Community structures evaluation in complex networks: A descriptive approach


DAO Vinh Loc1,2, BOTHOREL Cécile1,2, LENCA Philippe1,2

Type de document

Communication dans une conférence avec acte


NetSci-X 2017 : 3rd International Winter School and Conference on Network Science, 15-18 january 2017, Tel Aviv, Israel, 2017, vol. Springer Proceedings in Complexity , pp. 11-19




Evaluating a network partition just only via conventional quality metrics - such as modularity, conductance or normalized mutual of information - is usually insufficient. Indeed, global quality scores of a network partition or its clusters do not provide many ideas about their structural characteristics. Furthermore, quality metrics often fail to reach an agreement especially in networks whose modular structures are not very obvious. Evaluating the goodness of network partitions in function of desired structural properties is still a challenge. Here, we propose a methodology that allows one to expose structural information of clusters in a network partition in a comprehensive way, thus eventually helps one to compare communities identified by different community detection methods. This descriptive approach also helps to clarify the composition of communities in real-world networks. The methodology hence bring us a step closer to the understanding of modular structures in complex networks.


1 : LUSSI(TB) - Dépt. Logique des Usages, Sciences Sociales et de l'Information (Institut Mines-Télécom-Télécom Bretagne-UBL)
2 : Lab-STICC(TB) - Laboratoire en sciences et technologies de l'information, de la communication et de la connaissance (UMR CNRS 6285 - Télécom Bretagne - Université de Bretagne Occidentale - Université de Bretagne Sud - ENSTA Bretagne - Ecole Nationale d'ingénieurs de Brest)

Mots clés

Network Science, Community Detection, Community Structure, Network Partition, Understanding of modular structure, Descriptive KPI



retour à la liste des publications
  • Institut Carnot Télécom & Société numérique
  • Université Bretagne Loire
  • Institut Mines-Télécom