首页|基于LSTM神经网络的超市蔬菜类商品收益预测

基于LSTM神经网络的超市蔬菜类商品收益预测

扫码查看
目前深度学习预测技术已经十分成熟,适用于多种类型的预测任务,可以对时间序列数据进行准确预测.长短期记忆神经网络(LSTM)在循环神经网络(RNN)的基础上解决了梯度消失和梯度爆炸的问题,可用于预测长时间序列数据.文章提出了基于长短期记忆网络的超市蔬菜类商品的收益预测算法.首先利用长短期记忆(LSTM)对蔬菜单品净藕的批发单价、销售单价、销售总量进行预测,在此结果的基础上对花叶类蔬菜进行预测,从结果中可以看出长短期记忆网络(LSTM)对蔬菜类商品的预测效果表现十分出色,最后得到花叶类蔬菜的商品总收益.
Revenue Prediction of Supermarket Vegetable Commodities Based on LSTM Neural Network
In today's highly competitive market environment,accurate sales forecasting is crucial for the efficient operation of supermarkets,especially in the field of vegetable commodities.To optimize inventory control of vegetable commodities in supermarkets,it is essential to accurately predict the demand for these products.This study utilizes the LSTM neural network to analyze and forecast historical sales data based on time series,predicting the total sales volume,wholesale prices,and unit selling prices of vegetable commodities.The approach further refines predictions for categorized vegetable commodities and individual items,forecasting replenishment quantities and pricing strategies for each item within the categorized commodities.The objective is to cal-culate the maximum revenue for supermarkets,providing supermarket management with more precise inventory management decisions for procurement and sales strategies.This,in turn,serves as a reliable basis for maximizing sales profit in supermarkets.

LSTM neural networksupermarket revenuereplenishment quantity

茹慧英、马嘉灏、贺豪、安彦宇、郑佳琦

展开 >

河北建筑工程学院,河北张家口 075000

长短期记忆神经网络 循环神经网络 预测

河北省高等学校科学技术研究项目

QN2022097

2024

长江信息通信
湖北通信服务公司

长江信息通信

影响因子:0.338
ISSN:2096-9759
年,卷(期):2024.37(4)
  • 4