Intelligent Recommendation Of Tourist Attractions Based On Association Rule Mining Algorithm

With the arrival of the smart era,tourism has also shifted from traditional information services to intelligence.The concept of "smart travel" was proposed to solve this problem.Smart Travel aims at realizing the real-time interaction between tourists and the Internet.It uses current high-tech methods to integrate a large number of tourism resources,and through intelligent optimization,strengthens the connection between users and travel information,and then provides tourists with suitable travel product needs.The recommendation system is one of the important ways to make use of the wisdom of tourism.The recommendation system can effectively solve the problem of "information overload" based on information such as user needs and behavior.With the increase of the amount of tourism data and the diversification of tourism information,the traditional classification recommendation can no longer meet the individual needs of users in tourism.Considering the shortcomings of the traditional tourism information service in the recommendation of tourist attractions,this paper proposes and designs a smart tourist attraction based on association rule mining algorithm under the background of smart tourism,using data mining technology and taking the tourism in Gansu Province as an example.Recommended system.Through the analysis of historical operation behaviors of tourists and current interactions with the system,visitors' interest is obtained.Then,the association rules are used to mine the hidden relationships among the attractions and recommend the attractions.The recommendation system not only meets the common needs of most tourists,but also can achieve personalized requirements based on individual interest of visitors,and the recommendation results have better accuracy and user satisfaction.The specific work of this article is as follows:First,the recommendation system is studied.By comparing the advantages and disadvantages of each recommendation method,Select association rules as the recommended method in this article.Then the user's operating behavior is acquired and a personalized recommendation model is established.Then,the FP_growth algorithm is analyzed by examples and combined with the algorithm defect and various factors in the recommendation of attractions and attractions,the user interest degree was introduced to weight the importance of the attractions,and the multi-minimum support weighted FP_growth algorithm was used as the mining algorithm recommended by the site.The algorithm takes into account that the user's preference for all attractions is not the same as the browsing probability.Multi-minimum support can simultaneously take into account the needs of various users,can effectively improve the quality of recommendations,and achieve personalized recommendations;weighted values are used to distinguish The user's interest in different attractions makes the recommendation more accurate.Finally,using the collected scenic spot data,the algorithm is used for simulation testing and mining analysis.The analysis shows that compared with the traditional association rules recommendation algorithm,the user has a better satisfaction with the recommendation results recommended by the algorithm,improves the user experience,Accuracy reaches 96.53%,nearly 4% higher than traditional recommendation and shows that the recommendation method has greatly improved the accuracy of the recommendation and can be improved.which can better to meet the needs of users.