Local Properties of Social Networks

Abstract

Due to the development of information technology, we have a large amount of data available in recent years, perhaps from all areas of research. Many real world data and processes have a network structure and can usefully be represented as graphs. The network representation of complex systems provides a useful model for studying many processes, including biological, technological and social networks. Network analysis focuses on the relations among the nodes exploring the properties of each network. Owing to enormous and sustained growth of real world networks, the current trend in analyzing networks is to focus on local methods. This thesis is focused on investigating local characteristics of social networks and application of local methods on large datasets. Two novel local measures for node importance in the network are presented. The first one can be used for ranking of nodes or as an approach to transforming an unweighted network to weighted one, or to assist community detection. The second one can be utilized for network sampling or graph construction. Another topic of this thesis is also the analysis of large co-authorship datasets in order to develop a model capable of generating realistic graphs.