技术指标, C4.5, 股票预测

Abstract: Using data mining algorithms to analyze and forecast the stock still has problems in technical indicators and quantity of data. Based on the analysis of stock market data, this paper selected certain indicators as decision attribute, and used C4.5 decision tree to classify and forecast the stock. This article mainly optimized the indicators of stock, and improved the efficiency of C4.5 algorithm. Optimized algorithm combining with improved indicators not only enhances the efficiency of data mining, also gets better returns in stock forecasting.

Key words: data mining, decision tree, indicators, C4.5, stock forecasting

中图分类号: