| With the rapid development of information technology,e-commerce platforms can achieve deep integration of logistics,capital flow and information flow through the application of information technology.At present,the recommendation technology can analyze the user's interest preferences and implement personalized recommendation for customers.It plays an important role in the e-commerce platform and begins to play a role in logistics and other platforms.Improving the optimization recommendation technology algorithm and introducing it into logistics,industrial products and other platforms can help enterprises.Neighbor-based collaborative filtering recommendation is a very common recommendation method in information filtering technology of current recommendation systems.There are two shortcomings in the Bhattacharyya coefficient collaborative filtering algorithm.This thesis researches and improves the shortcomings of the algorithm.This paper makes two improvements on the basis of the Bhattacharyya coefficient coefficient collaborative filtering algorithm: Considering the situation when the number of two user rating items is different,the probability density of the user rating is processed to achieve the purpose of making full use of all the user's rating information.Consider the user's positive scoring preference for the project,further screen the data,and reduce the system's computational load.The experimental results show that through these two improvements,the improved Bhattacharyya coefficient collaborative filtering algorithm can obtain better recommendation results on the data set and improve the accuracy of the recommendation.In this thesis,the improved Bhattacharyya coefficient collaborative filtering algorithm is introduced into the B2 B procurement platform of A industrial products,and a framework for recommending technologies for this platform is designed to provide a reference for related platforms to use recommending technologies. |
PDF Full Text Request