Research On The Development Of Commodity Recommendation System Using JAVA

Nowadays,with the rapid development of computer technology and the network,the quantity of network information is soaring and a growing number of information appears in the network,but valuable information is difficult to be captured instantly,how to extract the useful information through vast quantities of information is becoming a very important issue.E-commerce recommendation system is one of the main ways to solve the information overload issue,e-commerce recommendation system brings a lot of benefits to users ans helps users find the goods they need,improves the marketing effect.In the e-commerce recommendation system,the user's preferences information can be obtained by the recommended algorithm of the website and then speculate the goods that the user maybe like,and finally recommend to them.And this method is widely used now.However,so far there are a lot of problems worthy of our study,in this paper,on the one hand,we have improved some of the algorithm,on the other hand,according to the previous recommendation algorithm,we put forward a new recommendation algorithm,Finally,the algorithm is evaluated.In our work,the collaborative filtering recommendation of the goods,collaborative filtering recommendation of the user and the recommendation based on the Slopone algorithm are combined,meanwhile,according to the previous recommendation algorithm,we introduce the Markov logical network model into the recommended algorithm,construct the recommendation model,and evaluate the algorithm through the experimental results.The system uses J2EE architecture technology and B/S-based three-tier system interface model to build the overall framework,completed the design of the functional modules of the telecom website.