摘要
伴随着中国航空管制的放松和自由化,中国民航市场竞争愈加激烈,因此航空公司针对性改进服务显得尤为重要。通过分析微博评论数据,深入了解航空旅客对航空服务需求的关注点,以提供有价值的见解帮助航空公司改进服务。首先,采用网络爬虫方法收集大量微博评论数据进行文本数据预处理;然后,分析基础统计,包括词频统计和词云生成,以识别旅客的核心关注点;最后,分析情感,通过情感得分判定评论的情感极性,使用隐含狄利克雷分布(LDA)主题模型对消极评论文本进行主题分析,以揭示航空服务需求的不同维度。实验结果表明,航空旅客最关注的问题包括售后服务、行李限额、服务水平等。航空公司需要时刻关注细节差异化服务,提高服务质量,以满足旅客个性化需求。这将为航空公司提供更有价值的指导,以改进服务,提高客户满意度。
With the liberalization of China’s aviation management control,competition in the Chinese civil aviation market becomes increasingly intense.Therefore,it is crucial for airlines to have their services improved in a targeted manner.By analyzing Weibo comment data,an in-depth understanding of the focus areas of air travelers’demands for air services was obtained,which provided valuable insights for airlines to enhance their services.A large amount of Weibo comment data was collected using Web scraping methods,followed by text data preprocessing.Then,basic statistical analysis,including word frequency statistics and word cloud generation,was conducted to identify passengers’core regions of interest.Subsequently,sentiment analysis was performed to determine the sentiment polarity of comments through sentiment scores.Using the Latent Dirichlet Allocation(LDA)topic model,negative comment texts were subjected to topic analysis to reveal different dimensions of air service demands.Experimental results show that air passengers are most concerned about after-sales service,baggage allowances,and service levels.Airlines need to pay constant attention to differentiating their services and improving service quality to meet passengers’personalized needs,which will provide airlines with more valuable guidance to enhance services and increase customer satisfaction.
作者
孙天笑
田勇
陈锦辉
黄钲翔
SUN Tianxiao;TIAN Yong;CHEN Jinhui;HUANG Zhengxiang(College of Civil Aviation,Nanjing University of Aeronautics and Astronautics,Nanjing Jiangsu 211106,China)
出处
《计算机应用》
CSCD
北大核心
2024年第S01期49-53,共5页
journal of Computer Applications
基金
南京航空航天大学科研与实践创新计划项目(xcxih20220731)。
关键词
文本挖掘
服务需求
差异化竞争
情感分析
自然语言处理
text mining
service demand
diversified competition
sentiment analysis
natural language processing
