Deep learning-based similarity evaluation in decentralized identity graphs

1. A computer-implemented method comprising:

representing a first identity profile from a first network as a first identity graph;

representing a second identity profile from a second network as a second identity graph, wherein the first identity profile and the second identity profile are decentralized identity profiles, wherein the first identity graph and the second identity graph are decentralized identity graphs;

representing a first neighborhood of nodes of the first identity graph as a first neighborhood vector and a second neighborhood of nodes of the second identity graph as a second neighborhood vector;

performing a contrastive loss analysis of the first identity graph to the second identity graph to describe similarities in the first and second neighborhood vectors;

determining a similarity score between the first identity profile and the second identity profile based on the similarities of the first and second identity graphs, wherein the similarity score is across multiple identity profiles represented in the decentralized identity graphs; and

implementing a security action based on the similarity score.

2. The method of claim 1 , wherein the first identity graph and the second identity graph are graph neural networks.

3. The method of claim 1 , further comprising:

determining that the second identity profile is legitimate based on the similarity score being more than a predetermined value.

4. The method of claim 1 , wherein:

the similarity score is less than a predefined value, and

the security action is blocking a release of the identity profile to a requester of the identity profile in response to the similarity score being less than the predefined value.

5. The method of claim 1 , further comprising:

linking a first node and a second node in the first identity graph with a deep neural network; and

predicting a content of the second node based on an input from the first node to the deep neural network.

6. A computer program product comprising a computer readable storage medium having program code embodied therewith, wherein the computer readable storage medium is not a transitory signal per se, and wherein the program code is readable and executable by a processor to perform a method comprising:

representing a first identity profile from a first network as a first identity graph;

representing a second identity profile from a second network as a second identity graph, wherein the first identity profile and the second identity profile are decentralized identity profiles, wherein the first identity graph and the second identity graph are decentralized identity graphs;

representing a first neighborhood of nodes of the first identity graph as a first neighborhood vector and a second neighborhood of nodes of the second identity graph as a second neighborhood vector;

performing a contrastive loss analysis of the first identity graph to the second identity graph to describe similarities in the first and second neighborhood vectors;

determining a similarity score between the first identity profile and the second identity profile based on the similarities of the first and second identity graphs, wherein the similarity score is across multiple identity profiles represented in the decentralized identity graphs; and

implementing a security action based on the similarity score.

7. The computer program product of claim 6 , wherein the first identity graph and the second identity graph are graph neural networks.

8. The computer program product of claim 6 , the method further comprising:

determining that the second identity profile is legitimate based on the similarity score being more than a predetermined value.

9. The computer program product of claim 6 , wherein:

the similarity score is less than a predefined value, and

the security action is blocking a release of the identity profile to a requester of the identity profile in response to the similarity score being less than the predefined value.

10. The computer program product of claim 6 , wherein the method further comprises:

linking a first node and a second node in the first identity graph with a deep neural network; and

predicting a content of the second node based on an input from the first node to the deep neural network.

11. The computer program product of claim 6 , wherein the program code is provided as a service in a cloud environment.

12. A computer system comprising:

one or more processors, and

a computer readable storage medium,

wherein:

the one or more processors are structured, located, connected, and/or programmed to run program instructions stored on the computer readable storage medium; and

the program instructions which, when executed by the one or more processors, cause the one or more processors to perform a method comprising:

representing a first identity profile from a first network as a first identity graph;

representing a second identity profile from a second network as a second identity graph, wherein the first identity profile and the second identity profile are decentralized identity profiles, wherein the first identity graph and the second identity graph are decentralized identity graphs;

representing a first neighborhood of nodes of the first identity graph as a first neighborhood vector and a second neighborhood of nodes of the second identity graph as a second neighborhood vector;

performing a contrastive loss analysis of the first identity graph to the second identity graph to describe similarities in the first and second neighborhood vectors;

determining a similarity score between the first identity profile and the second identity profile based on the similarities of the first and second identity graphs, wherein the similarity score is across multiple identity profiles represented in the decentralized identity graphs; and

implementing a security action based on the similarity score.

13. The computer system of claim 12 , wherein the first identity graph and the second identity graph are graph neural networks.

14. The computer system of claim 12 , the method further comprising:

determining that the second identity profile is legitimate based on the similarity score being more than a predetermined value.

15. The computer system of claim 12 , wherein:

the similarity score is less than a predefined value, and

the security action is blocking a release of the identity profile to a requester of the identity profile in response to the similarity score being less than the predefined value.

16. The computer system of claim 12 , wherein the method further comprises:

linking a first node and a second node in the first identity graph with a deep neural network; and

predicting a content of the second node based on an input from the first node to the deep neural network.

17. The computer system of claim 12 , wherein the stored program instructions are provided as a service in a cloud environment.