Computational Social Science Winter Symposium

Subgroup and Community Analytics

Organizers: Martin Atzmüller (University of Kassel)


Subgroup discovery and community detection are two approaches having been studied in different research areas like data mining and social network analysis. In this context, these techniques are especially helpful in order to provide for analytical and explorative data mining approaches, and for extracting knowledge for humans. We present an organized picture of recent research in subgroup discovery and community detection. Specifically, we also include recent advances on mining attributed graphs. That is, we include complex relational graphs that are annotated with additional information, e.g., attribute information on the nodes and/or edges of the graph. Then, descriptive patterns can be extracted using a variety of techniques, ranging from structural approaches to description-based methods. This includes detecting cohesive subgroups, correlated patterns, subspace clustering, and exceptional model mining methods.

Tutorial Outline:

  • Motivation & Introduction
    • Exploratory Data Analysis – Overview
    • SNA: Structural and Compositional Perspective
  • Subgroup Discovery: Introduction to Exploratory Subgroup Analysis
  •  Community Detection & Analysis
    • Methodological Overview: Core principles, Categorization & Comparison
    • Partitioning Algorithms: Methodological Introduction & Algorithmic Pointers
    • Overlapping Community Detection: Methodological Introduction & Algorithmic Pointers
    • Description-Oriented Community Detection: Methodological Introduction & Algorithmic Pointers
  • Community Evaluation: Methods and Techniques
  • Summary & Conclusions       

Organizational Details:

Target audience: The intended audience includes students, junior and senior researchers, and practitioners interested in the topics data mining and analysis on social networks and social media data.
Prerequisite: Knowledge about basic social network analysis principles and a basic data mining background are helpful, while important central concepts in the scope of the tutorial will be introduced and discussed.

Date and time:
Dec 1st, 09:00-11:00
GESIS Cologne, West II

PD Dr. Martin Atzmüller, Research Center for Information System Design, Knowledge and Data Engineering Group,
University of Kassel, Wilhelmshoeher Allee 73, 34121 Kassel, atzmueller(at)cs.uni-kassel(dot)de,