Simula@BI - Clustering via Coherent Network Partitions
Simula@BI invites Angela Angeleska, University of Tampa, to give a talk about computer science
Simula@BI invites Associate Professor Angela Angelska from University of Tampa, to give a talk about computer science.
Coherent partition of a graph G is recently defined as a vertex partition that corresponds to a partition composed only of disconnected subgraphs in the complement of G. An optimal coherent partition is defined as a coherent partition induced by an edge cut with the smallest number of edges. Here we characterize some relations between coherent and other types of partitions, including: clique partitions, biclique partitions, and modular graph decompositions.
We show that the problem of finding optimal coherent partition in graphs is NP-hard, but it reduces to a problem of finding optimal coherent partition in prime graphs. In addition, we demonstrate an application of coherent partitions in systems biology by introducing a family of parameter-free algorithms to efficiently predict protein complexes in protein-protein interaction networks.
Registration form coming soon.