Community detection is a fundamental problem in network data
analysis. Most of the existing community detection methods focus only
on the network structure without considering features of the nodes.
However, many real-world networks contain feature information on
nodes, which are often related to the community structure. To address
this problem, we propose a joint community detection criterion that
combines the network structure and the feature information on nodes.
The new criterion is shown to perform well in simulation studies and
on several data examples.
News & Events
- Department Seminar Series
- Graduate Student Seminar Series
- Statistics Career Fair
- Dissertation Defenses and Oral Preliminary Examinations
- Michael Woodroofe Lecture Series
- International Year of Statistics Events
- Seminars in Other Departments
- Archived Events