基于情感分析的酒店顾客满意度评估

随着信息技术发展和社交平台多方面渗入,在线评论凭借真实客观的优点已成为商家和消费者的主要信息来源。结合TF-IDF、K-means算法获取酒店顾客满意度影响因素;采用基于监督学习的分类算法与百度自然语言处理API得出情感极性值;利用多元线性回归建立满意度评估模型,并将该模型应用于“2019哈尔滨冰上冰雪嘉年华”期间754家酒店的满意度分析中。研究结果表明:与现有酒店预订平台相比,该评价体系更加客观全面,顾客对此阶段的酒店总体满意度中等偏上,单维度分析中,总体感受是顾客最为关注的问题,但是酒店设施和卫生环境等因素也会影响酒店顾客满意度的提升。

HOTEL CUSTOMER SATISFACTION EVALUATION BASED ON SENTIMENT ANALYSIS

With the development of information technology and the infiltration of social platforms,online reviews have become the main source of information for businesses and consumers with their real and objective advantages.We combined TF-IDF and K-means algorithms to obtain the influencing factors of hotel customer satisfaction,then used the supervised learning-based classification algorithm and Baidu natural language processing API to obtain emotional polarity values.The multivariate linear regression was used to establish the satisfaction evaluation model.We applied the mode to the satisfaction analysis of 754 hotels during the 2019 Harbin Ice and Snow Carnival.The research results show that compared with the existing hotel reservation platform,our evaluation system is more objective and comprehensive.Customers overall satisfaction with the hotel at this stage is above average.In the one-dimensional analysis,the overall feeling is the most concerned issue for customers,but factors such as hotel facilities and sanitation environment can affect the improvement of hotel customer satisfaction.