Abstract: 【Purpose/significance】The explosive growth of online academic resources makes information selection more and more diffi⁃ cult. Efficient and rapid selection of high-quality academic resources help to reduce the cost of information searching and improve the work efficiency of scholars.【Method/process】In this paper, a framework model of online academic resource selection is constructed by combining knowledge graph and user profiling.【Result/conclusion】The experiment shows that the framework can effectively improve the quality and efficiency of academic resources selection by combining the semantic network of online academic resources with the needs and preferences of scholars.【Innovation/limitation】This framework not only overcome the incomplete problem of knowledge graph, but also take into account the needs of individual and group scholar, and effectively improve the efficiency, quality and accu⁃ racy of academic resource selection. However, this study has not considered the time dimension in the knowledge graph, and the user profiling needs to be further explored.