基于BiLSTM-CNN-MultiHeadAttention-Dropout的物流评论情感分析

Sentiment Analysis of Logistics Comments Based on BiLSTM-CNN-MultiHeadAttention-Dropout

物流文本情感分析在快速发展的电商行业中愈加重要,为更好捕获局部情感特征并充分挖掘全局语义信息,提出一种基于BiLSTM-CNN-MultiHeadAttention-Dropout的物流评论情感分析模型。该模型对现有模型进行了改进,通过BiLSTM进行特征获取,对重要部分使用MultiHeadAttention机制捕获特征,采用Dropout机制来防止过拟合,最后用CNN提取特征,并应用于物流领域。为验证该模型有效性,对某电商平台的物流评论进行了实验分析,结果表明,该模型的准确率较高,可以为企业处理物流评论数据提供有效支撑。

Logistics text sentiment analysis is becoming more and more important in the rapidly developing e-commerce industry.In order to better capture local emotional features and fully exploit global semantic information,a logistics comment sentiment analysis model based on BiLSTM-CNN-MultiHeadAttention-Dropout is proposed.This model improves the existing model,acquires features through BiLSTM,captures features using the MultiHeadAttention mechanism for important parts,and uses the Dropout mechanism to prevent over fitting.Finally,features are extracted using CNN and applied to the logistics field.In order to verify the effectiveness of the model,an experimental analysis was conducted on the logistics review of an e-commerce platform.The results show that the accuracy of the model is high,which can provide effective support for enterprises to process logistics review data.