Users leave traces when they participate in activities in a collaborative digital environment. These traces in return offer a clue for recommending a user by competency on a subject. This mechanism helps further collaboration because knowing the specialization of users ameliorates distributing tasks reasonably. In this article we propose a semantic model containing activity, trace of interaction and competency. Applying Logistic Regression, we exploit this model to offer recommendation of a competent person in a digial ecosystem.