关键词: 可信等级, 行为特征, 用户行为动态鉴别, 多因素鉴别

Abstract:

With the development of Internet technology, there have been more kinds of Internet applications, and the ways of users participated in the Internet become richer. It is the basis of network trusted system to grating trusted level of network identity. This paper proposed an evaluation model of the identity trusted level based on user behavior. Identity provider (IDP) according to the entity's identity information and behavior determines the highest trusted level of entities. When relying party (RP) needs to identify the entity identity, IDP combines the login environment with the highest confidence level to provide RP a dynamic trusted level. RP may manage the rights according to the dynamic trust level. The model using the ideas of classification, can meet the demand of application to the different trusted level of user identity, achieve reasonable security, and improve the efficiency of identity management and ease of use.

Key words: trusted level, behavior features, dynamic authentication, multiple factors to identify