A KGE Based Knowledge Enhancing Method for Aspect-Level Sentiment Classification

ALSC (Aspect-level Sentiment Classification) is a fine-grained task in the field of NLP (Natural Language Processing) which aims to identify sentiment toward given aspect. In addition exploiting sentence semantics and syntax, current methods focus on introducing external knowledge as supplementary information. However, integration three categories information still challenging. this paper, novel method devised effectively combine sufficient semantic syntactic well use knowledge. The proposed model contains encoder, learning module, syntax enhancement an fusion module classifier. are respectively extracted via self-attention network graphical convolutional network. Specifically, KGE (Knowledge Graph Embedding) employed enhance feature representation Then, attention-based gate mechanism taken fuse types We evaluated benchmark datasets experimental results establish strong evidence high accuracy.