A Dynamic Community Detection Method for Complex Networks Based on Deep Self-Coding Network

Table 1

Dynamic community detection algorithms.

Algorithm	Description
Line graph neural network (LGNN) [12]	The high-level graph algorithm is realized by sparse matrix multiplication and message passing
Context path-based graph neural network (CP-GNN) [13]	The nodes of an input graph are classified by weak supervision
Graph convolution network (GCN) algorithm [14]	Local parameters are shared and the receptive field is proportional to the number of layers
The proposed model	The temporal information is effectively combined to transition oversmooth process