Predicting and recommending collaborations: An author-, institution-, and country-level analysis
作者：
Highlights：
• We study dynamic aspect of scientific collaborations at author-, institution-, and country-levels.
• Eight link predictors are applied and evaluated.
• For the employed data set, denser networks yield more precise collaboration predictions.
• Neighbor-information-based predictors produce more similar outcomes than topology-based ones.
• Author-, institution-, and country-level collaborations are recommended.
摘要
•We study dynamic aspect of scientific collaborations at author-, institution-, and country-levels.•Eight link predictors are applied and evaluated.•For the employed data set, denser networks yield more precise collaboration predictions.•Neighbor-information-based predictors produce more similar outcomes than topology-based ones.•Author-, institution-, and country-level collaborations are recommended.
论文关键词：Collaboration,Link prediction,Coauthorship,Networks,Dynamics
论文评审过程：Received 13 September 2013, Revised 8 January 2014, Accepted 13 January 2014, Available online 4 February 2014.
论文官网地址：https://doi.org/10.1016/j.joi.2014.01.008
