Thursday, February 22, 2024

 

Adaptive fusion of structure and attribute guided polarized communities search


Peer-Reviewed Publication

HIGHER EDUCATION PRESS

figure 

IMAGE: 

AN EXAMPLE OF THE POLARIZED COMMUNITIES IN AN ATTRIBUTED SIGNED NETWORK. SOLID EDGES ARE POSITIVE, WHILE DASHED EDGES ARE NEGATIVE. THE POLARIZED COMMUNITIES WHERE THE GREEN BOX IS LOCATED UTILIZE TOPOLOGY, WHILE THE POLARIZED COMMUNITIES WHERE THE RED BOX IS LOCATED CONSIDER BOTH TOPOLOGY AND ATTRIBUTE.

view more 

CREDIT: FANYI YANG, HUIFANG MA, WENTAO WANG, ZHIXIN LI, LIANG CHANG




The exploration of polarized communities, which consist of two antagonistic subgraphs and include a set of query nodes, is a crucial task in community search on signed networks. Most existing methods either predominantly rely on topological structure while disregarding node attributes or tend to prioritize the global identification of all polarized communities. Thus, they fail to consider two crucial insights. Firstly, integrating node attributes with network structure can enhance the search quality for polarized communities in attributed signed networks by leveraging complementary information. Secondly, global criteria-based polarized community detection aims to identify all polarized communities, neglecting personalized analyses centered around individual users.

To solve the problems, a research team led by Huifang MA published their new research on 15 Feb 2024 in Frontiers of Computer Science co-published by Higher Education Press and Springer Nature.

The research team introduced a novel community search framework termed "Polarized Communities via Adaptively Fusing Structure and Attribute in Attributed Signed Networks." Their approach starts by performing statistical analyses on the signed graph at the attribute level, establishing associations with the topological relationships. Subsequently, an adaptive fusion mechanism of the topological structure and attribute information is introduced. In order to detect polarized communities that include query nodes, a sparse indicator-vector is formulated based on the generalized Rayleigh quotient in local spectral eigenspaces, achieved by solving a linear programming problem.

DOI: 10.1007/s11704-023-2776-7

No comments:

Post a Comment