Research On Personalized Recommendation Method For Enterprise Users’ Scientific And Technological Service Demand

In recent years,as an important part of modern service industry,science and technology service industry has a good trend of development.However,China’s science and technology service industry is in a transitional stage,the science and technology service in the science and technology service platform show exponential growth and diversified service types And the enterprise user faces the massive science and technology service resources to display the demand personalization,the user demand is difficult to express accurately the question.How to solve the problem of accurate matching between enterprise users and service resources in science and technology service platform,and how to design an efficient and accurate recommendation algorithm is the key to improve service efficiency and resource utilization ratio.In view of the above problems,this paper based on the perspective of enterprise users,analyzes the background information of enterprise users and quantifies the demand of enterprise users for scientific and technological services,a personalized recommendation model for enterprise science and technology service is constructed based on enterprise scientific and technological service demand and its special attributes.The aim is to respond quickly and accurately to the needs of enterprise users,and to recommend the scientific and technological service resources to meet the corresponding needs of enterprise users,so as to improve the accuracy of the recommendation results.The main research contents of this thesis include:Firstly,Based on the triangular fuzzy method,the paper presents a method for obtaining the scientific and technological service demand of enterprise users,which solves the problems of Fuzzy,abstract and difficult to express accurately.The background information of enterprise users will be identified,classified,cleaned and transformed,and the needs of enterprise users will be analyzed and the service needs of users in selecting service resources will be determined,based on this,three dimensions are proposed to describe the demand preference of enterprise users,and the triangular fuzzy number is used to measure the demand preference of enterprise users.Secondly,Constructs a personalized recommendation model based on enterprise users’ scientific and technological service demand,improves the utilization rate of scientific and technological service,and optimizes the resource allocation.Designs a personalized recommendation Algorithm for scientific and technological services based on the needs of enterprise users,considering the needs of both suppliers and demanders.By introducing the quality of service Qo S information with special attributes of sci-tech service,combining the concept relation of ontology and cosine similarity,the similarity of service type and Qo S is solved,and the comprehensive weighted method is used to calculate the comprehensive similarity,get the best service resources to meet the needs of enterprise users.Finally,Through the technology service personalized recommendation Algorithm(Eursts)based on the needs of enterprise users,service Qo S based Algorithm,content based Recommendation Algorithm(CB),different types of technology services are compared.The results show that the performance of the technology service personalized recommendation Algorithm(EURSTS)based on enterprise user demand is better than the other two algorithms in the accuracy rate and recall rate,which can improve the effect of the recommendation Algorithm and improve the recommendation accuracy.