| The rapid development of the mobile Internet has promoted the reform and development of business models in many industries,and many traditional industries have gradually joined the Internet,including the catering industry.But it also brings problems at the same time of promoting development-information overload.Faced with these overloaded information,it is difficult for users to quickly find the information they want,and the effective use of information will be reduced.How to let users quickly find the restaurant they are interested in in these massive information is exactly the problem that the catering recommendation needs to solve.Since most of the catering activities are conducted in the form of groups,most of the current catering recommendation systems,such as the well-known Meituan reviews,recommend individual users.In response to this problem,this article focuses on group users ' Catering recommendations have been researched and improved on the basis of the group recommendation algorithm.A group catering recommendation algorithm(GCRA-BC)based on user check-in data is proposed,and the group catering recommendation system is designed and implemented,The details are as follows:(1)Group catering recommendation algorithm based on user's check-in data(GCRA-BC)By analyzing the user's preference information through the user's check-in data,I used a weighted hybrid fusion strategy for group recommendation during group preference fusion to optimize the group recommendation effect,and I assigned different users based on user activity.The weight and the activity are reflected by the number of check-in times.The more the number of check-in times,the more active the user.Then we calculate the degree of divergence among group members and select different preference fusion strategies for group recommendation based on the difference in divergence.Considering that the interaction behavior between group users will affect the group recommendation,I have further improved the group recommendation algorithm,incorporating the preference interaction between users into the above recommendation algorithm,and finally I am in the Yelp dataset The algorithm proposed in this paper is compared with other algorithms.The experimental results show that the GCRA-BC algorithm has higher accuracy.(2)Design and implementation of catering recommendation systemAfter verifying the effectiveness of the algorithm,this paper designs and implements a catering recommendation system for group users,applies the proposed improved algorithm to the recommendation module of the catering recommendation system,and recommends the group users through the system recommendation module.service.The system verifies that the improved algorithm proposed in this paper is also meaningful in real life.In this thesis,there are 31 figures,17 tables and 64 references. |
