Recently, the INFORMS Journal on Computing, a journal under the International Society for Operations Research and Management Sciences (INFORMS), published online the latest research findings by Geng Ruibin, an associate professor and doctoral supervisor at the School of Management, Northwestern Polytechnical University, and his co-authors. The paper, titled "Eliminating Social Popularity Bias in Recommendation: Causal Inference-Based Social Graph Neural Networks", is the communication authorship of Geng Ruibin. This research focuses on the issue of "social popularity bias" in social recommendation systems. Traditional recommendation algorithms tend to favor popular content, and in the context of social networks, this bias is further amplified by "social influence", leading to recommendation results that are concentrated on content followed by opinion leaders or high-influence users, thereby undermining information diversity and fairness.

Addressing this critical issue, the research team innovatively: constructed a Causal Graph model, systematically revealing the interference mechanism of "item popularity" and "social influence" as confounding factors on user preference learning; proposed a social graph neural network model (CISGNN) based on "Backdoor Adjustment" and "Counterfactual Reasoning", achieving the fundamental elimination of social popularity bias; validated the model's effectiveness through four real-world social datasets (Ciao, Epinions, Yelp-Philadelphia, Yelp-Tucson), and the results showed that CISGNN significantly outperforms the existing optimal model in terms of recommendation accuracy (HR, NDCG) and diversity metrics (Novelty, NicheRate). The framework proposed in this study not only promotes the integration of causal inference and recommendation systems at the algorithmic level, but also provides important managerial insights for platform algorithm governance, fairness-based recommendation, and the design of long-tail content incentive mechanisms.

"INFORMS Journal on Computing" is one of the top international journals in the field of management science and operations research. It is also ranked among the 24 most influential core journals for business schools globally (UTD-24) by the University of Texas at Dallas (UTD) School of Business. It holds a high academic reputation in the interdisciplinary research area of artificial intelligence and management decision-making.

[About the Author]

Geng Ruibin is an associate professor and doctoral supervisor at the School of Management, Northwestern Polytechnical University. His research interests include digital economy and digital governance, data sharing and information security, and AI-driven decision analysis. He has published over 10 papers in internationally renowned journals such as Information Systems Research (UTD-24), INFORMS Journal on Computing (UTD-24), European Journal of Operational Research, and Information & Management. His research has been cited over 1047 times by SCI, with the highest citation reaching 655 times. One of his papers was selected as an ESI Highly Cited Paper in 2024. He has led over 10 projects, including the National Natural Science Foundation of China's Youth Program, General Program, and the Ministry of Education's Humanities and Social Sciences Youth Fund. He has received honors such as the Third Prize for Excellent Achievements in Humanities and Social Sciences Research in Universities of Shaanxi Province (2021), the Emerald Outstanding Paper Award (2021), and the Nomination Award for Excellent Doctoral Dissertation in Zhejiang Province (2017).

(Content/Geng Ruibin Review/Jia Ming)