基于PSO优化双子支持向量机的电商经济预测研究  

Research on E-commerce Economic Forecasting Based on PSO Optimized Twin Support Vector Machine

在线阅读下载全文

作  者:巢瑞云[1] 刘源 CHAO Rui-yun;LIU Yuan(School of Economics and Management,Guangzhou NanYang Polytechnic College,Guangdong,Guangzhou 510900;School of Intelligent Medicine and Biotechnology Guilin Medical University,Guangxi,Guilin 541199)

机构地区:[1]广州南洋理工职业学院经济管理学院,广东广州510900 [2]桂林医学院智能医学与生物技术学院,广西桂林541199

出  处:《贵阳学院学报(自然科学版)》2024年第2期11-15,42,共6页Journal of Guiyang University:Natural Sciences

基  金:国家自然科学基金(项目编号:61474032);桂林医学院博士启动基金(项目编号:31304019011);2022年广东省社科规划项目“数商兴农”工程下广东省农村电子商务竞争力评价研究(项目编号:GD22XYJ28)。

摘  要:为提高电商经济预测性能,解决因时序复杂、特征量多及用户需求复杂等带来的预测精度偏低的问题,采用双子支持向量机进行电商经济预测。首先,获取电商经济数据特征,接着构建双子支持向量机(TWSVM)的电商经济预测模型,提取TWSVM的正则因子等参数,随机初始化多组参数,并构建多个粒子。然后借助粒子群优化(PSO)算法搜索最优TWSVM参数,以生成适合电商经济预测的最佳TWSVM模型,通过PSO优化获得最优TWSVM参数。最后采用最佳TWSVM模型进行电商经济预测,并对预测结果进行评价。在实例仿真中,以电商经济销售金额和增长率两个指标为主,PSO优化的TWSVM算法的预测准确度均高于90%。In order to improve the performance of e-commerce economic forecasting and solve the problem of low forecasting accuracy caused by complex time series,many features and complex user needs,Gemini support vector machine was used to forecast e-commerce economy.Firstly,the economic characters of e-commerce were obtained and then the economic prediction model of Twin Support Vector Machines(TWSVM)was established,and the parameters such as the canonical factor of TWSVM were extracted.Several groups of parameters were randomly initialized to build a particle swarm.Particle swarm optimization(PSO)algorithm was used to search the best parameters of TWSVM to generate the best TWSVM model suitable for e-commerce economic prediction.Through PSO optimization,the optimal TWSVM parameters were obtained.Finally,the best TWSVM model was used to predict e-commerce.In the example simulation,the prediction accuracy of TWSVM algorithm optimized by PSO was higher than 90%,mainly based on the economic sales amount and growth rate of e-commerce.

关 键 词:电商经济 双子支持向量机 粒子群优化 超平面 

分 类 号:TP391[自动化与计算机技术—计算机应用技术]

 

参考文献:

正在载入数据...

 

二级参考文献:

正在载入数据...

 

耦合文献:

正在载入数据...

 

引证文献:

正在载入数据...

 

二级引证文献:

正在载入数据...

 

同被引文献:

正在载入数据...

 

相关期刊文献:

正在载入数据...

相关的主题
相关的作者对象
相关的机构对象