Dual Cloud Bibliographic Network Model for Citation Recommendation Systems

Finding relevant and appropriate papers for citation is becoming more and more difficult for researchers as a result of the exponential growth in the number of scientific papers. To tackle this problem, citation recommendations have been developed, aiming to provide a curated selection of relevant reference papers tailored to a specific manuscript. Citation recommendation algorithms could raise privacy concerns, particularly when working with sensitive or personal data. In this paper, we propose a dual cloud-based bibliographic network representation (BNR) model for effective recommendation and to ensure data protection. The BNR model in the proposed work deals with the relationships, paper-to-paper, paper-author, paper-key terms, etc. The system is evaluated with academic corpus namely ACM, AAN, and DBLP. The system proposes a privacy-preserved BNR citation recommendation approach that outperforms previous citation recommendation approaches by a wide margin.