期刊文献+

Lasso-BP组合模型在鲤鱼价格预测中的应用研究 被引量:1

Research on the Application of Lasso-BP Combined Model in Carp Price Forecast
在线阅读 下载PDF
导出
摘要 为了对水产品市场中的水产品价格进行预测,本文以北京地区的鲤鱼市场为例,基于Lasso回归和BP神经网络组合模型,对鲤鱼价格和影响因素进行预测研究。通过Lasso回归对影响因素进行筛选,将筛选出的影响因素输入到BP神经网络中进行训练,并将影响因素数据输入训练后的模型中,即可输出预测价格。同时,选取2017年10月1日至2018年6月17日数据作为测试样本,对测试样本进行归一化处理,并采用本文提出的模型进行实验。实验结果表明,Lasso-BP模型预测值与真实值整体平均误差为2.59%,在价格波动趋势方面,预测数据准确率达96.53%,在价格波动幅度方面,预测结果的波动幅度与真实数据基本相同,平均误差为2.51%,说明模型预测精度较高,预测效果良好,该文模型满足实际生产工作需要,证明Lasso-BP模型能够对鲤鱼价格进行准确预测。该研究具有一定的理论意义和实际意义。 In order to predict the price of aquatic products in the aquatic product market,this paper takes the carp market in Beijing as an example,based on the Lasso regression and BP neural network combination model,to predict the price of carp and its influencing factors.The influencing factors are screened through Lasso regression,the selected influencing factors are input into the BP neural network for training,and the influencing factor data is input into the trained model,and the predicted price can be output.At the same time,the data from October 1,2017 to June 17,2018 was selected as the test sample,and the test sample was normalized and the model proposed in this paper was used for experiments.The experimental results show that the overall average error between the predicted value of the Lasso-BP model and the actual value is 2.59%.In terms of price fluctuation trend,the accuracy rate of the predicted data is 96.53%.In terms of price fluctuation range,the fluctuation range of the predicted result is basically the same as the real data with the average error being 2.51%,indicating that the model has high prediction accuracy and good prediction effect.The model proposed in this paper meets the needs of actual production work,which proves that the Lasso-BP model can accurately predict the price of carp.The research has certain theoretical and practical significance.
作者 李海涛 刘昌年 LI Haitao;LIU Changnian(School of Information Science&Technology,Qingdao University of Science and Technology,Qingdao 266061,China)
出处 《青岛大学学报(工程技术版)》 CAS 2022年第4期1-8,29,共9页 Journal of Qingdao University(Engineering & Technology Edition)
基金 农业部水产养殖数字建设试点项目(2017-A2131-130209-K0104-004) 青岛市创新创业领军人才(15-07-03-0030)。
关键词 Lasso回归模型 BP神经网络 ARIMA模型 价格预测 水产品价格 Lasso regression model BP neural network ARIMA model price prediction aquatic product price
  • 相关文献

参考文献15

二级参考文献110

共引文献609

同被引文献14

引证文献1

二级引证文献1

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部