Research On Personalized Recommendation System Of Electric Business Based On High-speed Railway Multimedia Service Platform

With the rapid development of China's high-speed railway network,high-speed rail has become the way that most people choose to travel with,according to the survey,people who often take the high-speed rail to travel most received higher education,have a certain economic capacity and social network,friends' advice on their shopping is very important.The formation of decision-making plays a key role.At the same time,the vigorous development of the Internet industry has made great changes in the traditional e-commerce model,the Internet industry and high-speed rail system combined to form a high-speed rail user characteristics based on the recommendation algorithm,and the algorithm used in e-commerce recommendation system is the mainpoint of this paper.Specific work is as follows:1.This paper studies and improves the personalized recommendation algorithm based on social network.The trustworthiness of traditional algorithm is jaccarad formula.The improved algorithm uses the similarity,reliability and intimacy ternary social information as the criterion of calculating trust degree.In calculating the user similarity,a time weighting factor is added to characterize the attenuation of the user's preferences over time.The trustworthy user reliability is calculated based on the degree of difference between the predicted score value and the actual score value by using the similarity degree of the trusted user and the target user to predict the score value of the target user to the common score product.The calculation of intimacy uses the improved jaccard formula to combine the degree of intimacy with reliability to achieve improved trust.2.The use of improved personalized recommendation algorithm to calculate the target users and social friends of the direct trust and indirect trust,and build social networks.Traversing social networks with Trust Walker's random walk model not only improves coverage,but also can compensate for the impact of the missing score on the forecast results by returning similar item ratings.Finally,the improved algorithm is tested with Epinions data set.The experimental results show that the improved algorithm improves the accuracy and coverage of the prediction compared with the traditional recommendation algorithm based on social network.3.In this paper,based on the network plan and system architecture of the project,this paper designs the e-commerce recommendation module of the project on the Android platform,which is used as a part of the high-speed mobile media service providing and operation platform project.And through the demand analysis and function analysis of the recommendation system,choose to use Springmvc + Spring + Hibernate three framework to complete the system design.And the improved recommendation algorithm of this paper is applied to the personalized recommendation subsystem of the server,and the personalized recommendation of the user is completed,and the feasibility of the proposed algorithm is verified.