Suppose if we have a visual scene graph and we model each component in the scene as a node of a graph and edges which are relationship between the visual scene components. Some of the nodes are like agent who can move(say human as node can move but not tree in scene graph). However, their movement or action, like who will move first is based on some rules. How to approach the problem of node priority list generation in this case. Like node A will move first, followed by node B and C. Usually graph are modelled as classification problem or link prediction problem. Not able to find a way in which such kind of priority list generation is tackled over graphs (or knowledge graph). Or suggest any other approach to tackle it without graph network or knowledge graph.



You must log in to answer this question.