酒店在线评论数据的特征挖掘

PDF下载HTML

Title:

Feature mining based on online hotel review

作者:

秦海菲¹, 杜军平²

1. 楚雄师范学院 信息科学与技术学院, 云南 楚雄 675000;
2. 北京邮电大学 计算机学院, 北京 100876

Author(s):

QIN Haifei¹, DU Junping²

1. School of Information Science and Technology, Chuxiong Normal University, Chuxiong 675000, China;
2. School of Computer Science, Beijing University of Posts and Telecommunications, Beijing 100876, China

关键词:

酒店; 在线点评; 数据获取; 特征抽取; 特征挖掘; 聚类分析; 分类; 智能推荐

Keywords:

hotel; online review; data capture; feature extract; feature mining; cluster analysis; classification; intelligent recommendation

分类号:

TP391

DOI:

10.11992/tis.201806016

摘要:

论文以酒店在线评论数据为研究对象，对酒店在线评论数据的特征挖掘进行了研究。论文首先从酒店在线评论数据的获取出发，经过数据清洗、词性分析、特征抽取、指标确定、特征筛选、特征确定、特征校验几个环节，实现了酒店在线评论数据特征挖掘的目的。论文以词频为基础，融合了词性分析、聚类分析等方法，利用词频数（TF）、词频率（TF₁）、词频权重（TTW）、评论频率（DF）、逆文档频率（IDF）和TF1-IDF等指标对候选特征词进行降维，得出酒店在线评论数据的特征，并对特征词进行校验，完成了酒店在线评论数据的特征挖掘的过程。论文将为以评论为依据的客户分类、酒店分类、智能推荐奠定基础。

Abstract:

In this study, the feature mining of online hotel review data is investigated. First, online hotel reviews data were obtained. To mine features from the review data, data cleaning, part-of-speech analysis, feature extraction, index determination, feature selection, feature determination, feature checking were carried out. Based on the word frequency, integrating part-of-speech analysis, and cluster analysis, the word frequency (TF), word frequency rate (TF₁), word frequency weight (TTW), comment frequency (DF), inverse document frequency (IDF), and TF1-IDF of candidate feature words were applied to reduce dimension. The online hotel review data features were obtained, and then the feature words were verified. This paper will lay a solid foundation for the classification of hotels and customers and intelligent recommendation based on online reviews.