Study On Personalized Recommendation Of Tourism Destination Based On Case-Based Reasoning

With the rapid development of China’s economy and technology,the increasing progress in Internet technology,and the gradual improvement of people’s living standard,people’s demand for self-help travel which can satisfy individual interests and preferences also increases.Because of huge amount of information filling the internet,users can hardly attain effective travel information which suits their need when they look through related travel information online.As a way to solve this problem,the tourism recommendation system has become the focus of attention of scholars,and how to recommend to the user the exact travel information that meets the needs of their personalized tourism need has become the key point of tourism research.The travel recommendation system usually has a cold start and the data sparse problem,and the recommended content is mainly based on tourism products,the user-based recommended information of some tourist destination systems is too single,not rich enough.This paper focuses on the user’s interest,and constructs case database based on travel notes,constructs the personalized recommendation model of the tourist destination based on case-based reasoning method,and provides users with information of the tourist destination which satisfies its individual needs and rich content,to a certain extent,solve the problem of data sparsity and cold start.The main work of this paper is as follows:(1)Construct travel destination model and case-user preference model which suits case-based reasoning method.From the "mafengwo" website to obtain information from users and travel notes to form the basic case database,and get the travel preferences of case-user by using the user’s travel preferences algorithm.The case-users are classified according to the type of tourism destination based on the user preference and the improved K-Means algorithm,forming various types of tourism destinations sub-case-database.When retrieving a case,the search efficiency is improved only by searching the sub-case-database of the user’s type of tourist destination.(2)The case attribute weights algorithm is constructed,and the weights of the case attributes are determined by the user’s evaluation of tourism elements in questionnaires and the weight algorithm of case attributes.(3)An improved trust algorithm is constructed,and the trust degree is introduced into the case-based reasoning of tourism destination personalized recommendation system,and a case similarity algorithm is constructed based on trust degree,improving the accuracy of recommendation results.(4)The method of case-based reasoning is applied to the personalized recommendation system of the tourist destination,the relevant algorithms are constructed and expressed by Mathematica software.The example is verified by user data,and the recommendation of travel destination and travel notes is realized initially.This paper successfully realizes the function of the recommendation system,making a personalized tourist destination and travel notes recommendations to tourists.The function was based on the combination of case-based reasoning technology,user interest and trust.The paper not only meets the needs of users,but also provide rich tourism information to users.The effectiveness and accuracy of the proposed algorithm are proved by case recommendation and empirical results.The research of this paper provides reference value for personalized recommendation system of the tourist destination to some degree.