The detection of community structure is probably one of the hottest trends in complex network research as it reveals the internal organization of people, molecules or processes behind social, biological or computer networks\dots 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 communities under this connectivity rule. The main reason is that the connectivity density is not correlated to a shared property or purpose. Therefore, another paradigm is required for properly formalize this issue in order to meaningfully detect these communities. In this article we study the community 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 \emph{chromarity} assessing 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.

翻译：社区结构检测可能是复杂网络研究中最热门的趋势之一，因为它揭示了社会、生物或计算机网络背后的人、分子或过程的内部组织。问题在于提供一种能够代表这种组织的网络划分，使得每个社区可能聚集了共享共同使命、目标或属性的节点。通常，这种识别基于社区内部与边界连接密度的差异。确实，共享共同目标或属性的节点预期会密切交互。尽管这一规则似乎大多相关，但一些基础科学问题（如疾病模块检测）凸显了在这种连接规则下无法显著确定社区的局限性。主要原因是连接密度与共享属性或目标并不相关。因此，需要另一种范式来恰当形式化这一问题，以便有意义地检测这些社区。在本文中，我们基于这一新原则研究社区形成。考虑到颜色形式上代表了共享属性，问题在于最大化社区内具有相同颜色的节点组。我们通过引入一种名为\emph{色性}的新度量来研究这一新颖的社区框架，该度量评估了在此约束下社区结构的质量。接下来，我们提出了一种基于这一新社区形成范式解决社区结构检测的算法。