This is not the final version of this paper. You can find the final version on the publisher web page. A Method for Group Extraction in Complex Social Networks Piotr Bródka 1 , Katarzyna Musial 2 , Przemysław Kazienko 1 1 Institute of Informatics, Wrocław University of Technology Wyb.Wyspiańskiego 27, 50-370 Wrocław, Poland 2 School of Design, Engineering & Computing, Bournemouth University, Poole, Dorset, BH12 5BB, United Kingdom piotr.brodka@pwr.wroc.pl, kmusial@bournemouth.ac.uk, kazienko@pwr.wroc.pl Abstract. The extraction of social groups from social networks existing among employees in the company, its customers or users of various computer systems became one of the research areas of growing importance. Once we have discovered the groups, we can utilise them, in different kinds of recommender systems or in the analysis of the team structure and communication within a given population. The shortcomings of the existing methods for community discovery and lack of their applicability in multi-layered social networks were the inspiration to create a new group extraction method in complex multi-layered social networks. The main idea that stands behind this new concept is to utilise the modified version of a measure called by authors multi-layered clustering coefficient. Keywords: multi-layered social network, groups discovery in social network, multi-layered clustering coefficient, social network analysis 1 Introduction In the recent few decades, the area of complex networks has attracted more and more scientists from different research fields. All complex networked systems have some common features such as: (i) skewed distribution of connections, (ii) small degree of separation between vertices, (iii) high clustering rate, (iv) non-trivial temporal evolution and last but not the least (v) presence of motifs, hierarchies and communities [3], [10]. The feature that is investigated by authors in this paper is the last enumerated one, i.e. the existence of communities whereas the subset of complex systems that is analysed are social networks. A social network (SN) is one of the type of complex networks in which nodes are people (social entities) and the edges denote the relationships between various people [13]. The concept of SN, first coined in 1954 by J. A. Barnes [1], has been in a field of study of modern sociology, geography, social psychology, organizational studies and computer science for the last few decades. Social networks and social network analysis supported by computer science provide the opportunity to expand other