基于集合预测的方面级情感三元组提取

Set Prediction for Aspect Sentiment Triplet Extraction

近年来,基于方面级别的情感分析(ABSA)任务受到越来越多的关注。其中,方面级情感三元组提取(ASTE)是ABSA任务中最新的子任务,其要求同时提取出句子的方面词、观点词并输出对应的情感极性。先前的工作大多采用pipeline方式进行提取,忽略了方面词和观点词之间的联系,且容易产生误差传播的问题。对此,该文提出一种基于集合预测的方法,将方面级情感三元组提取问题转换成集合预测问题,以端到端的方式进行三元组提取。在多个基准数据集上的实验表明,该文提出的模型取得了较为先进的结果。

In recent years,aspect-based sentiment analysis(ABSA)has received more and more attention.Among them,aspect sentiment triplet extraction(ASTE)is the key subtask that requires extracting the aspect words and opinion words of the sentence at the same time and outputting the corresponding sentiment polarity.Most previous work has used the pipeline approach for extraction,ignoring the connection between aspect words and opinion words.This paper proposes to treat the aspect sentiment triplet extraction problem as a set prediction issue,and performs triplet extraction in an end-to-end manner.Experiments on multiple benchmark datasets show that the model proposed in this paper has achieved better results.