Application Of Personalized Hybrid Recommendation Algorithm In Tourism

With the advent of the 21st century, we have truly entered the Internet era, Internet in constantly changing people’s life. How much faster better for the user to provide high quality service has become the key, mobile app is developing rapidly under such a background, APP carrying and convenient mobile service, good app gradually become the people daily life essential part of, how will mobile app really into the vast users are large in life, is an important problem faced by all developers.On the other hand, China’s rapid economic development, people’s standard of living increased, more and more people willing to out of the circle of their own lives, to domestic other scenic places or even abroad to tourism, then in order to be able to in the journey can be comfortable journey of body and mind, a fit tourists need their own personalized travel service is all the travelers are badly in need of, then how can we provide the paste and tourists of different needs of their own service? It has also become the problem that tourism service providers want to solve.In order to solve the above two problems, mobile personalized travel recommendation system will become the solution of their most suitable scheme, this recommendation system can for tourists in the choice of season of tourism service can save a lot of time, bring convenience and fast, but also for the developers to save a large amount of marketing costs, but also for the tourist attractions bring huge profits, to promote the development of China’s tourism industry. Therefore, mobile personalized recommendation system not only has theoretical research value but also has a very high commercial value.In this paper, in the traditional research of recommendation system found that single recommendation system on the recommendation effectiveness or efficiency has various shortcomings, so we on the basis of this, research and analysis of the advantages and disadvantages of different recommendation algorithms, combined with the advantages of mobile terminal, based on feature scene hybrid recommendation algorithm. We first use the user of the mobile terminal operation and record the behavior characteristics, according to these characteristics based on the user preference model was constructed based on the content of the recommendation algorithm. At the same time, sing the tag recommendation algorithm using scene spots information construction of situation model. Then the user preference model began to recommend for the first time, and recommended the results as input, use scenario models for secondary recommendation, the recommendation results is finally get personalized recommendation results (sparse user context model in major, whereas preference model in major).Then we design the detailed algorithm experiments, the algorithm of single recommendation results and our hybrid recommendation compared. Experimental results show that this hybrid recommendation algorithm is recommended for the higher accuracy, and to recommend a high efficiency, so as to prove the effectiveness of the hybrid algorithm.