Exploiting Semantics of Special Characters to Strengthen Passwords

Special characters plays an important role in ensuring the complexity of passwords in text-based password authentication approaches. However, designing effective user guidance for creating secure and dependable passwords incorporating special characters remains an ongoing challenge. Hence, this paper presents our proposed method to construct a special character semantic dictionary based on the characteristics of special character segments. Specifically, our method considers the common special character semantic patterns and integrates scenarios involving Leet (substitution of letters with visually similar characters) to create a more comprehensive and logical special character semantic dictionary. We validate the soundness, and practicality of our special character semantic dictionary through experiments based on Zipf's law and password attack simulations. Results demonstrate that our proposed special character semantic dictionary significantly improves the effectiveness of password attack tools like PCFG<inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$_{v4}$</tex-math></inline-formula>, PassGAN and GENPass against passwords with special characters. Additionally, the special character blacklists derived from our dictionary enhance the strength of passwords with special characters to a notable extent. This research also introduces a novel approach to optimizing normal password dictionaries.