计算机与数字工程2026，Vol.54Issue(4)：923-927,983,6.DOI:10.3969/j.issn.1672-9722.2026.04.003
基于改进的GWO和LSTM的加密货币预测模型
Cryptocurrency Prediction Model Based on Improved GWO and LSTM
许增晖 1王保卫 2张骏豪 1沈露遥1
作者信息
- 1. 南京信息工程大学软件学院 南京 210044
- 2. 南京信息工程大学计算机学院 南京 210044
- 折叠
摘要
Abstract
Cryptocurrency is a digital exchange medium using blockchain technology,which has the advantages of safe and transparent transaction process and immutable transaction records.Although cryptocurrency is recognized by many financial institu-tions,the uncertainty of its price poses a major risk to investment,so the price prediction of cryptocurrency has become a research hotspot.This paper proposes a cryptocurrency prediction model based on GWO,LSTM,and Attention mechanism.This paper uses the Attention mechanism to improve the LSTM model,which improves the mining performance of the long-term dependence of the LSTM model on time series data,thereby improving the prediction accuracy of the model.This paper designs an improved algorithm of GWO,IGWO,to improve the optimization performance of the GWO algorithm for hyperparameters of the network model.Finally,the IGWO algorithm is used to optimize the hyperparameters of LSTM-Attention to further improve the prediction accuracy of the model.Compared with CNN,RNN,LSTM single model,the prediction accuracy of the model proposed in this paper is significantly improved.关键词
加密货币/价格预测/灰狼算法/长短期记忆网络Key words
cryptocurrency/price prediction/GWO/LSTM分类
信息技术与安全科学引用本文复制引用
许增晖,王保卫,张骏豪,沈露遥..基于改进的GWO和LSTM的加密货币预测模型[J].计算机与数字工程,2026,54(4):923-927,983,6.
