A graph convolutional network (GCN) is a neural network explicitly developed for processing graph-structured data input. Graph-structured data is represented as a graph, with each node denoting an entity and each edge signifying a relationship between two entities. Graph Convolutional Networks (GCNs) are applicable to numerous tasks involving graph-structured data, including node classification, link prediction, and graph classification. Graph Convolutional Networks (GCNs) perform by executing a sequence of convolutional operations on graph data. These convolution techniques resemble those utilized in convolutional neural networks (CNNs), although they are modified to function on graph-structured input. The convolution procedures in Graph Convolutional Networks (GCNs) consider the interconnections among nodes in the graph, enabling GCNs to acquire more intricate features from the graph data. Graph Convolutional Networks (GCNs) exhibit efficacy across numerous tasks using graph-structured data. For instance, GCNs have been employed for node classification in social networks, link prediction in recommendation systems, and graph classification in natural language processing. The various types of graph convolution networks like Spatial graph convolutional networks: These networks employ a spatial convolution operation to consolidate the properties of adjacent nodes. Spectral graph convolutional networks: These networks employ a spectral convolution process, which is more potent yet computationally intensive. Message passing neural networks: These networks are a variant of graph convolutional networks (GCN) that employs a message passing mechanism to consolidate the features of adjacent nodes. In graph computation, GCNs denote the application of a spatially moving filter across the nodes of a graph, which contains embeddings or data pertinent to each node, to derive a feature representation for each node. Information from bigger neighbourhoods can also be integrated by stacking many convolutional layers, akin to the construction of a conventional CNN.