The embodiment of the invention discloses a graph alignment method and device and a storage medium, and the method comprises the steps: obtaining a source domain graph and a target domain graph whichneed to be aligned; performing unsupervised learning on the source domain graph to obtain a source domain node feature set, and performing unsupervised learning on the target domain graph to obtain atarget domain node feature set; constructing a generative adversarial network, and training the generative adversarial network; predicting an anchor link between the source domain graph and the targetdomain graph according to a generator in the trained generative adversarial network, and optimizing the trained generator according to a source domain node feature and a target domain node feature corresponding to the anchor link to obtain an optimized generator; and aligning the source domain nodes in the source domain graph with the target domain nodes in the target domain graph according to the optimized generator. Therefore, graph alignment under an unsupervised condition is realized, dependence on annotation data in the prior art is eliminated, the labor cost for realizing graph alignment is reduced, and the graph alignment efficiency is also improved.