作者：Guoxiu Hea,b,∗, Zhikai Xuea, Zhuoren Jiangc, Yangyang Kangd, Star Zhaoe,fand Wei Lug
作者单位：
a: Faculty of Economics and Management, East China Normal University, Shanghai, 200062, China
b: Institute of AI for Education, East China Normal University, Shanghai, 200062, China
c: School of Public Affairs, Zhejiang University, Hangzhou, 310058, China
d: Alibaba Group, Hangzhou, 310058, China
e: Institute of Big Data (IBD), Fudan University, Shanghai, 200433, China
f: National Institute of Intelligent Evaluation and Governance, Fudan University, Shanghai, 200433, China
g: School of Information Management, Wuhan University, Wuhan, 430072, China
摘要：The potential impact of a paper is often quantified by how many citations it will receive. However, most commonly used models may underestimate the influence of newly published papers over time, and fail to encapsulate this dynamics of citation network into the graph. In this study, we construct hierarchical and heterogeneous graphs for target papers with an annual perspective. The constructed graphs can record the annual dynamics of target papers’ scientific context information. Then, a novel graph neural network, Hierarchical and Heterogeneous Contrastive Graph Learning Model (H2CGL), is proposed to incorporate heterogeneity and dynamics of the citation network. H2CGL separately aggregates the heterogeneous information for each year and prioritizes the highly-cited papers and relationships among references, citations, and the target paper. It then employs a weighted GIN to capture dynamics between heterogeneous subgraphs over years. Moreover, it leverages contrastive learning to make the graph representations more sensitive to potential citations. In particular, co-cited or co-citing papers of the target paper with large citation gaps are taken as hard negative samples, while randomly dropping low-cited papers could generate positive samples. Extensive experimental results on two scholarly datasets demonstrate that the proposed H2 CGL significantly outperforms a series of baseline approaches for both previously and freshly published papers. Additional analyses highlight the significance of the proposed modules. Our codes and settings have been released on Github. (https://github.com/ECNU-Text-Computing/H2CGL）
关键词：Impact prediction; Citation network; Hierarchical and heterogeneous graph; Graph neural network; Contrastive learning
来源期刊：Information Processing and Management
发表时间：2023-10-08
DOI：https://doi.org/10.1016/j.ipm.2023.103512
收录数据库：elsevier
