Interest-driven community detection on attributed heterogeneous information networks

Mengyue Liu, Jun Liu, Yixiang Dong, Rui Mao, Erik Cambria

Abstract: Highlights•Detecting communities in attributed heterogeneous information networks (AHINs) requires user guidance or supervision for meaningful results.•Joint optimization of network embedding and community detection achieves better performances.•Integrate semantic-related structures with attributes helps to sample informative nodes for community detection tasks.•Gaussian mixture distribution is more effective for community assignments than Student’s t-distribution.