The observations among some animals and living bugs in colony caused our interest to end to understand the principle controling the phenomenon of the collective intelligence and its use in the domain of the artificial intelligence for the classification of the data. It exists, besides, of other algorithms in particular, of the algorithms inspired of the working of the immune system. Our contribution in the setting of this thesis will be therefore hybridization between two approaches: one bio-inspired of one social bug and the other of the immune system and this in order to solve a problem of classification