首页|基于用户性格和语义-结构特征的文本评论情感分类方法

基于用户性格和语义-结构特征的文本评论情感分类方法

扫码查看
原文链接
  • 国家科技期刊平台
  • NETL
  • NSTL
  • 万方数据
由于传统文本评论情感分类方法通常忽略用户性格对于情感分类结果的影响,提出一种基于用户性格和语义-结构特征的文本评论情感分类方法(User Personality and Semantic-structural Features based Sentiment Classifica-tion Method for Text Comments,BF_BiGAC).依据大五人格模型能够有效表达用户性格的优势,通过计算不同维度性格得分,从评论文本中获取用户性格特征.利用双向门控循环单元(Bidirectional Gated Recurrent Unit,BiGRU)和卷积神经网络(Convolutional Neural Network,CNN)可以有效提取文本上下文语义特征和局部结构特征的优势,提出一种基于BiGRU、CNN和双层注意力机制的文本语义-结构特征获取方法.为区分不同类型特征的影响,引入混合注意力层实现对用户性格特征和文本语义-结构特征的有效融合,以此获得最终的文本向量表达.在IMDB、Yelp-2、Yelp-5及Ekman四个评论数据集上的对比实验结果表明,BF_BiGAC在分类准确率(Accuracy)和加权macro F1值(Fw)上均获得较好表现,相对于拼接BiGRU、CNN的情感分类方法(Sentiment Classification Method Concatenating BiGRU and CNN,BiG-RU_CNN)在Accuracy值上分别提升0.020、0.012、0.017及0.011,相对于拼接CNN、BiGRU的情感分类方法(Sentiment Classification Method Concatenating CNN and BiGRU,ConvBiLSTM)Fw值上分别提升0.022、0.013、0.028及0.023;相对于预训练模型BERT和RoBERTa,BF_BiGAC在保证分类精度的情况下获得了较高的运行效率.
A Sentiment Classification Method for Text Comments Based on User Personality and Semantic-Structural Features
Since the traditional sentiment classification methods for text comments usually ignore the influence of us-er personality on sentiment classification results,a sentiment classification method for text comments based on user person-ality and semantic-structural features is proposed.According to the advantage of Big Five personality model on effectively expressing the user personality,the user personality feature is obtained from the comment texts by calculating the personali-ty scores from different dimensions.Moreover,the advantages of bidirectional gated recurrent unit(BiGRU)and convolu-tional neural network(CNN)on effectively extracting the contextual semantic features and the local structural features are taken,and a new text semantic-structural feature acquisition method based on BiGRU,CNN and two-layer attention mecha-nism is proposed.Finally,in order to distinguish the influence of the features with different types,the hybrid attention layer is introduced to obtain the final text vector representation by integrating the user personality feature and the textural seman-tic-structural feature effectively.The experimental results on the datasets of IMDB,Yelp-2,Yelp-5 and Ekman show that BF_BiGAC achieves good performance when the measurements of Accuracy and weighted macro F1(Fw)are used.Specifi-cally,it achieves the improvements of 0.020,0.012,0.017 and 0.011 compared to sentiment classification method concate-nating BiGRU and CNN(BiGRU_CNN)on accuracy,and achieves the improvements of 0.022,0.013,0.028 and 0.023 compared to sentiment classification method concatenating CNN and BiGRU(ConvBiLSTM)on Fw.Moreover,when com-paring with the pre-trained models of BERT and RoBERTa,BF_BiGAC achieves higher executing efficiency while ensur-ing the classification accuracy.

sentiment classificationBig Five personality modelbidirectional gated recurrent unitconvolutional neural networkattention mechanism

王友卫、刘瑞、凤丽洲

展开 >

中央财经大学信息学院,北京 100081

天津财经大学统计学院,天津 300222

情感分类 大五人格模型 双向门控循环单元 卷积神经网络 注意力机制

国家自然科学基金教育部人文社科资助项目国家社会科学基金中央财经大学新兴交叉学科建设项目

6190622019YJCZH17818CTJ008

2024

10.12263/DZXB.20220645

电子学报
中国电子学会

电子学报

CSTPCD北大核心
影响因子：1.237
ISSN：0372-2112
年,卷(期)：2024.52(5)