Social networks are large structures that depict social linkage between millions of actors. Social network analysis came out as a tool to study and monitor the patterning of such structures. One of the key problems handled to understand social networks evolving over time is the prediction of a future association between unlinked nodes, known as the link prediction problem. Traditional link prediction methods are designed to deal with social networks under a certain framework. Yet, data of such networks are usually noisy, missing and prone to observation errors causing distortions and likely unaccurate results. The belief function theory is an appealing framework for reasoning under uncertainty that permits to represent, quantify and manage imperfect evidence. The main aim of these works is to build models and approaches to deal with the problem of links prediction under uncertainty which describe the data provided in social network.