From Sentiment Annotations to Sentiment Prediction through Discourse Augmentation

2020/11/05 by Patrick Huber, Huber, Patrick, Giuseppe Carenini +1
Computer Science · #Computation and Language (cs.CL) #FOS: Computer and information sciences #cs.CL

paper · pdf · doi:10.48550/arxiv.2011.03021

In Proceedings of the 28 International Conference on Computational Linguistics (COLING). 10 pages

arxiv created 2020/11/05 · arxiv updated 2020/11/06

Abstract

Sentiment analysis, especially for long documents, plausibly requires methods capturing complex linguistics structures. To accommodate this, we propose a novel framework to exploit task-related discourse for the task of sentiment analysis. More specifically, we are combining the large-scale, sentiment-dependent MEGA-DT treebank with a novel neural architecture for sentiment prediction, based on a hybrid TreeLSTM hierarchical attention model. Experiments show that our framework using sentiment-related discourse augmentations for sentiment prediction enhances the overall performance for long documents, even beyond previous approaches using well-established discourse parsers trained on human annotated data. We show that a simple ensemble approach can further enhance performance by selectively using discourse, depending on the document length.

Related

• MEGA RST Discourse Treebanks with Structure and Nuclearity from Scalable Distant Sentiment Supervision
• Predicting Discourse Structure using Distant Supervision from Sentiment
• Better Document-level Sentiment Analysis from RST Discourse Parsing
• Sentiment Analysis in the Era of Large Language Models: A Reality Check
• Fortunately, Discourse Markers Can Enhance Language Models for Sentiment Analysis