基于机器学习的股票量化交易信息管理系统  

A Stock Quantitative Trading Information Management System Based on Machine Learning

作　　者：熊峰 季振山[1,2] XIONG Feng;JI Zhenshan(Hefei Institutes of Physical Science,Chinese Academy of Sciences,Hefei 230031,China;University of Science and Technology of China,Chinese Academy of Sciences,Hefei 230026,China)

机构地区：[1]中国科学院,合肥物质科学研究院,安徽合肥230031 [2]中国科学院,中国科学技术大学,安徽合肥230026

出　　处：《微型电脑应用》2022年第12期132-135,共4页Microcomputer Applications

摘　　要：针对传统信息管理系统准确性和时效性较差的问题,提出一种基于机器学习的股票量化交易系统。系统硬件部分采用Web端服务器和100Base-T广域网的网络服务器,并搭载含有数据决策功能的信息控制机制;系统软件部分利用优化数据查询代码的方式强化信息储存和管理能力,并通过朴素贝叶斯算法结合模拟假设法,建立基于股票信息特征的数据样本和代表集合,通过包含所有非负整数的集合范围,获得最小方差错误概率,确定朴素贝叶斯特征分类规则,实现对样本数据有效分类和管理。实验结果表明,测试结果可靠性较强、检测率较高、误报率较小,能够实现信息的有效管理。Aiming at the problem of poor accuracy and timeliness of traditional information management system,a stock quantitative trading system based on machine learning is proposed.The hardware part of the system adopts Web server and 100Base-T WAN network server,and carries information control mechanism with data decision-making function.In the software part of the system,the ability of information storage and management is strengthened by optimizing the data query code,and the data samples and representative sets based on the stock information characteristics are established by combining the naive Bayesian algorithm with the simulation hypothesis method.Through the set range containing all non negative integers,the minimum variance error probability is obtained,and the naive Bayesian feature classification rules are determined to effectively classify and manage the sample data.The experimental results show that the reliability of the test results is strong,the detection rate is high,and the false alarm rate is small,which can realize the effective management of information.

关 键 词：股票量化交易 网络服务器 查询代码 朴素贝叶斯算法 干扰数据 

分 类 号：TP365.1[自动化与计算机技术—计算机系统结构]