Location-based ranking of search results on online social networks
In one embodiment, a computing system may access a social graph comprising 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, where the nodes comprise a first node corresponding to a first user of an online social network, and a plurality of second nodes that each correspond to a concept or a second user associated with the online social network. The computing system may receive a search query from the first user. The computing system may generate one or more search results corresponding to the search query, where each search result corresponds to a node of the plurality of second nodes. The computing system may score each search result based on a proximity coefficient between the first node and the node corresponding to the search result.
1. A method for disambiguating similar search results, the method comprising, by one or more computing systems:
accessing, by the one or more computing systems, a social graph comprising 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, the nodes comprising:
a first node corresponding to a first user associated with an online social network; and
a plurality of second nodes that each correspond to a concept or a second user associated with the online social network;
receiving, at the one or more computing systems from a client system of the first user, a search query from the first user;
generating, by the one or more computing systems, one or more search results corresponding to the search query, wherein each search result corresponds to a node of the plurality of second nodes;
accessing, by the one or more computing systems, a first location history of the first node, wherein the first location history of the first node comprises:
a first set of geographic locations of the first user; and
one or more timestamps each corresponding to one of the first set of geographic locations;
accessing, by the one or more computing systems, one or more second location histories of the one or more second nodes corresponding to the one or more search results, wherein each second location history of each second node comprises:
a second set of geographic locations of second user corresponding to the second node; and
one or more timestamps each corresponding to one of the second set of geographic locations;
calculating, by the one or more computing systems, a proximity coefficient for each search result corresponding to one of the second plurality of nodes, wherein the proximity coefficient is calculated based on a sum of distances between the first set of geographic locations corresponding to the first node and the second set of geographic locations corresponding to the one of the second nodes during a time interval defined by the one or more timestamps of first location history of the first node and the one or more timestamps of second location history of the second node;
scoring, by the one or more computing systems, each of the search results based on the corresponding proximity coefficient to disambiguate similar search results; and
sending, from the one or more computing systems to the client system of the first user, instructions for presenting one or more of the search results to the first user based on the scores of the respective search results.
2. The method of claim 1 , further comprising:
ranking a plurality of search results based at least in part on the proximity coefficient calculated for each search result.
3. The method of claim 1 , wherein the location history of the first node further comprises at least one timestamp corresponding to a past time period.
4. The method of claim 1 , wherein the proximity coefficient is based on a function ƒ(d 1 ,t 1 ), (d 2 ,t 2 ) . . . (d i ,t i )), wherein (d 1 , d 2 . . . d i ) corresponds to the distance between a geographic location of the first user and a geographic location of each of the search results at time periods (t 1 ,t 2 . . . t i ).
5. The method of claim 1 , wherein the proximity coefficient is further based at least in part on a geographic location of the first user being within a threshold distance of a geographic location associated with the search result for at least a threshold amount of time.
6. The method of claim 1 , wherein the proximity coefficient is further based on a time decay factor.
7. The method of claim 1 , wherein the proximity coefficient is adjusted based on a determination of whether the first user is traveling.
8. The method of claim 7 , wherein determining whether the first user is traveling is based at least in part on the distance between a current location of the first user and a location determined to be the first user's home.
9. The method of claim 1 , wherein each of the search results corresponding to a second node comprises a location history of a second user associated with the search result.
10. The method of claim 9 , wherein the proximity coefficient is based at least in part on determining that a geographic location of the first user was within a threshold distance of a geographic location of the location history of the second user for at least a threshold amount of time.
11. The method of claim 9 , wherein the proximity coefficient is updated in response to the search query.
12. The method of claim 9 , wherein the proximity coefficient is updated periodically without any user input for one or more location histories of one or more particular second users.
13. The method of claim 12 , wherein the one or more particular second users comprise users having an affinity coefficient with respect to the first user exceeding a threshold affinity coefficient.
14. The method of claim 12 , wherein the one or more particular second users comprise users having one or more current locations within a predetermined distance of a current location of the first user.
15. The method of claim 1 , wherein the proximity coefficient determined for the first location history of the first node with respect to the a second location history of a second node is determined to be the proximity coefficient for the second location history of the second node with respect to the first location history of the first node.
16. The method of claim 1 , wherein the search results are scored and ranked based on their respective proximity coefficients.
17. The method of claim 1 , wherein the search results are scored further based at least in part on an affinity coefficient between the first user and each search result.
18. The method of claim 17 , wherein the proximity coefficient is used to adjust an affinity coefficient between the first user and each search result.
