| With the prosperity of tourism industry and the progress of Internet business model,the market size of the online tourism is also expanding,and online hotel booking business is an important part of the online tourism.In order to improve the competitiveness of enterprises and optimize operational strategies,more and more enterprises are investing in the study of user behavior.With the development of the online tourism,online travel companies need to adopt more advanced means to analyze the website user behavior from more angles.At present,the research on user reservation behavior in the online tourism is mainly aimed at active users,and there are relatively few researches and applications on waking up sleeping users.Therefore,this paper mainly analyzes the behavior of inactive users,groups different users by clustering.Then wake up half of the users by sending messages and coupons.Then analyze the test results,evaluation of the promotion efficiency and user value of users in different groups.This paper first expounds the development of online tourism,the development of user behavior in tourism and the problems encountered at home and in the world,and the research status of clustering data mining methods in the field of user classification.Then,nine factors related to the behavior of the user were selected to study the inactive users who booked the hotel website online.With the true user order data of Company X,the K-means clustering algorithm is used to classify users,and finally five types of users are obtained.Next,this paper designs a promotion plan based on the ABTest experimental idea.Send messages and coupons directly to test group users and record the order data of the users over a period of time.According to the above experimental results,the user of different groups is evaluated by using the analytic hierarchy process,such as the user wake-up rate,the income before coupon,the cost to increase a roomnight,and roomnights.Finally obtains the user value of each type of user and provide suggestions for user selection in users wake-up projects.Finally,this paper summarizes the shortcomings in the research process and looks forward to the future research direction.User behavior analysis and user classification are very necessary for online travel enterprises.The research methods and final conclusions in this paper can provide support for online tourism enterprises in the research of waking up sleeping users,and it has certain guiding significance for the decision-making in the operation of online tourism enterprises. |
