Systems and methods for evaluating page content

1. A computer-implemented method comprising:

determining, by a computing system, a set of candidate values for a field in a page;

evaluating, by the computing system, the set of candidate values for accuracy based at least in part on a machine learning model, wherein the evaluating further comprises:

obtaining, by the computing system, a score for a candidate value from the machine learning model based on a feature vector that represents the candidate value, wherein the feature vector is associated with at least one feature that represents a data pipeline endorsement of the candidate value by at least one data pipeline, wherein the data pipeline endorsement is weighted based on a consensus score associated with the at least one data pipeline, wherein the consensus score measures a rate at which the at least one data pipeline has historically provided data pipeline endorsements for candidate values that were endorsed by a threshold amount of users;

determining, by the computing system, a best scoring candidate value from the set of candidate values; and

associating, by the computing system, the field in the page with the best scoring candidate value.

2. The computer-implemented method of claim 1 , wherein the field corresponds to at least one of: a page category field, a website field, a phone number field, an hours of operation field, and a physical address field.

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

causing, by the computing system, the field in the page to be populated with the best scoring candidate value.

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

providing, by the computing system, the best scoring candidate value as a recommendation for populating the field in the page.

5. The computer-implemented method of claim 1 , wherein the feature vector includes a feature representing a set of weighted user endorsements for the candidate value, wherein a user endorsement is weighted based on a credibility score associated with the user, the credibility score measuring a credibility of the user.

6. The computer-implemented method of claim 5 , wherein the feature representing the set of weighted user endorsements for the candidate value is determined based at least in part on an activation function or a logit function.

7. The computer-implemented method of claim 1 , wherein the feature vector includes a feature representing a set of weighted data pipeline endorsements for the candidate value, wherein the set of weighted data pipeline endorsements includes the at least one data pipeline endorsement.

8. The computer-implemented method of claim 7 , wherein the feature representing the set of weighted data pipeline endorsements is determined based at least in part on an activation function or a logit function.

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

causing, by the computing system, one or more users to be polled to confirm or improve an accuracy of the best scoring candidate value.

10. A system comprising:

at least one processor; and

a memory storing instructions that, when executed by the at least one processor, cause the system to perform:

determining a set of candidate values for a field in a page;

evaluating the set of candidate values for accuracy based at least in part on a machine learning model, wherein the evaluating further comprises:

obtaining a score for a candidate value from the machine learning model based on a feature vector that represents the candidate value, wherein the feature vector is associated with at least one feature that represents a data pipeline endorsement of the candidate value by at least one data pipeline, wherein the data pipeline endorsement is weighted based on a consensus score associated with the at least one data pipeline, wherein the consensus score measures a rate at which the at least one data pipeline has historically provided data pipeline endorsements for candidate values that were endorsed by a threshold amount of users;

determining a best scoring candidate value from the set of candidate values; and

associating the field in the page with the best scoring candidate value.

11. The system of claim 10 , wherein the field corresponds to at least one of: a page category field, a website field, a phone number field, an hours of operation field, and a physical address field.

12. The system of claim 10 , wherein the instructions further cause the system to perform:

causing the field in the page to be populated with the best scoring candidate value.

13. The system of claim 10 , wherein the instructions further cause the system to perform:

providing the best scoring candidate value as a recommendation for populating the field in the page.

14. A non-transitory computer-readable storage medium including instructions that, when executed by at least one processor of a computing system, cause the computing system to perform a method comprising:

determining a set of candidate values for a field in a page;

evaluating the set of candidate values for accuracy based at least in part on a machine learning model, wherein the evaluating further comprises:

obtaining a score for a candidate value from the machine learning model based on a feature vector that represents the candidate value, wherein the feature vector is associated with at least one feature that represents a data pipeline endorsement of the candidate value by at least one data pipeline, wherein the data pipeline endorsement is weighted based on a consensus score associated with the at least one data pipeline, wherein the consensus score measures a rate at which the at least one data pipeline has historically provided data pipeline endorsements for candidate values that were endorsed by a threshold amount of users;

determining a best scoring candidate value from the set of candidate values; and

associating the field in the page with the best scoring candidate value.

15. The non-transitory computer-readable storage medium of claim 14 , wherein the field corresponds to at least one of: a page category field, a website field, a phone number field, an hours of operation field, and a physical address field.

16. The non-transitory computer-readable storage medium of claim 14 , wherein the instructions further cause the computing system to perform:

causing the field in the page to be populated with the best scoring candidate value.

17. The non-transitory computer-readable storage medium of claim 14 , wherein the instructions further cause the computing system to perform:

providing the best scoring candidate value as a recommendation for populating the field in the page.