Cross-Modal Multitask Transformer for End-to-End Multimodal Aspect-Based Sentiment Analysis

作者：

Highlights：

• We study a new task named End-to-End Multimodal Aspect Based Sentiment Analysis (MABSA).

• We propose a Cross-Modal Multitask Transformer (CMMT) framework for End-to-End MABSA.

• Experimental results show that CMMT outperforms a number of existing approaches.

摘要

•We study a new task named End-to-End Multimodal Aspect Based Sentiment Analysis (MABSA).•We propose a Cross-Modal Multitask Transformer (CMMT) framework for End-to-End MABSA.•Experimental results show that CMMT outperforms a number of existing approaches.

论文关键词：Fine-grained opinion mining,Aspect-Based Sentiment Analysis,Multimodal Sentiment Analysis

论文评审过程：Received 7 March 2022, Revised 25 June 2022, Accepted 19 July 2022, Available online 2 August 2022, Version of Record 2 August 2022.

论文官网地址：https://doi.org/10.1016/j.ipm.2022.103038