基于改进TF-IDF与BERT的领域情感词典构建方法  

Construction Method of Domain Sentiment Lexicon Based on Improved TF-IDF and BERT

作　　者：蒋昊达 赵春蕾[1,2] 陈瀚 王春东 JIANG Haoda;ZHAO Chunlei;CHEN Han;WANG Chundong(Key Laboratory of Computer Vision and System of Ministry of Education,Tianjin University of Technology,Tianjin 300384,China;Tianjin Key Laboratory of Intelligent Computing and Novel Software Technology,Tianjin 300384,China)

机构地区：[1]天津理工大学教育部计算机视觉与系统省部共建重点实验室,天津300384 [2]天津市智能计算与软件新技术重点实验室,天津300384

出　　处：《计算机科学》2024年第S01期150-158,共9页Computer Science

基　　金：国家重点研发计划“科技助力经济2020”重点专项项目(SQ2020YFF0413781,SQ2020YFF0401503)。

摘　　要：领域情感词典的构建是领域文本情感分析的基础。现有的领域情感词典构建方法存在所筛选候选情感词冗余度高、情感极性判断失准、领域依赖性强等问题。为了提高所筛选候选情感词的领域性和判断领域情感词极性的准确程度,提出了一种基于改进词频-逆文档频率(TF-IDF)与BERT的领域情感词典构建方法。该方法在筛选领域候选情感词阶段对TF-IDF算法进行改进,将隐含狄利克雷分布(LDA)算法与改进后的TF-IDF算法结合,进行领域性修正,提升了所筛选候选情感词的领域性;在候选情感词极性判断阶段,将情感倾向点互信息算法(SO-PMI)与BERT结合,利用领域情感词微调BERT分类模型,提高了判断领域候选情感词情感极性的准确程度。在不同领域的用户评论数据集上进行实验,结果表明,该方法可以提高所构建领域情感词典的质量,使用该方法构建的领域情感词典用于汽车领域和手机领域文本情感分析的F1值分别达到78.02%和88.35%。The construction of a domain sentiment lexicon is the foundation of domain text sentiment analysis.The existing me-thods for constructing domain sentiment lexicon have problems such as high redundancy of selected candidate sentiment words,inaccurate judgment of sentiment polarity,and high domain dependency.In order to improve the domain specificity of selected candidate sentiment words and the accuracy of judging the polarity of domain sentiment words,a domain sentiment lexicon construction method based on improved term frequency-inverse document frequency(TF-IDF)and BERT is proposed.This method improves the TF-IDF algorithm in the phase of selecting domain candidate sentiment words.The latent dirichlet allocation(LDA)algorithm is combined with the improved TF-IDF algorithm to perform domain corrections,improves the domain specificity of the selected candidate sentiment words.In the polarity judgment stage of candidate sentiment words,the semantic orientation pointwise mutual information(SO-PMI)algorithm is combined with BERT.By fine-tuning the BERT classification model using domain sentiment words,the accuracy of judging the sentiment polarity of domain candidate sentiment words is improved.Experiments are conducted on user comment datasets in different domains,and the experimental results show that this method can improve the quality of the constructed domain sentiment lexicon,and the F1 value of the domain sentiment lexicon constructed by this method for text sentiment analysis in the automotive field and mobile phone field reaches 78.02%and 88.35%,respectively.

关 键 词：情感分析 领域情感词典 词频-逆文档频率 隐含狄利克雷分布 情感倾向点互信息算法 BERT模型 

分 类 号：TP391.1[自动化与计算机技术—计算机应用技术]