Heterogeneous dynamical academic network for learning scientific impact propagation

作者：

Highlights：

• A novel and scalable heterogeneous dynamical graph learning framework.

• A fast weighted contextualized heterogeneous node sampling strategy.

• Scientific impact prediction with the knowledge from large-scaled academic network.

• Influential factors analysis for complex author impact prediction.

摘要

•A novel and scalable heterogeneous dynamical graph learning framework.•A fast weighted contextualized heterogeneous node sampling strategy.•Scientific impact prediction with the knowledge from large-scaled academic network.•Influential factors analysis for complex author impact prediction.

论文关键词：Scientific impact prediction,Heterogeneous information network,Graph neural network,Information diffusion,Science of science

论文评审过程：Received 29 June 2021, Revised 24 November 2021, Accepted 27 November 2021, Available online 12 December 2021, Version of Record 27 December 2021.

论文官网地址：https://doi.org/10.1016/j.knosys.2021.107839