【摘要】 针对方面级情感分析任务中语义信息难以提取以及方面词信息难以和上下文信息相关联的问题，提出一种融合双通道的语义信息模型(FDCS)。通过BERT预训练模型搭建两个通道获取不同层次的语义信息，一个是全局信息通道，另一个是句子信息通道；使用语义注意力融合双通道中不同层次的语义信息，将融合后的语义信息再次分别融入全局信息和句子信息；根据每个通道语义信息的不同分别提取相应的特征信息。在3个基准数据集上的实验结果表明，该模型的性能优于其它模型。更多还原

【Abstract】 Aiming at the problems that semantic information is difficult to extract, and that aspect word information is difficult to associate with context information in aspect-level sentiment analysis tasks, a fused dual-channel semantic information model(FDCS) was proposed. Two channels were built through the BERT pre-training model to obtain semantic information at diffe-rent levels, one was the global information channel, and the other was the sentence information channel. Semantic attention was used to fuse the semantic information of different levels in the dual channels, and the fused semantic information was re-integrated into the global information and sentence information respectively. The corresponding feature information was extracted according to the different semantic information of each channel. Experimental results on three benchmark datasets show that this model outperforms other models.更多还原