Emotion-Cause Pair Extraction as Sequence Labeling Based on A Novel Tagging Scheme

Abstract

The task of emotion-cause pair extraction deals with finding all emotions and the corresponding causes in unannotated emotion texts. Most recent studies are based on the likelihood of Cartesian product among all clause candidates, resulting in a high computational cost. Targeting this issue, we regard the task as a sequence labeling problem and propose a novel tagging scheme with coding the distance between linked components into the tags, so that emotions and the corresponding causes can be extracted simultaneously. Accordingly, an end-to-end model is presented to process the input texts from left to right, always with linear time complexity, leading to a speed up. Experimental results show that our proposed model achieves the best performance, outperforming the state-of-the-art method by 2.26% (p < 0.001) in F 1 measure.

Clause Level Attention Network for Emotion Cause Pair Extraction

“…Ding et al [7] combined construction, interplay and projections of EC pairs into a two-dimensional table. Yuan et al [8] (2020) and Fan et al [9] (2021) designed a cause-oriented labeling scheme to solve the ECPE task.…”

“…The greater the relative distance between the two clauses, the lower the probability of matching. Following Chen et al [8] (2020), we refer to sentences with a relative position less than three to the current sentence as neighbouring sentences. To establish the interaction between sentiment causes, we connect the sentiment/cause clause with its neighboring clauses to obtain the initial representation of both pairs.…”

“…Our model is trained by jointly optimizing the task ECPE using the Adam optimizer. We calculate the loss as in equation (8).…”

A Multi-turn Machine Reading Comprehension Framework with Rethink Mechanism for Emotion-Cause Pair Extraction

“…However, the average ratio of invalid/valid pairs is more than 200:1 in the ECPE benchmark corpus , which leads to a label sparsity problem. Besides, some works (Chen et al, 2020a;Yuan et al, 2020) proposed sequence labeling methods based on novel tagging schemes. However, they cannot effectively model the corresponding relation between emotions and causes.…”

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

“…These methods all solved multiple tasks in a framework, which achieved good performance. Hence, most recent studies focused on extracting emotion-cause pairs in an end-to-end manner [45]- [53].…”