Ranking of place-entities on online social networks

1. A method comprising, by one or more computing devices of an online social network:

receiving, from a client system of a first user of the online social network, comparison information associated with two or more particular place-entities, each particular place-entity being associated with a particular score on a first scoring scale and a first feature;

ranking a scored list of place-entities associated with the online social network based in part on the comparison information, wherein each place-entity of the scored list is associated with a score on the first scoring scale, and wherein the place-entities of the scored list comprise at least the two or more particular place-entities; and

determining a place-classifier algorithm based at least in part on the ranked list of place-entities and one or more features associated with each place-entity of the ranked list, wherein the one or more features comprises at least the first feature.

2. The method of claim 1 , wherein the first user is connected to each particular place-entity within a social graph of the online social network, wherein the social graph comprises a plurality of nodes and a plurality of edges connecting the nodes, each of the edges between two of the nodes representing a single degree of separation between them, and wherein the first user corresponds to a first node in the social graph and a plurality of place-entities correspond to a plurality of second nodes in the social graph, respectively.

3. The method of claim 2 , wherein the first user being connected to each particular place-entity within the social graph corresponds to the first user having previously performed, via the online social network, a social-networking interaction with each of the two or more particular place-entities.

4. The method of claim 3 , wherein the social-networking interaction comprises one or more of a check-in, a like, or a review submitted by the first user.

5. The method of claim 1 , wherein the received comparison information comprises an indication of which place-entity, of the two or more particular place-entities, the first user prefers.

6. The method of claim 1 , wherein the each particular place-entity being associated with the particular score on the first scoring scale corresponds to each particular place-entity having a score on the first scoring scale within a threshold of the particular score.

7. The method of claim 6 , wherein each particular place-entity having the score on the first scoring scale within the threshold of the particular score corresponds to the score being within 10% of the particular score.

8. The method of claim 1 , wherein the each particular place-entity being associated with the particular score on the first scoring scale corresponds to each particular place-entity having a score on the first scoring scale equal to the particular score.

9. The method of claim 1 , further comprising receiving a plurality of other comparison information from a respective plurality of other users of the online social network.

10. The method of claim 1 , wherein the ranked list of place-entities further comprises a plurality of other place-entities about which other comparison information was received from other users of the online social network.

11. The method of claim 1 , wherein ranking the scored list of place-entities comprises performing a topological sorting operation based on the comparison information and a plurality of other comparison information received from other users of the online social network.

12. The method of claim 1 , further comprising:

receiving a request for a list of top N place-entities having a particular feature, wherein N is a positive integer; and

filtering the ranked list based on the particular feature to determine the top N place-entities having the particular feature.

13. The method of claim 1 , further comprising:

receiving a request for a list of top N place-entities near a particular location, wherein Nis a positive integer; and

filtering the ranked list to determine the top N place-entities having a location within a threshold distance from the particular location.

14. The method of claim 1 , further comprising applying the place-classifier algorithm to predict a result of a pairwise comparison of two place-entities presented to the first user.

15. The method of claim 1 , further comprising applying the place-classifier algorithm to generate a list of top-ranked place-entities.

16. The method of claim 1 , wherein the place-classifier algorithm is a global classifier configured to generate a non-personalized list of place-entities based at least in part on global features of the place-entities.

17. The method of claim 1 , wherein the place-classifier algorithm is a personalized classifier configured to generate a list of place-entities customized for the first user of the online social network based at least in part on user-specific features.

18. The method of claim 1 , wherein the client system is a mobile computing system, and wherein the comparison information is provided by the first user via a webpage of the online social network or on an interface of a native application associated with the online social network.

19. One or more computer-readable non-transitory storage media embodying software that is operable when executed to:

receive, from a client system of a first user of the online social network, comparison information associated with two or more particular place-entities, each particular place-entity being associated with a particular score on a first scoring scale and a first feature;

rank a scored list of place-entities associated with the online social network based in part on the comparison information, wherein each place-entity of the scored list is associated with a score on the first scoring scale, and wherein the place-entities of the scored list comprise at least the two or more particular place-entities; and

determine a place-classifier algorithm based at least in part on the ranked list of place-entities and one or more features associated with each place-entity of the ranked list, wherein the one or more features comprises at least the first feature.

20. A system comprising: one or more processors; and a memory coupled to the processors comprising instructions executable by the processors, the processors operable when executing the instructions to:

receive, from a client system of a first user of the online social network, comparison information associated with two or more particular place-entities, each particular place-entity being associated with a particular score on a first scoring scale and a first feature;

rank a scored list of place-entities associated with the online social network based in part on the comparison information, wherein each place-entity of the scored list is associated with a score on the first scoring scale, and wherein the place-entities of the scored list comprise at least the two or more particular place-entities; and

determine a place-classifier algorithm based at least in part on the ranked list of place-entities and one or more features associated with each place-entity of the ranked list, wherein the one or more features comprises at least the first feature.