access control based on trail inference

Professionals are constantly seeking qualification and consequently increasing their knowledge in their area of expertise. Thus, it is interesting to develop a computer system that knows its users and their work history. Using this information, even in the case of professional role change, the system could allow the renewed authorization for activities, based on previously authorized use. This article proposes a model for user access control that is embedded in a context-aware environment. The model applies the concept of trails to manage access control, recording activities usage in contexts and applying this history as a criterion to grant new accesses. Despite the fact that previous related research works consider contexts, none of them uses the concept of trails. Hence, the main contribution of this work is the use of a new access control criterion, namely, the history of previous accesses (trails). A prototype was implemented and applied in an evaluation based on scenarios. The results demonstrate the feasibility of the proposal, allowing for access control systems to use an alternative way to support access rights.