Research On Recommendation Systems Based On Dynamic Trust

The development of the Internet makes online social networks more and more important.The recommendation system plays an important role in the use of multiple coping strategies for online social networking.Multiple online social applications use the recommendation system algorithm to make the user group's expansion and personalized recommendations.The user's acceptance of the recommended results also affects the performance of the recommended algorithm.And trust,as a subjective concept,plays a very important role in the decision-making process of social networking participants.At the same time,its subjectivity makes it worth discussing the indirect transfer and mapping problems of trust.Therefore,trust is more important to the researchers concerned than other factors that need to be considered in the decision-making process.However,in the current trust recommendation system,the critical issues of some trust communication processes have not been effectively resolved,which makes it impossible for some trust models to give accurate and high-quality recommendations.In the development of the trust recommendation system,many scholars have proposed different trust mapping and communication methods for this problem,and have built different trust communication models.They want to use the proposed models to make more personalized speculation,enabling participants to get more accurate trust recommendations.The recommendation system,which combines trust,is an important recommendation system application based on social networking,which combines the trust relationship between users to recommend the project to the user,in order to be more accurate and more consistent with the recommended results of user preferences.However,previous studies generally assumed that the trust value between users was fixed,unable to respond to the dynamic changes of user trust and preferences,and then influence the recommendation effect.In fact,when the user is recommended,the experience user's trust in the requester will increase and the other will decline when the actual evaluation is higher than the psychological expectation.Aiming at this problem,and focusing on the dynamic nature of trust change process and trust in users,a combination of user trust enhancement method is proposed.Using the method of strengthening learning method,the study method is used to improve the value of trust in the recommended process,and puts forward a polynomial level algorithm to calculate the value and recommendation of the user,and can motivate the recommender to learn the user's preferences and ensure that the trust of the requester is always high.The experiment shows that the method can quickly respond to the dynamic changes of user preferences,and when it is applied to the recommendation system,it can provide more timely and accurate recommendations for users compared to other methods.