Advances in Network Clustering and Blockmodeling

2Bibliometric Analyses of the Network Clustering Literature

Vladimir Batagelj^1,2,5, Anuška Ferligoj^3,5 and Patrick Doreian^3,4

¹IMFM Ljubljana

²IAM, University of Primorska, Koper

³FDV, University of Ljubljana

⁴University of Pittsburgh

⁵NRU HSE Moscow

2.1 Introduction

Partitioning networks is performed in many disciplines, as is evidenced by the chapters of this book. The data we consider here are from the network clustering literature. Our focus here is the large set of publications identified in the area of graph/network clustering and blockmodeling, and included in the Web of Science¹ (WoS) through February 2017. The two dominant approaches for clustering networks are found in the “social” social network literature and the literature featuring physicists and other scientists examining networks. Blockmodeling is an approach that partitions the nodes of a network into positions (clusters of nodes) with the blocks being the sets of relationships within and between positions. The result is a simplified image of the whole network. Community detection, associated with the work of physicists studying networks, aims to identify communities composed of nodes having a higher probability of being connected to each other than to members of other communities. In identifying the literature featuring the clustering of networks we ensured the inclusion of both of these approaches.

The rest of the chapter is structured as follows: Section 2.2 outlines steps in the collection of data and ...