Interest for computational trust and reputation models is on the rise. One of the most important aspects of these models is how they deal with information received from other individuals. More generally, the critical choice is how to represent and how to aggregate social evaluations. In this article, we make an analysis of the current approaches of representation and aggregation of social evaluations under the guidelines of a set of basic requirements. Then we present two different proposals of dealing with uncertainty in the context of the Repage system [J. Sabater, M. Paolucci, R. Conte, Repage: Reputation and image among limited autonomous partners, Journal of Artificial Societies and Social Simulation 9 (2). URL http://jasss.soc.surrey.ac.uk/9/2/3.html], a computational module for management of reputational information based on a cognitive model of imAGE, REPutation and their interplay already developed by the authors. We finally discuss these two proposals in the context of several examples.
