We present in this paper a new approach for constructing implicit social networks from electronic sources, like emails, SMS and phone calls. The main novelty is the use of Social Interaction Analysis to assist end-users in their communication needs. We discuss how a social proximity can be calculated between two persons in a social network and show how a weighted, directed network is constructed based on interactions between people. After a description of the architecture of the framework, we show how a contextual, weighted social network can help in automatically finding the contact with highest probability to know the whereabouts of a person who is currently not reachable. The implemented social helper application uses a specific contextual view of the social network, exploiting only interactions that occurred in the last 48 hours.