Mutually reinforced network embedding: An integrated approach to research paper recommendation
作者:
Highlights:
• A novel network embedding model is proposed for scientific paper recommendation.
• Our model integrate co-authorship, text and network with mutual reinforcement.
• A inductive method is developed to recommend reference papers for cold-start papers.
• Our method achieves state-of-art performance in HepTh and AAN datasets.
摘要
•A novel network embedding model is proposed for scientific paper recommendation.•Our model integrate co-authorship, text and network with mutual reinforcement.•A inductive method is developed to recommend reference papers for cold-start papers.•Our method achieves state-of-art performance in HepTh and AAN datasets.
论文关键词:Research paper recommendation,Network embedding,Mutually reinforced model
论文评审过程:Received 24 January 2022, Revised 16 May 2022, Accepted 16 May 2022, Available online 24 May 2022, Version of Record 27 May 2022.
论文官网地址:https://doi.org/10.1016/j.eswa.2022.117616