Object risk identification method, device and equipment based on graph neural network

The invention discloses an object risk identification method, device and equipment based on a graph neural network. The method comprises the steps that a trigger relation structure diagram is obtained, nodes of the trigger relation structure diagram comprise a plurality of object nodes and a plurality of application use event nodes, the node content of each object node is attribute feature information of a corresponding object, and the node content of each application use event node is event feature information of a corresponding application use event; the edge of the trigger relation structure diagram represents the trigger relation between each object node and each application use event node; inputting the trigger relation structure diagram into a graph neural network to perform overdue risk feature extraction of the target object node to obtain overdue risk feature information of the target object node; and based on the overdue risk feature information, performing overdue risk identification on a target ob