摘要:In this paper, we propose an online accounts management method that uses a risk-based approach to minimize any risk of personal information leakage, while permitting users to reuse passwords. The proposed method clusters some accounts to reduce the number of passwords a user has to remember. This is achieved without increasing the risk value of any leaked personal information because the clustered accounts have similar attributes. Further, the proposed method suppresses any increase in risk by optimizing the cluster even if the password is shared by numerous accounts, and can reduce the number of passwords to approximately 30% - 40%. In evaluations conducted, 91% of participants accepted our method, indicating that it balances usability with security.