CPC G16C 20/70 (2019.02) [G06F 16/9024 (2019.01); G06F 18/24 (2023.01); G16C 20/80 (2019.02); G16C 20/90 (2019.02)] | 18 Claims |
1. A method for identifying a heterogeneous graph, performed by computing device, the method comprising:
obtaining a heterogeneous graph of a molecular space structure, the molecular space structure comprising a chemical molecular structure and a protein molecular structure;
characterizing a molecular structure included in the heterogeneous graph to generate feature information;
generating feature vectors corresponding to key nodes on a topology structure according to sampling information obtained by sampling the heterogeneous graph and the feature information;
aggregating the feature vectors to generate a graph representation vector corresponding to the heterogeneous graph; and
classifying the heterogeneous graph according to the graph representation vector to obtain a classification prediction result of the heterogeneous graph,
wherein the classifying the heterogeneous graph includes:
training and optimizing a neural network, the neural network being a group of given heterogeneous graphs, and
performing a tag prediction on the graph representation vector by using the trained and optimized neural network to obtain the classification prediction result of the heterogeneous graph relative to given tag data, the classification prediction result being indicated as a tag predicted by the heterogeneous graph.
|