Research On Personalized Tourist Attractions Recommendation Based On Behavior Analysis

As the mobile Internet technology has been more and more popular, the mobile travel applications has been developed rapidly. Travel enthusiasts can stay at home, plan a journey and start it anytime. But in this process, they will encounter the "information overload" problem which can not be avoided and results that the travel plan has to be postponed or even be canceled. Personalized tourist attractions recommendation is an effective solution to the "information overload" problem of mobile travel app, which can make the Tourism Industry, the sunrise industry, achieve sustained and healthy development. It should be noted that the mobile travel applications contain food, housing, transportation, travel, shopping and five functional modules, and improving the defect of tourism function module is the research purpose of the paper. In addition, this module has the intelligent guide features, so it is replaced by the tourist resort intelligent guide system to clarify the research object in the paper.The purpose of this study is to improve the classic Apriori association rules algorithm which is used by personalized tourist attractions recommendation, construct the personalized scenic recommendation architecture based on the behavior-preference model and obtain the recommended method of personalized tours. That can enhance the accuracy of tourist attractions recommendation, improve the experience of visitors, which makes the mobile travel application is important for the development of tourism industry and the tourists travel. This paper will use comparative analysis method, the model constructing and the actual data verification method to complete this research.Firstly, the research background and significance are elaborated in detail. Further the paper compares and analyzes the research and development circumstances from data mining, intelligent guide system and personalized recommendation technology based on tourist behavior separately. And the defect of personalized tourist attractions recommendation in intelligent guide system are identified. Meanwhile elaborating data mining algorithm, intelligent guide system and personalized recommendation technology is the theoretical basis for research purposes of the paper.Secondly, analyzing the limitations of the Apriori algorithm used in intelligent guide system, considering the actual circumstances of each resort intelligent guide system effects of different factors contribute value, proposing the improved Apriori algorithm based on the behavior of visitors, this algorithm is combined with the establishment of tourists' behavior-preference personalized recommendation model. Validating of the model further, analyzing the personalized tourist attractions recommenddation architecture and excluding invalid data mining lift the level of personalized tourist attractions recommendation.Lastly, based on the tourists' behavior-preference model, experiments are carried on to analyze characteristic scenic recommendation architecture based on the actual data mining. And the results of the experiments help to analyze the target groups of tourists, tourism services and system interface thoroughly. Thus this paper give improved personalized tourist attractions recommendation method finally.