Identifying unreliable contributors of user-generated content

• US 8,862,492 B1
• Filed: 04/29/2011
• Issued: 10/14/2014
• Est. Priority Date: 04/29/2011
• Status: Active Grant

• Current Assignee: Google LLC (Alphabet Inc.)
• Original Assignee: Google Inc. (Alphabet Inc.)
• Inventors: Kulshreshtha, Ashutosh, Slep, Mitchell, Cornea, Nicu D., de Alfaro, Luca
• Primary Examiner(s): Koneru, Sujay
• Application Number:
  US13/098,339
• Time in Patent Office: 1,264 Days
• Field of Search: 705 711- 742
• US Class Current: 705/7.32
• CPC Class Codes: G06Q 30/0609 Buyer or seller confidence or verification
• CPC Class Codes: G06Q 50/01 Social networking

Abstract

A system and method for updating and correcting facts that receives proposed values for facts from users and determines a probability that the user providing the values provides unreliable information.

Claims

1. A method, executed by a computer system, for determining the unreliability of a user providing edits to an online map, the method comprising:

analyzing interaction sessions of unreliable users with an online map editing service to identify session characteristics of the unreliable users;

receiving a plurality of interactions by the user with the online map editing service;

identifying a plurality of interaction sessions by the user based on a timing of the plurality of interactions;

determining the identified characteristics for each of the plurality of interaction sessions by the user;

determining by one or more processors a score for each determined characteristic;

determining by the one or more processors using a model trained on the interaction sessions of unreliable users an unreliability score indicative of a probability that the user provides intentionally incorrect edits based at least in part on the determined scores of the determined characteristics; and

responsive to the unreliability score exceeding a threshold, identifying the user as unreliable.

2. The method of claim 1 further comprising:

receiving a second plurality of interactions of the user with an online map hosting service;

identifying the plurality of interaction sessions further based on a second timing of the second plurality of interactions;

identifying types of interactions in the second plurality of interactions of the user; and

wherein the unreliability score is further based on the identified types of interactions in the second plurality of interactions.

3. The method of claim 1 wherein the determined characteristics comprise a measure of elapsed time between interactions within the one of the plurality of interaction sessions and wherein the probability that the user provides intentionally incorrect edits is higher if the elapsed time is lower.

4. The method of claim 1 wherein the determined characteristics comprise a time of day of each of the plurality of interaction sessions.

5. The method of claim 1 wherein the determined characteristics comprise a number of interactions in each of the plurality of interaction sessions and wherein the probability that the user provides intentionally incorrect edits is higher if the number of interactions in each of the plurality of sessions is higher.

6. The method of claim 1 wherein receiving a plurality of interactions comprises retrieving a plurality of interactions.

7. The method of claim 1 wherein the determined characteristics comprise a measure of elapsed time between two of the plurality of sessions and wherein the probability that the user provides intentionally incorrect edits is higher if the elapsed time is lower.

8. A computer system for determining the unreliability of a user providing edits to a map, the system comprising:

one or more processors for executing program code; and

a non-transitory computer-readable storage medium storing executable program code for;

analyzing interaction sessions of unreliable users with an online map editing service to identify session characteristics of the unreliable users;

receiving a plurality of interactions by the user with the on-line map editing service;

identifying a plurality of interaction sessions by the user based on a timing of the plurality of interactions;

determining the identified characteristics for each of the plurality of sessions by the user;

determining a score for each determined characteristic;

determining using a model trained on the interaction sessions of unreliable users an unreliability score indicative of a probability that the user provides intentionally incorrect edits based at least in part on the determined scores for the determined characteristics; and

responsive to the unreliability score exceeding a threshold, identifying the user as unreliable.

9. The system of claim 8 further comprising program code for:

receiving a second plurality of interactions of the user with an online map hosting service;

identifying the plurality of interaction sessions further based on a second timing of the second plurality of interactions;

identifying types of interactions in the second plurality of interactions of the user; and

wherein the unreliability score is further based on the identified types of interactions in the second plurality of interactions.

10. The system of claim 8 wherein the determined characteristics comprise a measure of elapsed time between interactions within the one of the plurality of interaction sessions and wherein the probability that the user provides intentionally incorrect edits is higher if the elapsed time is lower.

11. The system of claim 8 wherein the determined characteristics comprise a time of day of each of the plurality of interaction sessions.

12. The system of claim 8 wherein the determined characteristics comprise a number of interactions in each of the plurality of interaction sessions and wherein the probability that the user provides intentionally incorrect edits is higher if the number of interactions in each of the plurality of sessions is higher.

13. The system of claim 8 wherein the determined characteristics comprise a measure of elapsed time between two of the plurality of sessions and wherein the probability that the user provides intentionally incorrect edits is higher if the measure of elapsed time is shorter.

14. The system of claim 8 wherein receiving a plurality of interactions comprises retrieving a plurality of interactions.

15. A non-transitory computer-readable storage medium storing executable program code for determining the unreliability of a user providing edits to a map, the computer program code comprising program code for:

analyzing interaction sessions of unreliable users with an online map editing service to identify session characteristics of the unreliable users;

receiving a plurality of interactions by the user with the on-line map editing service, the interactions involving edits to the map;

identifying a plurality of interaction sessions by the user based on a timing of the plurality of interactions;

determining the identified characteristics for each of the plurality of interaction sessions by the user;

determining a score for each determined characteristic of the interaction sessions;

determining using a model trained on the interaction sessions of unreliable users an unreliability score indicative of a probability that the user provides intentionally incorrect edits based at least in part on the determined scores for the determined characteristics of the interaction sessions; and

responsive to the unreliability score exceeding a threshold, identifying the user as unreliable.

16. The non-transitory computer-readable storage medium of claim 15 further comprising program code for:

receiving a second plurality of interactions of the user with an online map hosting service;

identifying the plurality of interaction sessions further based on a second timing of the second plurality of interactions;

identifying types of interactions in the second plurality of interactions of the user; and

wherein the unreliability score is further based on the identified types of interactions in the second plurality of interactions.

17. The non-transitory computer-readable storage medium of claim 15 wherein the determined characteristics comprise a time of day of an interaction session.

18. The non-transitory computer-readable storage medium of claim 15 wherein the determined characteristics comprise a number of interactions in an interaction session that the user provides intentionally incorrect edits is higher if the number of interactions is higher.

19. The non-transitory computer-readable storage medium of claim 15 wherein the determined characteristics comprise a measure of elapsed time between two of the plurality of interaction sessions and wherein the probability that the user provides intentionally incorrect edits is higher if the elapsed time is lower.

20. The non-transitory computer-readable storage medium of claim 15, further comprising:

determining characteristics of interactions of the user occurring within an interaction session; and

determining the unreliability score based in part on the characteristics of the user occurring within the interaction session.