Aspect Sentiment Triplet Extraction: A Seq2Seq Approach With Span Copy Enhanced Dual Decoder

Aspect Sentiment Triplet Extraction: A Seq2Seq Approach With Span Copy Enhanced Dual Decoder
复制标题

通过 Hawkes 增强序列模型进行电信欺诈检测

DOI:
10.1109/taslp.2022.3198802
复制
发表时间:
2022
期刊:
IEEE/ACM Transactions on Audio, Speech, and Language Processing
影响因子:
--
通讯作者:
Junjie Wu
Junjie Wu
中科院分区:
其他
文献类型:
--
作者:
Zhihao Zhang;Yuan Zuo;Junjie Wu

文献摘要

参考文献

相似文献

面向情感三联体提取(ASTE)是一项相对较新的、具有挑战性的任务,它试图为基于面向情感的分析提供一个完整的解决方案。一个句子中的情态三联体通常有重叠,例如,一个情态与多个观点相关联,反之亦然。近年来,端到端的ASTE方法因其能够避免基于管道的方法的错误传播问题而受到越来越多的关注。然而,现有的基于标记的端到端方法难以获得令人满意的召回,而生成方法未能充分考虑到方面、观点及其相应情绪之间的潜在相互作用。在本文中,我们将ASTE任务形式化为一个具有跨度复制机制的Seq2Seq学习问题,用于提取多个和可能重叠的三元组。针对ASTE任务,设计了一种新的双解码器,其中提出了一种基于多头注意的跨度复制机制来复制多个令牌方面和意见。双解码器得益于丰富的编码器输出,可以融合多种类型的信息,包括词语义、POS标签和BIO标签。在各种基准数据集上的实验表明,我们的方法达到了新的最先进的结果。我们还进行了分析实验,以验证各种模型组件的有效性,特别是对于重叠的三联体提取。我们发现我们的模型可以通过数据增强和后训练得到进一步的改进。
Aspect Sentiment Triplet Extraction (ASTE) is a relatively new and very challenging task that attempts to provide an integral solution for aspect-based sentiment analysis. Aspect sentiment triplets in a sentence usually have overlaps when, e.g., one aspect is associated with multiple opinions and vice versa. Recently, end-to-end ASTE methods are becoming more and more popular for they can avoid the error propagation problem of pipeline-based methods. However, existing tagging-based end-to-end methods face difficulty to obtain a satisfactory recall, and generative methods fail to take a full account of the underlying interactions between aspects, opinions and their corresponding sentiments. In this paper, we formalize the ASTE task as a Seq2Seq learning problem with span copy mechanism for extracting multiple and possibly overlapped triplets. A novel dual decoder is devised purposefully for the ASTE task, where a multi-head attention based span copy mechanism is proposed to copy multi-token aspects and opinions. The dual decoder benefits from the rich output of encoder that can fuse multi-type information including word semantic, POS tag and BIO tag. Experiments on various benchmark datasets demonstrate that our approach achieves new state-of-the-art results. We also conduct analytical experiments to verify the effectiveness of various model components particularly for overlapped triplets extraction. We find that our model can be further improved through data augmentation and post-training.
DOI: 10.18653/v1/d19-1464
发表时间: 2019-09
期刊: ArXiv
影响因子: --
作者:
Chen Zhang;Qiuchi Li;D. Song
通讯作者: Chen Zhang;Qiuchi Li;D. Song
DOI: --
发表时间: 2019-08
期刊: ArXiv
影响因子: --
作者:
A. Rietzler;Sebastian Stabinger;Paul Opitz;Stefan Engl
通讯作者: A. Rietzler;Sebastian Stabinger;Paul Opitz;Stefan Engl
DOI: 10.18653/v1/2021.emnlp-main.731
发表时间: 2021-10
期刊: --
影响因子: --
作者:
Rajdeep Mukherjee;Tapas Nayak;Yash Butala;Sourangshu Bhattacharya;Pawan Goyal
通讯作者: Rajdeep Mukherjee;Tapas Nayak;Yash Butala;Sourangshu Bhattacharya;Pawan Goyal
DOI: 10.18653/v1/2020.acl-main.340
发表时间: 2020-07
期刊: The Journal of Steroid Biochemistry and Molecular Biology
影响因子: --
作者:
Zhuang Chen-;T. Qian
通讯作者: Zhuang Chen-;T. Qian
DOI: 10.1609/aaai.v35i15.17606
发表时间: 2020-06
期刊: --
影响因子: --
作者:
Sheena Panthaplackel;Miltiadis Allamanis;Marc Brockschmidt
通讯作者: Sheena Panthaplackel;Miltiadis Allamanis;Marc Brockschmidt