arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2306.04152cs.AI
收藏

面向方面情感四元组预测的统一单步解决方案

A Unified One-Step Solution for Aspect Sentiment Quad Prediction

DataStory(数说故事);International Digital Economy Academy(国际数字经济研究院)

Junxian Zhou, Haiqin Yang, Yuxuan He, Hao Mou, Junbo Yang

AI总结:

针对 ASQP 数据规模小、密度低及现有方法误差传播或效率不足的问题,本文发布两个新数据集,并提出单步模型 One-ASQP,同时预测方面类别与 AOS 三元组。

摘要:

方面情感四元组预测(ASQP)是基于方面的情感分析中一项具有挑战性且意义重大的子任务,因为它能提供完整的方面级情感结构。然而,现有 ASQP 数据集通常规模较小且密度较低,阻碍了技术进步。为扩充容量,本文发布两个新的 ASQP 数据集,其特点包括:规模更大、每个样本包含更多词语以及密度更高。借助这些数据集,我们揭示了现有强 ASQP 基线的不足,并因此提出一种统一的单步 ASQP 解决方案,即 One-ASQP,用于同时检测方面类别并识别方面—观点—情感(AOS)三元组。我们的 One-ASQP 具有若干独特优势:(1)通过将 ASQP 分离为两个子任务并独立且同时求解,可避免基于流水线方法中的误差传播,并克服基于生成方法训练和推理缓慢的问题;(2)通过在基于词元对的二维矩阵中引入情感特定的角标记模式,可利用情感元素之间更深层的交互,并高效解码 AOS 三元组;(3)我们设计的“[NULL]”词元可帮助有效识别隐式方面或观点。在两个基准数据集和我们发布的两个数据集上的实验证明了 One-ASQP 的优势。两个新数据集公开发布于 \url{https://www.github.com/Datastastory-CN/ASQP-Datasets}。

英文摘要:

Aspect sentiment quad prediction (ASQP) is a challenging yet significant subtask in aspect-based sentiment analysis as it provides a complete aspect-level sentiment structure. However, existing ASQP datasets are usually small and low-density, hindering technical advancement. To expand the capacity, in this paper, we release two new datasets for ASQP, which contain the following characteristics: larger size, more words per sample, and higher density. With such datasets, we unveil the shortcomings of existing strong ASQP baselines and therefore propose a unified one-step solution for ASQP, namely One-ASQP, to detect the aspect categories and to identify the aspect-opinion-sentiment (AOS) triplets simultaneously. Our One-ASQP holds several unique advantages: (1) by separating ASQP into two subtasks and solving them independently and simultaneously, we can avoid error propagation in pipeline-based methods and overcome slow training and inference in generation-based methods; (2) by introducing sentiment-specific horns tagging schema in a token-pair-based two-dimensional matrix, we can exploit deeper interactions between sentiment elements and efficiently decode the AOS triplets; (3) we design ``[NULL]'' token can help us effectively identify the implicit aspects or opinions. Experiments on two benchmark datasets and our released two datasets demonstrate the advantages of our One-ASQP. The two new datasets are publicly released at \url{https://www.github.com/Datastory-CN/ASQP-Datasets}.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