A Knowledge Regularized Hierarchical Approach for Emotion Cause Analysis

Abstract

Emotion cause analysis, which aims to identify the reasons behind emotions, is a key topic in sentiment analysis. A variety of neural network models have been proposed recently, however, these previous models mostly focus on the learning architecture with local textual information, ignoring the discourse and prior knowledge, which play crucial roles in human text comprehension. In this paper, we propose a new method to extract emotion cause with a hierarchical neural model and knowledge-based regularizations, which aims to incorporate discourse context information and restrain the parameters by sentiment lexicon and common knowledge. The experimental results demonstrate that our proposed method achieves the state-of-the-art performance on two public datasets in different languages (Chinese and English), outperforming a number of competitive baselines by at least 2.08% in F-measure.

A Mutually Auxiliary Multitask Model With Self-Distillation for Emotion-Cause Pair Extraction

“…Xia et al [31] used the Transformer to encode clauses, which achieved excellent performance. In order to overcome the lack of training data, Fan et al [32] and Hu et al [33] coincidentally introduced external emotion knowledge on the basis of hierarchical design to improve the accuracy of the model. In addition, inspired by methods on other NLP tasks, researchers have made a lot of new attempts.…”

Effective Inter-Clause Modeling for End-to-End Emotion-Cause Pair Extraction

“…adopted Transformer encoder augmented with position information and integrated global prediction embedding to improve performance. Fan et al (2019) incorporated sentiment and position regularizers to restrain parameter learning. Hu et al (2019) exploited external sentiment classification corpus to pretrain the model.…”

SKEP: Sentiment Knowledge Enhanced Pre-training for Sentiment Analysis

“…Sentiment words with their polarity are widely used for sentiment analysis, including sentencelevel sentiment classification (Taboada et al, 2011;Shin et al, 2017;Lei et al, 2018;Barnes et al, 2019), aspect-level sentiment classification (Vo and Zhang, 2015), opinion extraction (Li and Lam, 2017), emotion analysis (Gui et al, 2017;Fan et al, 2019) and so on. Lexicon-based method (Turney, 2002;Taboada et al, 2011) directly utilizes polarity of sentiment words for classification.…”

“…There are many specific sentiment tasks, and these tasks usually depend on different types of sentiment knowledge including sentiment words, word polarity and aspect-sentiment pairs. The importance of these knowledge has been verified by tasks at different level, for instance, sentence-level sentiment classification (Taboada et al, 2011;Shin et al, 2017;Lei et al, 2018), aspect-level sentiment classification (Vo and Zhang, 2015;Zeng et al, 2019), opinion extraction (Li and Lam, 2017;Gui et al, 2017;Fan et al, 2019) and so on. Therefore, we assume that, by integrating these knowledge into the pre-training process, the learned representation would be more sentimentspecific and appropriate for sentiment analysis.…”