A weakly-supervised graph-based joint sentiment topic model for multi-topic sentiment analysis

Information Sciences（2022）

引用 3|浏览16

暂无评分

摘要

•A novel graph-based joint sentiment topic model, with an importance sampling-based training scheme and self-training approached, is proposed, that integrates an edge-gated graph convolutional network to better learn the representation of topics and works.

查看译文

关键词

Multi-topic sentiment analysis,Edge-gated graph convolutional network,Joint sentiment-topic model,Self-training

AI 理解论文

溯源树

样例

生成溯源树，研究论文发展脉络

Chat Paper

正在生成论文摘要