Research And Implementation On Tourism Product Recommendation System Based On Case-based Reasoning

With the development of society and improvement of people's income,torism has become an impaortant part of people's life.Many people choose the product of travel agency to save the energy during the trip.Hower,Few travel agency provide perfect service of the recommendation of torism product which meet consumer's personalized needs,so building intelligent system which meet user's personalized needs is valueable.Cased-based reasoning,which is an important one of reasoning ways of artificial intelligence,takes advantag of experience of solving past problem to solve the current problem,and it provides ideas to build torism product recommendation system.Firstly,this paper analyse the content of travle case and the influencing factors of the user's decision,express the travel case with framework method.Then this paper introduces the index structure of the travel case.This paper uses the source of the tourist and the destination as the index of the travel case.A case clustering method can be used to devide cases which have the same tourist's source and destination into sub-cases which use it's typical case as index.In this way,A multi-level-index is built which can improve the efficiency of case retrieval.This paper uses K-nearest neighbor strategy in case retrieval.The concept of dependency and importace of the attributes in the rough set theory is used to calculate the weight of the attributes in the travel case.An improved attributes reduction method based on discrepancy matrix is used to reduce the redundant attributes in the travel case.In this way,the key attributes are stressed and the complexity of calculating the similarity of the case is reduced.Finally,the overall design,the detailed design and implementation of the tourism product recommendation system is described,an example is given to prove the usability of the system.