Research On Tourism E-commerce Oriented Data Mining

With the rapid development of economy and the great improvement of thestandard of people’s life, tourism industry is growing fast, tourism e-commerceemerged. There is also growing competition between firms. How to integrateresources, to provide satisfactory service for the tourists, and form a stable Customersources becomes focus of competition. A common problem facing today is thetourism e-commerce systems has collected a lot of data, but do not get real valuableinformation. The application of data mining technology to the tourism e-commerce,through mining the related data, to provide personalized service to tourists hasbecome an effective method to improve their competitiveness.This paper introduces the basic concept and problems and development directionof tourism e-commerce, and analyzed the recommendation technology in the systemof tourism e-commerce. Detailed several kinds of recommendation algorithms.This paper mainly studies the algorithm of data mining and its application in thetourism e-commerce personalized recommendation system. Based on the research ofpersonalized recommendation technology, we constructed a tourism e-commercerecommendation method. First use the user’s browsing behavior data to establishuser-product interest degree matrix, and then use the clustering method to cluster theuser, and then mining the association rules in the same kind of user, finally onlinefiltering the association rules, and recommended the result to the user.Aiming at the problems of the user-product evaluation information is not easy toget, we propose use user-product interest degree matrix to replace the user-productevaluation matrix, and puts forward the method to create the user-product interestdegree matrix model to solve the problem.We improved the association rules algorithm. A new algorithm based ondecomposed transaction matrix is proposed. It can reduce the times of comparison andavoids scanning the database many times. Experiments have shown that it doesaccurately increased efficiency.