In the present paper, we introduce an extension of the conceptual graph model suitable to the representation of data which are modelized using fuzzy sets. We extend the specialization relation of the conceptual graph model to fuzzy conceptual graphs. Lastly we introduce a new way of comparing conceptual graphs, using the idea that a graph may be compatible with another graph with a given degree d, which allows to make more flexible comparisons of fuzzy conceptual graphs. This work takes place within a project that aims at building a tool for the analysis of microbial risks in food products