한국어 PII 기반의 향상된 비밀번호 취약성 평가모델

초록

Passwords are an essential means of protecting personal information in modern society. Many people tend to create passwords based on personally identifiable information (PII). However, such passwords can be easily guessed if the PII is leaked or hacked, making them vulnerable to security threats. To address this issue, it is necessary to assess the vulnerability of passwords generated using personal information. Existing password vulnerability assessment models do not take into account the inclusion of personal information in passwords and mainly evaluate vulnerabilities based on the English alphabet, which may not be suitable for non-English-speaking users. Therefore, this paper proposes a model that analyzes whether a user's password contains personally identifiable information and evaluates its vulnerability. Since most personal information, such as names, dates of birth, and phone numbers, is typically used when creating passwords, this model considers patterns that mix these elements with passwords, the length of the password, the complexity of character combinations, and the inclusion of personal information. To evaluate the model’s effectiveness, we conducted accuracy experiments, achieving an accuracy rate of 92.9%. Therefore, it is expected that using this model to assess password vulnerability when personal information is included in password creation could enhance security.

키워드