Research On User Behavior And User Value Of Tourist Network Based On Web Data Mining

Recently, the concepts and applications of the "Big Data" and "Internet plus" are frequently mentioned in tourism industry. Rather than blindly following the trends, most tourism enterprises would keep clamming down and think about the more efficient and feasible management solutions. Under the competitive environment that resources is extremely uneven, whether it is OTA or traditional tourism enterprises, the majority of tourism enterprises are difficult to win in the market by high-quality resources competition, therefore, enhance the efficiency of operation to obtain the relative competition advantage is necessary. There are many ways to improve the operational efficiency, faced with the rich data in tourism industry and tourism enterprises, using the methods of data mining to solve the problem will be more rapid and efficient, also more practical.In this paper, first of all, sort and review the relevant literature at home and abroad, defines the relevant concepts of tourist network users'behavior and value. Secondly, with the help of some statistical methods, such as correlation analysis, factor analysis and some data mining methods like cluster analysis and survival analysis. Simultaneously, based on the basic data mining process to build a model of tourism data processing, it is constructed for the purpose of providing the logical analysis framework. In the end, using the model to study about the tourist network user behavior and network user value, and using mafengwo.cn as an example.Through this study, we found that there are some different features in tourist's network user behavior. Female users in UGC Websites has higher viscosity, and better than male users in retained performance. All users has different stability and volatility in different time dimensions. The regional distribution of users conform to the basic law of tourism and tourist flow. Combined with the cluster analysis results, we also found that the different categories are obviously different. And then, using the calculation of Customer Lifetime Value and retained function to get the precise quantification of the user's value. Finally, proposing for the improvement of operational efficiency by the knowledge and information obtained from the data mining, it can effectively enhance the efficiency of the operation in enterprises.