Improved Scheme for Extracting Aspect Sentiment Triplets Based on Aspect Aware Attention Enhancement

In response to the problems of high model complexity, low computational efficiency, weak generalization ability, and inflexible parameter adapters in the aspect aware attention enhanced aspect sentiment triplet extraction scheme, this paper proposes an improvement scheme based on convolutional network approach for extracting aspect sentiment triplets which has been further optimized and improved. The model structure is optimized through transformer variants, the model generalization ability is enhanced through transfer learning methods, the effectiveness of the parameter adapter is improved through dynamic weight adjustment, a knowledge graph is introduced to test the benchmark dataset accurately, and the visualization tool Netron is introduced to enhance the interpretability of the model. The experimental parameters indicate that the scheme has been optimized in terms of performance and accuracy in extracting triplet data.