Social Networks Analysis and Machine Learning: an Overview of Approaches and Applications

A significant data mining challenge is social network analysis. To effectively retain the network topological structure and other attribute information, it is essential to encode network data into low-dimensional representations, or network embeddings, before doing social network analysis. Classification, link prediction, anomaly detection, and clustering are further applications made easier by network representation learning. Machine learning-based approaches have also gained a lot of attention in recent years. This study performs a thorough analysis of the recent literature on network representation learning by utilizing neural network models. A study of past research works in social network analysis reveals that most of them used methods based on matrices and graphs. The same methodology is used in several studies that concentrate on various facets of society. The antiquated techniques, which mostly rely on visual analysis, are suitable for usage with more controllable datasets. Although it is possible to automate the process of extracting the information while also reducing bias, one could argue that a study of social networks based on machine learning could perform better even now, in unpredictable circumstances. This is because the data extraction based on machine learning has delivered better results.