Journal of Computer Applications ›› 2018, Vol. 38 ›› Issue (9): 2464-2468.DOI: 10.11772/j.issn.1001-9081.2018020481

Previous Articles     Next Articles

Joint model of microblog emotion recognition of emoticons and emotion cause detection based on neural network

ZHANG Chen1, QIAN Tao2, JI Donghong1   

  1. 1. School of Cyber Science and Engineering, Wuhan University, Wuhan Hubei 430072, China;
    2. School of Computer Science and Technology, Hubei University of Science and Technology, Xianning Hubei 437100, China
  • Received:2018-03-12 Revised:2018-05-04 Online:2018-09-10 Published:2018-09-06
  • Contact: 钱涛
  • Supported by:
    This work is partially supported by the Surface Project of National Natural Science Foundation of China (61772378), the Surface Project of Hubei Natural Science Foundation (2018CFB690).

基于神经网络的微博情绪识别与诱因抽取联合模型

张晨1, 钱涛2, 姬东鸿1   

  1. 1. 武汉大学 国家网络安全学院, 武汉 430072;
    2. 湖北科技学院 计算机学院, 湖北 咸宁 437100
  • 通讯作者: 钱涛
  • 作者简介:张晨(1992—),男,江苏镇江人,硕士研究生,主要研究方向:自然语言处理、深度学习;钱涛(1975—),男,湖北潜江人,副教授,博士,CCF会员,主要研究方向:自然语言处理、机器学习;姬东鸿(1967—),男,湖北武汉人,教授,博士,主要研究方向:自然语言处理、数据挖掘。
  • 基金资助:
    国家自然科学基金面上项目(61772378);湖北省自然科学基金面上项目(2018CFB690)。

Abstract: As a deep text emotion understanding, emotion cause detection has become a hot issue in the field of text emotion analysis, but current research usually regards emotion cause detection and emotion recognition as two independent tasks, which easily leads to propagation of errors. Considering that emotion cause detection and emotion recognition are interacted, and that the emoticons in the microblog text usually express the emotion of the text, a joint model of emotion cause detection and emotion recognition of emoticons based on Bi-directional Long Short-Term Memory Conditional Random Field (Bi-LSTM-CRF) model was proposed. This model formalizes emotion cause detection and emotion recognition into a unified sequence labeling problem, it makes full use of the interaction between emotion causes and emotions and simultaneously processes the two tasks. The experimental results show that this model achieves the F score as 82.70% in emotion cause detection and 74.74% in emotion recognition of emoticons, compared with the serial model, the F score is enhanced by 5.82% and 17.12%, which means the joint model can effectively reduce propagation of errors and improve the F scores of emotion cause detection and emotion recognition of emoticons.

Key words: emotion cause detection, emotion recognition, emoticon, sequence labeling, Bi-directional Long Short-Term Memory Conditional Random Field (Bi-LSTM-CRF), joint model

摘要: 情绪诱因抽取作为深层次的文本情绪理解已成为情绪分析任务中的新热点,当前研究通常把诱因抽取和情绪识别看作两个独立的任务,容易导致错误在任务间的传播问题。考虑到情绪识别及诱因抽取是相互作用的,以及微博文本中表情符通常表达文本的情绪,提出了一种基于双向长短期记忆条件随机场(Bi-LSTM-CRF)模型的情绪诱因和表情符情绪识别的联合模型。该模型将情绪诱因抽取以及情绪识别形式化为一个统一的序列标注问题,充分利用了情绪诱因与情绪之间的互相作用,将情绪诱因的抽取和情绪识别同时进行。实验结果表明,该模型在诱因抽取任务中的F值为82.70%,在情绪识别任务中的F值为74.74%,相比串行模型的F值分别提高5.82和17.12个百分点,这个结果表明联合模型能够有效降低任务串行进行时的误差传递,同时提高了诱因抽取和情绪识别的F值。

关键词: 诱因抽取, 情绪识别, 表情符, 序列标注, 双向长短期记忆条件随机场, 联合模型

CLC Number: