MiMuSA—mimicking human language understanding for fine-grained multi-class sentiment analysis

• Main Authors: Wang, Zhaoxia, Hu, Zhenda, Ho, Seng-Beng, Cambria, Erik, Tan, Ah-Hwee
• Other Authors: School of Computer Science and Engineering
• Format: Article
• Language: English
• Published: 2024
• Subjects: Engineering::Computer science and engineering; Human-Like Understanding; Fine-Grained Sentiment Understanding
• Online Access: https://hdl.handle.net/10356/172942
• Institution: Nanyang Technological University

Sentiment analysis is an important natural language processing (NLP) task due to a wide range of applications. Most existing sentiment analysis techniques are limited to the analysis carried out at the aggregate level, merely providing negative, neutral and positive sentiments. The latest deep learn...