The Design And Implementation Of Tourist Attraction Recommender System Based On Collaborative Filtering

With the development of Internet technology, Internet has extended its influence to every field and aspect in our life. As the most dynamic "sun-rising industry", the tourism industry has also been swept into such a big wave of the Internet. In such an era of "information overload", how to find users'favorite in the mass of tourist attractions becomes a problem. Personalized Recommendation comes into being. It can recommend information to users which meet their needs. Collaborative filtering (CF) recommendation technology is one of the core technologies. Although the CF recommendation technology is widely and successfully used, the traditional algorithm still exist some problems, such as cold start, data sparse, changes of user interest, low recommendation accuracy and other issues.The main content of this paper is to help users to extricate themselves from the massive movie information. The system can automatically recommend the interested movies to the users. By studying of traditional CF algorithm, this paper improves the traditional CF algorithm based on the low accuracy of the recommendation and the changes of user interest. A Collaborative filtering recommendation algorithm based on multi-attribute rating which considering the changes of user interest is proposed. First, multi-attribute rating of the project was introduced because of the single overall rating of traditional collaborative filtering algorithms, which result in lacking of the multifaceted understanding of user interest, so that the recommendation was not accurate enough. The users'overall rating of the project would be computed by the rating of each attribute. Secondly, a time function based on the forgetting rules of Hermann Ebbinghaus is introduced to solve the problems caused by changes of user interest. The time function was adopted as the weight to the rating of the item to ensure the accuracy of recommendation when calculating the similarity of users.Recommendation system of tourist attractions is based on personalized recommendation system. First, this paper analyzed the system requirements. Secondly, the functions and database of the system were designed in detail. Finally, some main functions of the system are realized by using the object-oriented programming technology. The system will provide users services of recommending tourist attractions when they are using the system.