19. One or more computer-readable non-transitory storage media embodying software that is operable when executed to:
access a social graph comprising 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, the nodes comprising:
a first node corresponding to a first user associated with an online social network; and
a plurality of second nodes that each correspond to a concept or a second user associated with the online social network;
receiving, from a client system of the first user, a search query from the first user;
generate one or more search results corresponding to the search query, wherein each search result corresponds to a node of the plurality of second nodes;
access a first location history of the first node, wherein the first location history of the first node comprises:
a first set of geographic locations of the first user; and
one or more timestamps each corresponding to one of the first set of geographic locations;
access one or more second location histories of the one or more second nodes corresponding to the one or more search results, wherein each second location history of each second node comprises:
a second set of geographic locations of second user corresponding to the second node; and
one or more timestamps each corresponding to one of the second set of geographic locations;
calculate a proximity coefficient for each search result corresponding to one of the second plurality of nodes, wherein the proximity coefficient is calculated based on a sum of distances between the first set of geographic locations corresponding to the first node and the second set of geographic locations corresponding to the one of the second nodes during a time interval defined by the one or more timestamps of first location history of the first node and the one or more timestamps of second location history of the second node;
score each of the search results based on the corresponding proximity coefficient to disambiguate similar search results; and
send, to the client system of the first user, instructions for presenting one or more of the search results to the first user based on the scores of the respective search results.
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:
access a social graph comprising 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, the nodes comprising:
a first node corresponding to a first user associated with an online social network; and
a plurality of second nodes that each correspond to a concept or a second user associated with the online social network;
receiving, from a client system of the first user, a search query from the first user;
generate one or more search results corresponding to the search query, wherein each search result corresponds to a node of the plurality of second nodes;
access a first location history of the first node, wherein the first location history of the first node comprises:
a first set of geographic locations of the first user; and
one or more timestamps each corresponding to one of the first set of geographic locations;
access one or more second location histories of the one or more second nodes corresponding to the one or more search results, wherein each second location history of each second node comprises:
a second set of geographic locations of second user corresponding to the second node; and
one or more timestamps each corresponding to one of the second set of geographic locations;
calculate a proximity coefficient for each search result corresponding to one of the second plurality of nodes, wherein the proximity coefficient is calculated based on a sum of distances between the first set of geographic locations corresponding to the first node and the second set of geographic locations corresponding to the one of the second nodes during a time interval defined by the one or more timestamps of first location history of the first node and the one or more timestamps of second location history of the second node;
score each of the search results based on the corresponding proximity coefficient to disambiguate similar search results; and
send, to the client system of the first user, instructions for presenting one or more of the search results to the first user based on the scores of the respective search results.
21. The system of claim 20 , wherein the processors are further operable to:
rank a plurality of search results based at least in part on the proximity coefficient calculated for each search result.
22. The system of claim 20 , wherein the location history of the first node further comprises at least one timestamp corresponding to a past time period.
23. The system of claim 20 , wherein the proximity coefficient is based on a function ƒ(d 1 , t 1 ), (d 2 ,t 2 ) . . . (d i ,t i )), wherein (d 1 , d 2 . . . d i ) corresponds to the distance between a geographic location of the first user and a geographic location of each of the search results at time periods (t 1 ,t 2 . . . t i ).
24. The system of claim 20 , wherein the proximity coefficient is further based at least in part on a geographic location of the first user being within a threshold distance of a geographic location associated with the search result for at least a threshold amount of time.
25. The system of claim 20 , wherein the proximity coefficient is further based on a time decay factor.
26. The system of claim 20 , wherein the proximity coefficient is adjusted based on a determination of whether the first user is traveling.
27. The system of claim 26 , wherein determining whether the first user is traveling is based at least in part on the distance between a current location of the first user and a location determined to be the first user's home.
28. The system of claim 20 , wherein each of the search results corresponding to a second node comprises a location history of a second user associated with the search result.
29. The system of claim 28 , wherein the proximity coefficient is based at least in part on determining that a geographic location of the first user was within a threshold distance of a geographic location of the location history of the second user for at least a threshold amount of time.
30. The system of claim 28 , wherein the proximity coefficient is updated in response to the search query.
31. The system of claim 28 , wherein the proximity coefficient is updated periodically without any user input for one or more location histories of one or more particular second users.
32. The system of claim 31 , wherein the one or more particular second users comprise users having an affinity coefficient with respect to the first user exceeding a threshold affinity coefficient.
33. The system of claim 31 , wherein the one or more particular second users comprise users having one or more current locations within a predetermined distance of a current location of the first user.
34. The system of claim 20 , wherein the proximity coefficient determined for the first location history of the first node with respect to a second location history of a second node is determined to be the proximity coefficient for the second location history of the second node with respect to the first location history of the first node.
35. The system of claim 20 , wherein the search results are scored and ranked based on their respective proximity coefficients.
36. The system of claim 20 , wherein the search results are scored further based at least in part on an affinity coefficient between the first user and each search result.
37. The system of claim 36 , wherein the proximity coefficient is used to adjust an affinity coefficient between the first user and each search result.