Research On Hotel Customer Satisfaction Evaluation Based On Sentiment Analysis

With the development of information technology and the infiltration of social platforms,online hotel reservations came into being.Compared with traditional offline hotel booking,online booking attracts many consumers with its advantages of transparent prices and abundant choices.Hotel customer satisfaction is considered to be an important reference basis for measuring the vitality of the hotel,but with the vigorous development of the hotel industry,technology continues to empower consumers,consumers' voices continue to increase,customer needs cannot be met,and brand loyalty is gradually increasing reduce.The lack of information provided by hotel service providers,inaccurate publicity,and limitations on payment methods have limited the sustainable development of the hotel industry to a certain extent.Therefore,in the hotel booking transaction under the background of digital wave,how to enable consumers to obtain more authentic and reliable information,improve customer experience,improve customer satisfaction,and enhance customer choice tendencies to enhance customer loyalty,has become an urgent problem for the development of the hotel industry.Based on this background,this article is based on customer satisfaction theory,customer segmentation theory,demand hierarchy theory and STP marketing theory,and Ctrip.com hotel online review data as the research object to discuss how to use it in the context of gradually deepening the demand for quality tourism Information technology measures the problem of hotel customer satisfaction.First,TF-IDF,Word2 Vec,and K-means algorithms are used to obtain the influencing factors of hotel customer satisfaction.Secondly,Naive Bayes algorithm and sentiment analysis technology are used to obtain hotel customer satisfaction sentiment score.Finally,multiple linear regression is used to construct hotel customer satisfaction Degree evaluation model.In order to discuss the consumption needs of different consumer groups,subdivide the differentiated interest points of the customer group,obtain the hotel review characteristics of customers of different travel types by introducing statistical feature values,and use the Lasso algorithm to filter the factors affecting the satisfaction of customers of different travel types.The above model was applied to the customer satisfaction assessment of hotels in Central Street during the Harbin Ice and Snow Carnival in 2019,and the rationality of the model was verified.The research results show that,compared with the existing research,the evaluation system in this paper is more objective and comprehensive.The overall customer satisfaction of the hotel at this stage is above average,and the overall feeling in the single-dimensional analysis is the problem that customers are most concerned about,but factors such as hotel facilities and sanitary environment affect the improvement of hotel customer satisfaction;The overall customer satisfaction of hotels of different travel types is relatively high,but their focus is different.Hotel managers can use Internet thinking,combined with Internet word-of-mouth to develop a more humane marketing plan from the aspects of service level and facility construction,so as to effectively improve hotel customer satisfaction,promote the integration of the hotel industry and the Internet,at the same time,it conforms to the trend of the era of "Internet + Big Data".