Development of a system for monitoring information security of social networks

Abstract

This paper addresses the design of a monitoring system intended to assess and maintain the information security of social network platforms and their users. The proposed system architecture combines automated content monitoring, account and bot-behavior analysis, malicious-link and phishing detection, and data-leakage tracking into a unified pipeline built on classical rule-based filters and machine-learning classifiers. The paper describes the functional modules of the system, the data flow between them, and the criteria used to prioritize alerts for human moderators. Limitations related to scalability, privacy constraints, and evasion by malicious actors are discussed, along with directions for further development.