NFE-PCN: A Node Feature Enhanced Embedding Framework for Pattern Change in Dynamic Network

Dynamic networks are complex as their structures and node features change over time. However, they can better represent the real world, thus attracting interest of researchers. Although realistic dynamic often exhibit changes in patterns, existing network models tend to classify all snapshots having same pattern learn during embedding. These embedding ignore a large amount information about patterns networks. So, it is necessary design dedicated framework for learning Accordingly, this paper proposes new framework, namely NFE-PCN effectively extracting Specifically, first determines which snapshot located, then enhances between by maintaining pattern. We conduct experiments with both artificial datasets predicting links classifying nodes. The obtained results show that model under decreases computational effort performance improved up 29%, quite significant.