Abstract: 【Purpose/significance】The construction of accurate recommendation model of academic information based on user dynamic portrait is to meet the requirements of academic new media users for the precision,personalization and specialization of academic new media information resources,and it is also an effective way to improve the efficiency of academic information flow and value-added.【Method/process】This paper explores the information flow model of academic new media ,and further analyzes the user demand and hi⁃erarchical portrait of academic new media.On this basis,it reconstructs the steps of user portrait of academic new media,and constructs the recommendation model of academic new media information resources based on user dynamic profile【. Result/conclusion】The accu⁃
rate recommendation model of academic new media information based on user dynamic portrait can realize the accurate docking be⁃tween academic information resources and users,enhance user loyalty ,better serve the academic activities of scientific researchers【. In⁃novation/limitation】This paper analyzes and constructs the academic new media information recommendation model from the perspec⁃tive of theoretical framework,and will focus on the technical implementation and practical application of the model.