• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文会议>International conference on neural information processing;Annual conference of Asia-Pacific Neural Network Society >Aspect-Level Sentiment Classification with Dependency Rules and Dual Attention
【24h】

Aspect-Level Sentiment Classification with Dependency Rules and Dual Attention

机译：具有依赖规则和双重注意的方面级别的情感分类

获取原文
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

Aspect-level sentiment classification aims to predict the sentiment polarity towards the given aspects of sentences. Neural network models with attention mechanism have achieved great success in this area. However, existing methods fail to capture enough aspect information. Besides, it is hard for simple attention mechanism to model complex interaction between aspects and contexts. In this paper, we propose a Segment Model with Dual Attention (SegM-DA) to tackle these problems. We combine deep learning models with traditional methods by defining dependency rules to extract auxiliary words, which helps to enrich aspect information. In addition, in order to model structural relation between aspects and contexts, we introduce dependent attention mechanism. Coupled with standard attention mechanism, we establish the dual attention mechanism, which models the interaction from both word- and structure- dependency. We perform aspect-level sentiment classification experiments on two real datasets. The results show that our model can achieve the state-of-the-art. performance.
机译：方面级别的情感分类旨在预测句子给定方面的情感极性。具有注意力机制的神经网络模型在这一领域取得了巨大的成功。但是，现有方法无法捕获足够的方面信息。此外，简单的注意力机制很难对方面和上下文之间的复杂交互进行建模。在本文中，我们提出了一种具有双重注意力的细分模型（SegM-DA）来解决这些问题。我们通过定义依赖关系规则来提取辅助词，从而将深度学习模型与传统方法结合起来，这有助于丰富方面信息。另外，为了建模方面和上下文之间的结构关系，我们引入了依赖注意机制。结合标准的注意机制，我们建立了双重注意机制，该机制从单词和结构依赖性两个方面对交互进行了建模。我们在两个真实的数据集上执行方面级别的情感分类实验。结果表明，我们的模型可以达到最新水平。表现。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱：kefu@zhangqiaokeyan.com

京公网安备：11010802029741号 ICP备案号：京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有

  • 客服微信

  • 服务号