Abstract:In recent years,with the development of knowledge graph technology,based on the construction of a new technology identification model for the solar photovoltaic industry,the time dimension and field dimension of technological evolution are extracted based on the scientific knowledge graph generated by related software,which has strong research significance.At the same time,equipment selection and material procurement act make it complicated and time-consuming for users to select target information from a large amount of information.The recommendation system (RS) aims to solve the problem of information explosion by finding a set of items that satisfy the user’s personalized interests,but it still faces problems such as cold start.Using the knowledge graph to analyze and build a relationship model between users and items can effectively improve the accuracy of personalized recommendations.This article focuses on how to use knowledge graphs to make accurate and interpretable recommendations,and finally puts forward possible research directions in this field.