Detecting and Preventing Clone Profile Attacks in Bidirectional Social Networks: A Novel Graph-Based Technique

Counts differ because each database indexes a different set of publications. We treat OpenAlex as the canonical count; Google Scholar is not shown (no API, and crawling it violates its ToS).

Profile cloning attacks are a prevalent cybercrime in Online Social Networks (OSNs), in which attackers impersonate legitimate users to conduct social, financial, and psychological abuse. A common issue with this cybercrime is that existing prevention and detection methods often fail to capture relational inconsistencies inherent in cloned accounts. To address this challenge, this paper proposes a graph-theory-based Clone Security System (CSS) for bidirectional OSNs. The proposed framework models social networks as graphs and employs two dedicated algorithms to detect and prevent profile cloning attacks by analyzing structural and relational similarities between user profiles. Experimental evaluation on social network datasets demonstrates that the proposed approach effectively identifies cloned profiles with improved accuracy and reduced false-positive rates compared with conventional methods. The results indicate that graph-based analysis provides a robust and scalable solution for enhancing user security and trust in OSNs; however, the proposed work is limited to bidirectional graph-based social networks.