Venue prediction based on ranking
View Patent ↗A technique is disclosed for presenting possible cheek-in venues to a mobile device user, such as for the purpose of checking in with social networking services such as Foursquare. The disclosed technique operates by training a model during an initial training sequence. The trained model, representing a teamed hypothesis, describes venues that are ranked according to their relevance to users who are also described by the model. During a subsequent run-time sequence, the system receives real-time geolocation data that represents users, wherein the data includes user geolocation and the level of accuracy of the geolocation. The system generates a set of candidate venues, ranks the generated candidate venues by applying the learned hypothesis, which can be dependent on the level of accuracy, and presents the user with the ranked venues as possible cheek-in venues. The user can then select the venue actually being checked into from the presented check-in venues.
1. A system comprising:
a data-processing system comprising one or more processors communicatively coupled to a data store, the data-processing system further comprising memory coupled to the one or more processors and storing instructions that, when executed by the one or more processors, cause the data-processing system to:
receive a first geolocation of a first user;
generate a non-empty set of candidate venues based at least in part on the first geolocation;
determine whether a level of accuracy of the first geolocation of the first user satisfies an accuracy threshold;
consequent to a determination that the level of accuracy of the first geolocation of the first user does not satisfy the accuracy threshold, rank the candidate venues, wherein the ranking results in a non-empty set of possible check-in venues; and
transfer one or more of the possible check-in venues from the non-empty set of possible check-in venues to facilitate presentation of the one or more of the possible check-in venues with a device.
2. The system of claim 1 , wherein the ranking the candidate venues is based at least in part on applying to the candidate venues a hypothesis that is learned from locations of one or more venues that are visited by at least one of i) the first user and/or ii) one or more additional users, wherein the hypothesis is selected from a set of two or more hypotheses based at least in part on the first geolocation.
3. The system of claim 2 , wherein the hypothesis is selected from the set of two or more hypotheses based at least in part on the level of accuracy of the first geolocation or a distance of the first geolocation to at least one other venue.
4. The system of claim 2 , wherein the level of accuracy corresponds to an accuracy radius with respect to the first geolocation.
5. The system of claim 4 , wherein a confidence indicator is attributed to the level of accuracy.
6. The system of claim 1 , wherein the instructions further cause the data-processing system to:
receive a calendrical time identifier of when the first user was at the first geolocation;
determine, based at least in part on the level of accuracy of the first geolocation, whether to use the calendrical time identifier in processing the non-empty set of candidate venues; and
consequent to a determination based at least in part on the level of accuracy of the first geolocation to use the calendrical time identifier, process the non-empty set of candidate venues based at least in part on the calendrical time identifier.
7. The system of claim 6 , wherein the processing the non-empty set of candidate venues based at least in part on the calendrical time identifier comprises:
determining a set of venue categories based at least in part on the calendrical time identifier; and
including and/or excluding one or more candidate venues based at least in part on the determined set of venue categories; and/or
ranking the candidate venues based at least in part on the determined set of venue categories.
8. One or more non-transitory, processor-readable media storing instructions that, when executed by a data-processing system, cause the data-processing system to:
receive a first geolocation of a first user;
generate a non-empty set of candidate venues based at least in part on the first geolocation;
determine whether a level of accuracy of the first geolocation of the first user satisfies an accuracy threshold;
consequent to a determination that the level of accuracy of the first geolocation of the first user does not satisfy the accuracy threshold, rank the candidate venues, wherein the ranking results in a non-empty set of possible check-in venues; and
transfer one or more of the possible check-in venues from the non-empty set of possible check-in venues to facilitate presentation of the one or more of the possible check-in venues with a device.
9. The one or more non-transitory, processor-readable media of claim 8 , wherein the ranking the candidate venues is based at least in part on applying to the candidate venues a hypothesis that is learned from locations of one or more venues that are visited by at least one of i) the first user and/or ii) one or more additional users, wherein the hypothesis is selected from a set of two or more hypotheses based at least in part on the first geolocation.
10. The one or more non-transitory, processor-readable media of claim 9 , wherein the hypothesis is selected from the set of two or more hypotheses based at least in part on the level of accuracy of the first geolocation or a distance of the first geolocation to at least one other venue.
11. The one or more non-transitory, processor-readable media of claim 9 , wherein the level of accuracy corresponds to an accuracy radius with respect to the first geolocation.
12. The one or more non-transitory, processor-readable media of claim 11 , wherein a confidence indicator is attributed to the level of accuracy.
13. The one or more non-transitory, processor-readable media of claim 8 , wherein the instructions further cause the data-processing system to:
receive a calendrical time identifier of when the first user was at the first geolocation;
determine, based at least in part on the level of accuracy of the first geolocation, whether to use the calendrical time identifier in processing the non-empty set of candidate venues; and
consequent to a determination based at least in part on the level of accuracy of the first geolocation to use the calendrical time identifier, process the non-empty set of candidate venues based at least in part on the calendrical time identifier.
14. The one or more non-transitory, processor-readable media of claim 8 , wherein the processing the non-empty set of candidate venues based at least in part on the calendrical time identifier comprises:
determining a set of venue categories based at least in part on the calendrical time identifier; and
including and/or excluding one or more candidate venues based at least in part on the determined set of venue categories; and/or
ranking the candidate venues based at least in part on the determined set of venue categories.
15. A method comprising:
receiving by a data-processing system a first geolocation of a first user;
generating by the data-processing system a non-empty set of candidate venues based at least in part on the first geolocation;
determining by the data-processing system whether a level of accuracy of the first geolocation of the first user satisfies an accuracy threshold;
consequent to a determination that the level of accuracy of the first geolocation of the first user does not satisfy the accuracy threshold, ranking by the data-processing system the candidate venues, wherein the ranking results in a non-empty set of possible check-in venues; and
transmitting by the data-processing system one or more of the possible check-in venues from the non-empty set of possible check-in venues to facilitate presentation of the one or more of the possible check-in venues with a device.
16. The method of claim 15 , wherein the ranking the candidate venues is based at least in part on applying to the candidate venues a hypothesis that is learned from locations of one or more venues that are visited by at least one of i) the first user and/or ii) one or more additional users, wherein the hypothesis is selected from a set of two or more hypotheses based at least in part on the first geolocation.
17. The method of claim 16 , wherein the hypothesis is selected from the set of two or more hypotheses based at least in part on the level of accuracy of the first geolocation or a distance of the first geolocation to at least one other venue.
18. The method of claim 16 , wherein the level of accuracy corresponds to an accuracy radius with respect to the first geolocation, and wherein a confidence indicator is attributed to the level of accuracy.
19. The method of claim 15 , further comprising:
receiving, by the data-processing system, a calendrical time identifier of when the first user was at the first geolocation;
determining, by the data-processing system and based at least in part on the level of accuracy of the first geolocation, whether to use the calendrical time identifier in processing the non-empty set of candidate venues; and
consequent to a determination based at least in part on the level of accuracy of the first geolocation to use the calendrical time identifier, processing by the data-processing system the non-empty set of candidate venues based at least in part on the calendrical time identifier.
20. The method of claim 19 , wherein the processing the non-empty set of candidate venues based at least in part on the calendrical time identifier comprises:
determining a set of venue categories based at least in part on the calendrical time identifier; and
including and/or excluding one or more candidate venues based at least in part on the determined set of venue categories; and/or
ranking the candidate venues based at least in part on the determined set of venue categories.