A Unified Bayesian Model for Generalized Community Detection in Attribute Networks

Algorithm 1

Inference for GSC.

	Require: adjacency matrix , attributes matrix , iterations , and specified group number
Ensure: group assignment
0: initialize , , , , set , , , , , , and to 0
Initialize each node’s latent community label
(1)//sampling z, , , , , and
(2)for to do
(3) for to do
(4) //get the current community assignment of node
(5) update , , , , , , and
(6) for to do
(7) compute probability according to equation (8)
(8) end for
(9) Gibbs sampling for and obtain
(10) update , , , , , , and
(11) for to do
(12) //get the current topic assignment of attribute
(13) update , , , and
(14) for = 1 to do
(15) compute probability according to equation (9)
(16) end for
(17) Gibbs sampling for and obtain
(18) update , , , and
(19) end for
(20) end for
(21) slice sampling for , , , and in (0, 1)
(22)end for