Journal of Computer Science and Technology


Graph Enhanced Transformer for Aspect Category Detection

Chen Chen1, Hou-Feng Wang2, Qing-Qing Zhu3, and Jun-Fei Liu4   

  1. 1Office of the Cyberspace Affairs Commission, Peking University, Beijing 100871, China
    2School of Electronics Engineering and Computer Science, Peking University, Beijing 100871, China
    3School of Software and Microelectronics, Peking University, Beijing 100871, China
    4National Engineering Research Center for Software Engineering, Peking University, Beijing 100871, China

Aspect category detection is one challenging subtask of aspect based sentiment analysis, which categorizes a review sentence into a set of predefined aspect categories. Most existing methods regard aspect category detection as a flat classification problem. However, aspect categories are inter-related, and they are usually organized with a hierarchical tree structure. To leverage the structure information, this paper proposes a hierarchical multi-label classification model to detect aspect categories and uses a graph enhanced transformer network to integrate label dependency information into prediction features. Experiments have been conducted on four widely used benchmark datasets, showing that the proposed model outperforms all strong baselines.


方面类别检测是方面级情感分析的子任务之一. 给定一组预定义的方面类别,方面类别检测致力于将评论文本分类到一个或多个方面类别作为评价的目标。方面类别由对象实体和对象属性组合而成,实体对象可以是被评价的实体本身,也可以是实体的一部分、一个模块或相关实体,而对象属性是实体对象某个特定的属性。对于特定领域,预定义的方面类别之间通常是相互关联的,可组织成树状层次结构。然而,现有方面类别检测方法通常将方面类别检测任务视为平面分类问题,未利用到类别之间的关联信息。本文提出将预定义的方面类别归纳为树状结构,基于层次多标签分类框架对方面类别检测任务建模,并设计了一种图结构增强的自注意力机制应用于变换器网络,来学习类别之间的依赖关系表示,实现树结构下的类别信息交互,从而有效提高了方面类别检测的性能。我们在国际语义评测提供的四个基准数据集进行了充分的实验,结果证明了所提方法达到了当前最优的方面类别检查性能。

Key words: aspect based sentiment analysis, aspect category detection, hierarchical multi-label classification, transformer network


No related articles found!
Full text



[1] Zhou Di;. A Recovery Technique for Distributed Communicating Process Systems[J]. , 1986, 1(2): 34 -43 .
[2] Li Wei;. A Structural Operational Semantics for an Edison Like Language(2)[J]. , 1986, 1(2): 42 -53 .
[3] Feng Yulin;. Recursive Implementation of VLSI Circuits[J]. , 1986, 1(2): 72 -82 .
[4] Gao Qingshi; Zhang Xiang; Yang Shufan; Chen Shuqing;. Vector Computer 757[J]. , 1986, 1(3): 1 -14 .
[5] Jin Lan; Yang Yuanyuan;. A Modified Version of Chordal Ring[J]. , 1986, 1(3): 15 -32 .
[6] Pan Qijing;. A Routing Algorithm with Candidate Shortest Path[J]. , 1986, 1(3): 33 -52 .
[7] Zhang Cui; Zhao Qinping; Xu Jiafu;. Kernel Language KLND[J]. , 1986, 1(3): 65 -79 .
[8] Min Yinghua; Han Zhide;. A Built-in Test Pattern Generator[J]. , 1986, 1(4): 62 -74 .
[9] Lu Xuemiao;. On the Complexity of Induction of Structural Descriptions[J]. , 1987, 2(1): 12 -21 .
[10] Wu Yunzeng;. On the Development of Applications of Logic in Programming[J]. , 1987, 2(1): 30 -34 .

ISSN 1000-9000(Print)

CN 11-2296/TP

Editorial Board
Author Guidelines
Journal of Computer Science and Technology
Institute of Computing Technology, Chinese Academy of Sciences
P.O. Box 2704, Beijing 100190 P.R. China
  Copyright ©2015 JCST, All Rights Reserved