Generation of Believable Fake Integral Equations for Cyber Deception

Generation of Believable Fake Integral Equations for Cyber Deception

연구 분야: Strategies

학회: International Conference on Security and Privacy

Due to the increased sophistication of cyber attacks over the last few decades, there has been an exponential rise in data exfiltration incidents worldwide. Cyber attackers often remain undetected in enterprise networks for a significant amount of time (312 days for a zero-day attack), sufficient to compromise sensitive, business-critical, or mission-critical data such as customer data, scientific documents, trade secrets, proprietary research, etc. Protecting such information is paramount, necessitating techniques to secure it even after theft. Recent research suggests the automatic generation of fake documents to increase burden on the attacker, who needs to correctly identify the correct document from a set of legitimate and counterfeit documents. Numerous works have been proposed in the literature for automatically generating fake documents by manipulating text, equations, images, tables, circuit diagrams, etc. Among all types of equations present in the document, Integral equations are one of the core components of novel innovation and play a crucial role in a diverse set of domains such as risk management, stock market prediction, weather forecast, or even the prediction of natural disasters, etc. In this work, we introduce the Fake Integral Equation Generation Engine (FIEGE) to produce many plausible decoy documents by manipulating the integral equations present in the scientific document. The generated fake document can confuse and mislead the potential attackers if stolen, slow them down, waste their time and resources in identifying the original document, and thereby increase the overall cost of attack. The system employs algorithms for efficient and practical fake integral equation generation, while human evaluation studies demonstrate its effectiveness in deceiving experts. Future research directions are discussed to enhance the performance and resilience of the FIEGE system, making it a promising tool for improving cybersecurity in organizations.