基于表情和语气的情感词典用于弹幕情感分析  

Emotional Dictionary Based on Emoticons and Modal for Barrage Sentiment Analysis

作　　者：邱全磊 崔宗敏[1] 喻静 QIU Quan-lei;CUI Zong-min;YU Jing(School of Information Science and Technology,Jiujiang University,Jiujiang 332005,China)

机构地区：[1]九江学院信息科学与技术学院,江西九江332005

出　　处：《计算机技术与发展》2020年第8期178-182,共5页Computer Technology and Development

基　　金：国家自然科学基金(61762055)。

摘　　要：现有的方法没有考虑颜文字表情对情感分析的影响,同时没有考虑语气词在情感表达中的作用。这影响了弹幕情感分析的效果,降低了情感分析的准确率。为此,构建了一种新的基于表情和语气的情感词典用于弹幕情感分析。首先提出了一种新的应用于弹幕的情感词典构建方法,该方法在以往构建情感词典方法的基础上新增加了弹幕表情词典以及弹幕语气词典的构建方法;然后,提出了一种新的基于弹幕情感词典的情感值计算方法,该方法在对弹幕进行情感值计算时,考虑了表情和语气对情感分析的影响;最后,将提出的方法与现有的方法进行对比实验。实验结果表明,提出的方法在召回率、表情分析、语气词分析等方面比现有的方法在弹幕情感分析领域具有更好的性能。The existing methods do not take into account the influence of emoticons on sentiment analysis and the role of modal particles in emotional expression,which affects the effect of the sentiment analysis of the barrage and reduces the accuracy of sentiment analysis.To this end,we construct a new emotion dictionary based on emoticons and modal for barrage sentiment analysis.Firstly,a new method of constructing the emotion dictionary applied to the barrage is proposed.On the basis of the previous methods of constructing emotional dictionary,this method adds a new method of constructing a dictionary of emoticons and a modal dictionary.Then a new method based on the barrage sentiment dictionary is proposed,which takes into account the influence of emoticons and modal on sentiment analysis when calculating the sentiment value of the barrage.Finally,the proposed method is compared with the existing methods.The experiment shows that the proposed method has better performance than the existing methods in term of recall rate,expression analysis,modal analysis and so on in barrage sentiment analysis.

关 键 词：弹幕 表情 语气 情感词典 情感分析 

分 类 号：TP391.5[自动化与计算机技术—计算机应用技术]