基于网络共识的股票价格行为数据挖掘_英文_.pdfVIP

心智与计算 71 心智与计算, Vol.4，No.1 (2010), 71-77 文章编号：MC–2010 - 09 收稿日期：2010-03-12 出版日期：2010-03-31 © 2007 MC– 厦门大学信息科学与技术学院 Stock Price Behavior Data Mining Based on Network Consensus LAI Xing-rui, ZHANG Dong-zhan, DUAN Jiang-jiao (Department of Computer Science, Xiamen University, Xiamen 361005, China) zdz@ Abstract: Stock price behavior data mining has aroused great concern in research of computer science, economy, machine learning and other fields. However, due to the great uncertainty and complexity in stock market, researches from different points of view are desired to be integrated to make full use of different aspects of knowledge. In this paper, we propose to combine stock price and network consensus and create a stock behavior data mining platform based on network consensus. Stock price data is continuously extracted from real time stock transaction systems, and saved to the database in the platform. Online discussion threads in community forums are captured and stored to the database, too. The architecture, data structure and analysis supporting components are designed to meet the requirements in utilizing the two aspects of knowledge to mine the association. Key words: data mining; network consensus; stock price 基于网络共识的股票价格行为数据挖掘 赖兴瑞, 张东站，段江娇 （厦门大学计算机科学系，福建厦门 361005 ） zdz@ 摘要：股票价格行为数据挖掘激发了计算机科学、机器学习及其他领域研究的广泛关注。然而，由 于股票价格本身的不确定性和股市的复杂性，急需从不同角度将各方面知识综合利用。本文提出了股票 价格充分融合网络共识的策略，构建了基于网络共识的股票价格行为数据挖掘平台。从交易系统实时采 集股票交易价格数据，同时从网络智能获取当前的讨论热点线索一并存储到平台数据库。围绕融合两方 面数据进行股票价格行为数据挖掘，完成了体系架构、数据结构和分析支持组件的设计，实现了股票价 格行为数据挖掘原型系统。 关键词：数据挖掘；网络共识；股票价格 中图分类