Hello Titas De . You can think about representing the system as a heterogeneous graph and use a graph neural network for feature embedding. This page would be a good start to get your way https://pytorch-geometric.readthedocs.io/en/latest/notes/heterogeneous.html. You can also read some papers about GNN for Recommendation Systems. This survey paper may be helpful Article Graph Neural Networks in Recommender Systems: A Survey

Thanks Babatounde Moctard Oloulade for the direction. But the main challenge lies in how we are defining the graph from both user and item point of view.