摘要
为进一步提高LSTM网络预测的精度,提出一种基于ResNet和Attention改进LSTM的风险预测模型。其中,利用ResNet10网络对样本数据进行特征提取,然后引入Attention赋予不同特征权重,进而提高重要特征的权重,最后运用LSTM对样本数据进行预测。结果表明,本研究的方案得到的风险预测准确率、准确率等指标明显高于其他风险预测模型,由此得出本方改进方案可行。
To further improve the accuracy of LSTM network prediction,a risk prediction model based on ResNet and Attention improved LSTM is proposed.Among them,the ResNet10 network is used to extract features from the sample data,and then attention is introduced to assign different feature weights,thereby increasing the weight of important features.Finally,LSTM is used to predict the sample data.The results indicate that the accuracy and other indicators of risk prediction obtained by the proposed approach are significantly higher than those of other risk prediction models,indicating the feasibility of our improved approach.
作者
高敏
GAO Min(Heze Information Engineering School of Shandong Province,Heze 274000,Shandong China)
出处
《粘接》
2025年第3期193-196,共4页
Adhesion
