Social network analysis is a crucial aspect of data mining, requiring the
encoding of network data into low-dimensional representations called network
embeddings to preserve topology structure and attribute information. This enhances
applications like classification, link prediction, anomaly detection, and clustering, with
deep neural network techniques gaining interest. This survey explores the current
literature on network representation learning, focusing on neural network models. It
introduces basic models, extends them for complex scenarios, introduces embedding
techniques, discusses network representation learning applications, and suggests future
research directions.
Keywords: Deep learning, Deep social network analysis, Network embedding, Representation learning, Social networks.
