Tutorial on Community Detection: From Plain to Attributed Complex Networks @ WebSci'16

Tutorial on Community Detection: From Plain to Attributed Complex Networks @ WebSci'16


The ACM Web Science conference provides an interdisciplinary forum, connecting researchers from multiple disciplines, like computer science, sociology, economics, information science, or psychology. With the information-rich networks in the World Wide Web and associated rich content data, there is a growing interest in the comprehensive analysis of this data. Together with a variety of heterogeneous data sources complementing network data, this leads to the modeling of complex attributed graphs. The tutorial specifically addresses these issues and provides an overview on methods, techniques, example applications and tools for subgroup analysis and community detection on attributed graphs providing an organized picture on analytical methods, also including exemplary applications.

Tutorial Description

Cohesive subgroup analysis and community detection are two approaches having been studied in social network analysis. In this context, these techniques are especially helpful in order to provide for analytical analysis approaches. We present an organized picture of recent research on subgroup analysis and community detection, specifically targeting complex relational networks that include compositional information concerning actors or ties. These are annotated with additional information, e.g., attribute information on the nodes and/or edges of the corresponding graph. Then, patterns and communities can be extracted using a variety of techniques, ranging from structural approaches to description-based methods.

Tutorial Slides

The tutorial slides are available: Download slides


Martin Atzmueller is adjunct professor (Privatdozent) at the University of Kassel and heads the Ubiquitous Data Mining Team at the Research Center for Information System Design (ITeG, Hertie Chair for Knowledge and Data Engineering, University of Kasssel). He earned his habilitation (Dr. habil.) in 2013 at the University of Kassel, and received his Ph.D. in Computer Science from the University of Wuerzburg in 2006. He studied Computer Science at the University of Texas at Austin (USA) and at the University of Wuerzburg where he completed his MSc in Computer Science. His research areas include data mining, ubiquitous and social computing, mining social media, as well as web and network science. He has published a significant number of scientific articles in top venues, e.g., the International Joint Conference on Artificial Intelligence (IJCAI), the European Conference on Machine Learning and Principles and Practice on Knowledge Discovery in Databases (ECML PKDD), the IEEE Conference on Social Computing (SocialCom), the ACM/IEEE International Conference on Advances in Social Networks Analysis and Mining (ASONAM), the ACM International Conference on Information and Knowledge Management (CIKM) and the ACM Conference on Hypertext and Social Media (HT). He is the winner of several Best Paper and Innovation Awards. He regularly acts as PC member of several top-tier conferences and as co-organizer on a number of international workshops and conferences, e.g., AAAI, ACM HT, ECML/PKDD, ICML, MUSE, MSM, SocialCom. He has given tutorials on “Applied Descriptive Pattern Mining” at LWA 2011 in Magdeburg, Germany, on “Community and Pattern Analytics in Social Networks” at LWA 2013 in Bamberg, Germany, on “Subgroup Discovery and Community Detection on Attributed Graphs” at ASONAM 2015 in Paris, France, and on “Subgroup and Community Analytics” at the Computational Social Science Wintersymposium (CSSWS) 2015 in Cologne, Germany.

PD Dr. Martin Atzmueller, Research Center for Information System Design, Knowledge and Data Engineering Group, University of Kassel, Wilhelmshoeher Allee 73, 34121 Kassel, atzmueller@cs.uni-kassel.de, https://www.kde.cs.uni-kassel.de/atzmueller