Descriptive Community Detection

Subgroup discovery and community detection are standard approaches for identifying (cohesive) subgroups. This paper presents an organized picture of recent research in descriptive community (and subgroup) detection. Here, it summarizes approaches for the identification of descriptive patterns targeting both static as well as dynamic (sequential) relations. We specifically focus on attributed graphs, i. e., complex relational graphs that are annotated with additional information. This relates to attribute information, for example, assigned to the nodes and/or edges of the graph. Combining subgroup discovery and community detection, we also summarize an efficient and effective approach for descriptive community detection.