System and method for contrastive network analysis and visualization

1. A method of analyzing a target network relative to a background network of data using machine learning, the method comprising:

extracting a first feature matrix from an adjacency matrix representative of the target network;

extracting a second feature matrix from an adjacency matrix representative of the background network;

generating a projection matrix based on the first and second feature matrices using a contrastive learning algorithm;

generating a first contrastive matrix representation of the target network based on the projection matrix and the first feature matrix;

generating a second contrastive matrix representation of the background network based on the projection matrix and the second feature matrix; and

displaying a visualization of unique features of the target network relative to the background network based on the first contrastive matrix and the second contrastive matrix.

2. The method of claim 1 , wherein the first feature matrix is extracted from the adjacency matrix representative of the target network using a network learning algorithm; and

wherein the second feature matrix is extracted from the adjacency matrix representative of the target network using the network learning algorithm.

3. The method of claim 2 , wherein the network learning algorithm is a DeepGL algorithm.

4. The method of claim 1 , wherein the target network is a neuron network representation associated with a patient generated at a specific time;

wherein the background network is a neuron network representation associated with the patient generated at a prior time before the specific time; and

wherein the method further comprises identifying changes in the neuron network representation associated with the patient generated at a specific time based on the visualized unique features of the target network.

5. The method of claim 1 , wherein the target network is a social collaboration network associated with a research domain generated at a specific time;

wherein the background network is a social collaboration network associated with a research domain generated at a prior time before the specific time; and

wherein the method further comprises identifying changes in the social collaboration network associated with a research domain generated at a specific time based on the visualized unique features of the target network.

6. The method of claim 1 , wherein the target network is a network representation of phenotypic features of a first biological species;

wherein the background network is a network representation of phenotypic features of a second, related biological species; and

wherein the method further comprises identifying distinctive phenotypic features of the first biological species based on the visualized unique features of the target network.

7. The method of claim 1 , wherein the target network is a network representation of links between symptoms found in patients diagnosed with a first biological disease;

wherein the background network is a network representation of links between symptoms found in patients diagnosed with a second biological disease; and

wherein the method further comprises identifying distinctive symptoms unique to the first biological disease based on the visualized unique features of the target network.

8. The method of claim 1 , wherein the target network is a real world network being modeled by a network modeling application;

wherein the background network is a model network representative of the real world network generated by the network modeling application; and

wherein the method further comprises updating the network modeling application to include the visualized unique features of the target network.

9. A non-transitory computer readable medium encoded with instructions for making a computing device execute a method of analyzing a target network relative to a background network of data using machine learning, the method comprising:

extracting a first feature matrix from an adjacency matrix representative of the target network;

extracting a second feature matrix from an adjacency matrix representative of the background network;

generating a projection matrix based on the first and second feature matrices using a contrastive learning algorithm;

generating a first contrastive matrix representation of the target network based on the projection matrix and the first feature matrix;

generating a second contrastive matrix representation of the background network based on the projection matrix and the second feature matrix; and

displaying a visualization of unique features of the target network relative to the background network based on the first contrastive matrix and the second contrastive matrix.

10. The non-transitory computer readable medium of claim 9 , wherein the first feature matrix is extracted from the adjacency matrix representative of the target network using a network learning algorithm;

wherein the second feature matrix is extracted from the adjacency matrix representative of the target network using the network learning algorithm; and

wherein the network learning algorithm is a DeepGL algorithm.

11. The non-transitory computer readable medium of claim 9 , wherein the target network is a neuron network representation associated with a patient generated at a specific time;

wherein the background network is a neuron network representation associated with the patient generated at a prior time before the specific time; and

wherein the method further comprises identifying changes in the neuron network representation associated with the patient generated at a specific time based on the visualized unique features of the target network.

12. The non-transitory computer readable medium of claim 9 , wherein the target network is a social collaboration network associated with a research domain generated at a specific time;

wherein the background network is a social collaboration network associated with a research domain generated at a prior time before the specific time; and

wherein the method further comprises identifying changes in the social collaboration network associated with a research domain generated at a specific time based on the visualized unique features of the target network.

13. The non-transitory computer readable medium of claim 9 , wherein the target network is a network representation of phenotypic features of a first biological species;

wherein the background network is a network representation of phenotypic features of a second, related biological species; and

wherein the method further comprises identifying distinctive phenotypic features of the first biological species based on the visualized unique features of the target network.

14. The non-transitory computer readable medium of claim 9 , wherein the target network is a network representation of links between symptoms found in patients diagnosed with a first biological disease;

wherein the background network is a network representation of links between symptoms found in patients diagnosed with a second biological disease; and

wherein the method further comprises identifying distinctive symptoms unique to the first biological disease based on the visualized unique features of the target network.

15. The non-transitory computer readable medium of claim 9 , wherein the target network is a real world network being modeled by a network modeling application;

wherein the background network is a model network representative of the real world network generated by the network modeling application; and

wherein the method further comprises updating the network modeling application to include the visualized unique features of the target network.

16. A computing device comprising:

a storage device storing network data associated with a target network and a background network; and

a processor encoded to execute a method of automatically analyzing the target network relative to the background network of data using machine learning, the method comprising:

extracting a first feature matrix from an adjacency matrix representative of the target network;

extracting a second feature matrix from an adjacency matrix representative of the background network;

generating a projection matrix based on the first and second feature matrices using a contrastive learning algorithm;

generating a first contrastive matrix representation of the target network based on the projection matrix and the first feature matrix;

generating a second contrastive matrix representation of the background network based on the projection matrix and the second feature matrix; and

displaying a visualization of unique features of the target network relative to the background network based on the first contrastive matrix and the second contrastive matrix.

17. The computing device of claim 16 , wherein the target network is a neuron network representation associated with a patient generated at a specific time;

wherein the background network is a neuron network representation associated with the patient generated at a prior time before the specific time; and

wherein the method further comprises identifying changes in the neuron network representation associated with the patient generated at a specific time based on the visualized unique features of the target network.

18. The computing device of claim 16 , wherein the target network is a social collaboration network associated with a research domain generated at a specific time;

wherein the background network is a social collaboration network associated with a research domain generated at a prior time before the specific time; and

wherein the method further comprises identifying changes in the social collaboration network associated with a research domain generated at a specific time based on the visualized unique features of the target network.

19. The computing device of claim 16 , wherein the target network is a network representation of phenotypic features of a first biological species;

wherein the background network is a network representation of phenotypic features of a second, related biological species; and

wherein the method further comprises identifying distinctive phenotypic features of the first biological species based on the visualized unique features of the target network.

20. The computing device of claim 16 , wherein the target network is a network representation of links between symptoms found in patients diagnosed with a first biological disease;

wherein the background network is a network representation of links between symptoms found in patients diagnosed with a second biological disease; and

wherein the method further comprises identifying distinctive symptoms unique to the first biological disease based on the visualized unique features of the target network.

21. The computing device of claim 16 , wherein the target network is a real world network being modeled by a network modeling application;

wherein the background network is a model network representative of the real world network generated by the network modeling application; and

wherein the method further comprises updating the network modeling application to include the visualized unique features of the target network.