The detection of community structure is probably one of the central trends in complex network emphasizing the complex internal organization of people, molecules or processes behind social, biological or computer net- works. . . The issue is to provide a network partition representative of this organization so that each community presumably gathers nodes sharing a common mission, purpose or property. Usually the identification is based on the difference between the connectivity density of the interior and the boundary of a community. Indeed, nodes sharing a common purpose or property are expected to interact closely. Although this rule appears mostly relevant, some fundamental scientific problems like disease module detection highlight the inability to determine significantly the communi- ties under this connectivity rule. The main reason is that the connectivity density is not correlated to a shared property or purpose. Therefore, an- other paradigm is required for properly formalize this issue in order to meaningfully detect these communities. In this article we study the com- munity formation from this new principle. Considering colors formally figures the shared properties, the issue is thus to maximize group of nodes with the same color within communities. We study this novel community framework by introducing new measurement called chromatic entropy as- sessing the quality of the community structure regarding this constraint. Next we propose an algorithm solving the community structure detection based on this new community formation paradigm.