文献类型：期刊文献

中文题名：基于酒店评论大数据的游客评论主题挖掘与情感分析--以北京五星级酒店为例

英文题名：Topic Mining and Sentiment Analysis of Tourist Reviews Based on the Big Data of Hotel Reviews:A Case Study of Beijing Five-star Hotels

作者：马桂真[1];彭霞[1]

第一作者：马桂真

机构：[1]北京联合大学旅游学院,北京100101

第一机构：北京联合大学旅游学院

年份：2021

卷号：35

期号：2

起止页码：58-68

中文期刊名：北京联合大学学报

外文期刊名：Journal of Beijing Union University

基金：北京联合大学科研项目“大数据源分类特征与综合获取技术研究”(ZB10202005);北京市社会科学基金“北京市历史文化街区意象的游客感知研究”(17JDGLB002)。

语种：中文

中文关键词：酒店评论大数据;主题挖掘;情感分析

外文关键词：Hotel review big data;Text topic mining;Sentiment analysis

摘要：分析酒店评论数据可以挖掘游客的关注点、意见、建议、情感倾向等有价值的信息。结合对酒店评论数据进行主题挖掘和情感分析的交叉研究,提出一个包含数据采集、数据预处理、主题挖掘、情感倾向研究及可视化分析的集成框架。以Tripadvisor网站上北京地区50家五星级酒店的5万余条中文评论数据为研究对象,进行LDA主题挖掘,同时基于酒店领域扩充情感词典,判定评论文本三元情感极性,并在此基础上实现主题和情感的交叉分析。研究结果可降低潜在游客购买决策的风险,也为酒店管理者制定针对性的管理和营销策略提供重要参考依据。研究方法同样适用于景区及餐饮领域的在线评论数据分析,拓展评论大数据与自然语言处理技术在旅游业的应用范畴。
By analyzing hotel review data,it is very possible for one to mine valuable information such as tourists concerns,opinions,suggestions,and emotional tendencies.Based on the cross research of topic mining and sentiment analysis of hotel review data,this paper proposes an integrated framework including data collection,data preprocessing,topic mining,sentiment tendency research and visual analysis.LDA topic mining is carried out by taking more than 50000 reviews written in Chinese from 50 five-star hotels in Beijing on Tripadvisor as the research object.Meanwhile,on the basis of the hotel domain,the emotion dictionary is expanded to determine the ternary emotion polarity of the review text,and on this basis,the cross analysis of theme and emotion is realized.The results of this study can reduce the risk of potential tourists purchase decision and provide important reference for hotel managers to formulate targeted management and marketing strategies.The research method is also applicable to online review data analysis in scenic spots and catering fields,and conducive to expanding the application of review big data and natural language processing technology in tourism industry.