WordMarkov: A New Password Probability Model of Semantics
Jiahong Xie, Haibo Cheng, Rong Zhu, Ping Wang, Kaitai Liang
Abstract
To date there are few researches on the semantic information of passwords, which leaves a gap preventing us from fully understanding the passwords characteristic and security. We propose a new password probability model for semantic information based on Markov Chain with both generalization and accuracy, called WordMarkov, that can capture the semantic essence of password samples. Further, we evaluate our design via password guessing attacks, on six real-world datasets, and we show that WordMarkov obtains 24.29%–67.37% improvement over the state-of-the-art password probability models. Even more surprising is that WordMarkov achieves 75.35%–96.34% attack improvement on "long" passwords, indicating the importance of semantic parts in long passwords.