Publication Date
Fall 2024
Degree Type
Master's Project
Degree Name
Master of Science in Computer Science (MSCS)
Department
Computer Science
First Advisor
Faranak Abri
Second Advisor
Genya Ishigaki
Third Advisor
William Andreopoulos
Keywords
Hidden Markov Models, Social Engineering Attacks, Markov Decision Process, Deep Reinforcement Learning, Systematic Literature Review.
Abstract
The advent of the internet has revolutionized communication and connectivity on a global scale. Now every computer is connected to the internet. Although this technological advancement has made human life easier, this has also led to an increase in sophisticated methods of exploitation. Social Engineering is such a prominent threat to the human community. Social engineering attackers manipulate the victim into giving away sensitive details. Understanding the dynamics of social engineering is crucial for developing measures to help individuals and organizations avoid falling prey to these deceptive tactics. Hence it is essential to understand the attackers. Thus gaining insight into the mindset of attackers becomes imperative to proactively thwart these insidious social engineering attacks. This study aims to analyze the attacker’s mindset by modeling it as a Markov Decision Process and training Reinforcement Learning agents using both Model-Based and Model-Free learning methods. The approach enables a comparison of the results obtained from each method to identify which is better suited to the problem at hand. This work also explores an approach that combines Hidden Markov Models (HMMs) with Markov Decision Processes (MDPs) to infer hidden state sequences when direct state information is unavailable. By mapping each MDP state to an HMM state and using optimal actions as observations, the structure of the HMM can be used to estimate the sequence of underlying states.
Recommended Citation
Sachithanandam, Bharkavi, "Reinforcement Learning and Hidden Markov Models for Simulating and Analyzing Social Engineering Attacks." (2024). Master's Projects. 1453.
https://scholarworks.sjsu.edu/etd_projects/1453
