Machine learning for dynamically updating a user interface

1. A computer-implemented method for dynamically updating a user interface, the method comprising:

receiving, by one or more processors, a plurality of claims in real-time for electronic storage in an open claim inventory, the plurality of claims comprising a first claim and a second claim;

determining, by the one or more processors, a first predicted recovery score for the first claim and a second predicted recovery score for the second claim using one or more machine learning models, wherein each predicted recovery score is indicative of the likelihood that the corresponding claim is recoverable;

determining, by the one or more processors, a first priority score for the first claim and a second priority score for the second claim, wherein each priority score is determined based at least in part on the corresponding predicted recovery score and a corresponding recoverable amount;

inserting, by the one or more processors, the first claim in a first queue data structure, wherein the first queue data structure (a) is associated with a first plurality of claims, and (b) comprises an indication of the first priority score for the first claim and respective priority scores for the first plurality of claims;

inserting, by the one or more processors, the second claim in a second queue data structure, wherein the second queue data structure (a) is associated with a second plurality of claims, and (b) comprises an indication of the second priority score for the second claim and respective priority scores for the second plurality of claims;

prioritizing, by the one or more processors, (a) the first claim and the first plurality of claims in the first queue data structure, and (b) the second claim and the second plurality of claims in the second queue data structure; and

dynamically providing, by the one or more processors, an update to (a) a first user interface element associated with the first queue data structure, and (b) a second user interface element associated with the second queue data structure, wherein the user interface displays the update (a) to the first user interface element, and (b) the second user interface element.

2. The computer-implemented method of claim 1 further comprising:

extracting a first feature set from a set of training claims;

determining a recovery success rate for each claim of the set of training claims, wherein the recovery success rate is a target output of the one or more machine learning models; and

storing the first feature set and the recovery success rates for the set of training claims as a training dataset.

3. The computer-implemented method of claim 2 further comprising automatically detecting a change in the first feature set and the first claim.

4. The computer-implemented method of claim 2 further comprising training the one or more machine learning models based at least in part on the training dataset.

5. The computer-implemented method of claim 1 , wherein the priority score for the first claim is the product of the predicted recovery score for the first claim and the first recoverable amount for the first claim.

6. The computer-implemented method of claim 1 , wherein dynamically providing the update to the first user interface element and the second user interface element comprises pushing the update.

7. The computer-implemented method of claim 1 further comprising

determining a predicted recovery score for each of the first plurality of claims; and

determining a priority score for each of the first plurality of claims.

8. A computer program product comprising a non-transitory computer readable medium having computer program instructions stored therein, the computer program instructions when executed by a processor, cause the processor to:

receive a plurality of claims in real-time for electronic storage in an open claim inventory, the plurality of claims comprising a first claim and a second claim;

determine a first predicted recovery score for the first claim and a second predicted recovery score for the second claim using one or more machine learning models, wherein each predicted recovery score is indicative of the likelihood that the corresponding claim is recoverable;

determine a first priority score for the first claim and a second priority score for the second claim, wherein each priority score is determined based at least in part on the corresponding predicted recovery score for the first claim and a corresponding recoverable amount;

insert the first claim in a first queue data structure, wherein the first queue data structure (a) is associated with a first plurality of claims, and (b) comprises an indication of the first priority score for the first claim and respective priority scores for the first plurality of claims;

insert, by the one or more processors, the second claim in a second queue data structure, wherein the second queue data structure (a) is associated with a second plurality of claims, and (b) comprises an indication of the second priority score for the second claim and respective priority scores for the second plurality of claims;

prioritize (a) the first claim and the first plurality of claims in the first queue data structure, and (b) the second claim and the second plurality of claims in the second queue data structure; and

dynamically provide an update to (a) a first user interface element associated with the first queue data structure, and (b) a second user interface element associated with the second queue data structure, wherein the user interface displays the update (a) to the first user interface element, and (b) the second user interface element.

9. The non-transitory computer readable medium of claim 8 , wherein the computer program instructions when executed by a processor, further cause the processor to:

extract a first feature set from a set of training claims;

determine a recovery success rate for each claim of the set of training claims, wherein the recovery success rate is a target output of the one or more machine learning models; and

store the first feature set and the recovery success rates for the set of training claims as a training dataset.

10. The non-transitory computer readable medium of claim 9 , wherein the computer program instructions when executed by a processor, further cause the processor to automatically detect a change in the first feature set and the first claim.

11. The non-transitory computer readable medium of claim 9 , wherein the computer program instructions when executed by a processor, further cause the processor to train the one or more machine learning models based at least in part on the training dataset.

12. The non-transitory computer readable medium of claim 8 , wherein the priority score for the first claim is the product of the predicted recovery score for the first claim and the first recoverable amount for the first claim.

13. The non-transitory computer readable medium of claim 8 , wherein dynamically providing the update to the first user interface element and the second user interface element comprises pushing the update.

14. The non-transitory computer readable medium of claim 8 , wherein the computer program instructions when executed by a processor, further cause the processor to:

determine a predicted recovery score for each of the first plurality of claims; and

determine a priority score for each of the first plurality of claims.

15. A computing system comprising a non-transitory computer readable storage medium and one or more processors, the computing system configured to:

receive a plurality of claims in real-time for electronic storage in an open claim inventory, the plurality of claims comprising a first claim and a second claim;

determine a first predicted recovery score for the first claim and a second predicted recovery score for the second claim using one or more machine learning models, wherein each predicted recovery score is indicative of the likelihood that the corresponding claim is recoverable;

determine a first priority score for the first claim and a second priority score for the second claim, wherein each priority score is determined based at least in part on the corresponding predicted recovery score and a corresponding recoverable amount;

insert the first claim in a first queue data structure, wherein the first queue data structure (a) is associated with a first plurality of claims, and (b) comprises an indication of the first priority score for the first claim and respective priority scores for the first plurality of claims;

insert, by the one or more processors, the second claim in a second queue data structure, wherein the second queue data structure (a) is associated with a second plurality of claims, and (b) comprises an indication of the second priority score for the second claim and respective priority scores for the second plurality of claims;

prioritize the first claim and the first plurality of claims in the first queue data structure, and (b) the second claim and the second plurality of claims in the second queue data structure; and

dynamically provide an update to (a) a first user interface element associated with the first queue data structure, and (b) a second user interface element associated with the second queue data structure, wherein the user interface displays the update (a) to the first user interface element, and (b) the second user interface element.

16. The computing system of claim 15 , wherein the computing system is further configured to:

extract a first feature set from a set of training claims;

determine a recovery success rate for each claim of the set of training claims, wherein the recovery success rate is a target output of the one or more machine learning models; and

store the first feature set and the recovery success rates for the set of training claims as a training dataset.

17. The computing system of claim 16 , wherein the computing system is further configured to automatically detect a change in the first feature set and the first claim.

18. The computing system of claim 16 , wherein the computing system is further configured to train the one or more machine learning models based at least in part on the training dataset.

19. The computing system of claim 15 , wherein the priority score for the first claim is the product of the predicted recovery score for the first claim and the first recoverable amount for the first claim.

20. The computing system of claim 15 , wherein dynamically providing the update to the first user interface element and the second user interface element comprises pushing the update.

21. The computing system of claim 15 , wherein the computing system is further configured to:

determine a predicted recovery score for each of the first plurality of claims; and

determine a priority score for each of the first plurality of claims.