Gated relational stacked denoising autoencoder with localized author embedding for global citation recommendation
作者：
Highlights：
• A novel neural model is proposed for global citation recommendation task. The proposed model is essentially a combination of three neural networks with latent variables.
• A novel author embedding method that uses limited localized neighbors is developed. The extensibility of this method is verified.
• The learning algorithm of the proposed neural model is detailedly derived.
• The time complexity for the learning algorithm is analyzed.
• Extensive experiments manifest the superiority of the proposed model.
摘要
•A novel neural model is proposed for global citation recommendation task. The proposed model is essentially a combination of three neural networks with latent variables.•A novel author embedding method that uses limited localized neighbors is developed. The extensibility of this method is verified.•The learning algorithm of the proposed neural model is detailedly derived.•The time complexity for the learning algorithm is analyzed.•Extensive experiments manifest the superiority of the proposed model.
论文关键词：Global citation recommendation,Stacked denoising autoencoder,Topic model,Machine learning,Deep learning
论文评审过程：Received 30 December 2020, Revised 30 April 2021, Accepted 4 June 2021, Available online 26 June 2021, Version of Record 29 June 2021.
论文官网地址：https://doi.org/10.1016/j.eswa.2021.115359
