PersonalityGate: A general plug-and-play GNN gate to enhance cascade prediction with personality recognition task
作者：
Highlights：
• A multitask GNN framework is proposed for joint cascade and personality prediction.
• A plug-and-play GNN gate is designed to extract personality traits from cascade.
• Extracted personality traits enhance GNN-based cascade prediction models.
• Empirical study corroborates the intuitive assumption of the framework.
• Experimental results on real world datasets validate the framework’s usefulness.
摘要
•A multitask GNN framework is proposed for joint cascade and personality prediction.•A plug-and-play GNN gate is designed to extract personality traits from cascade.•Extracted personality traits enhance GNN-based cascade prediction models.•Empirical study corroborates the intuitive assumption of the framework.•Experimental results on real world datasets validate the framework’s usefulness.
论文关键词：Cascade prediction,Personality recognition,Graph neural network,Multitask learning,Big five model
论文评审过程：Received 22 August 2021, Revised 16 January 2022, Accepted 25 April 2022, Available online 5 May 2022, Version of Record 16 May 2022.
论文官网地址：https://doi.org/10.1016/j.eswa.2022.117381
