Identifying widely disseminated scientific papers on social media

作者：

Highlights：

• XGB can effectively identify widely disseminated papers by incorporating literature-related and social media-related features.

• Literature-related features have greater effects on identifying the long-term disseminated papers than social media-related features.

• Most social media-related features play more important roles in identifying broadly mentioned literature.

• Papers whose disseminators consist of numerous academic-related Twitter users or the proportion of such users is small tend to be widespread on social media.

摘要

•XGB can effectively identify widely disseminated papers by incorporating literature-related and social media-related features.•Literature-related features have greater effects on identifying the long-term disseminated papers than social media-related features.•Most social media-related features play more important roles in identifying broadly mentioned literature.•Papers whose disseminators consist of numerous academic-related Twitter users or the proportion of such users is small tend to be widespread on social media.

论文关键词：Widely disseminated papers,Social media,Machine learning,Predictive model,Twitter

论文评审过程：Received 29 November 2021, Revised 28 February 2022, Accepted 11 April 2022, Available online 19 April 2022, Version of Record 19 April 2022.

论文官网地址：https://doi.org/10.1016/j.ipm.2022.102945