Statistical Filtering of Search Results on Online Social Networks

• US 20150363402A1
• Filed: 06/13/2014
• Published: 12/17/2015
• Est. Priority Date: 06/13/2014
• Status: Abandoned Application

• Current Assignee: Meta Platforms, Inc. (f/k/a Facebook, Inc.)
• Original Assignee: Meta Platforms, Inc. (f/k/a Facebook, Inc.)
• Inventors: Jackson, Thomas E., Mashiach, Li-Tal
• Application Number:
  US14/304,596
• Publication Number:
  US 20150363402A1
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06F 16/9535 Search customisation based on user profiles and personalisation
• CPC Class Codes: G06Q 50/01 Social networking

Abstract

In one embodiment, a method includes receiving a query from a first user of an online social network and identifying a set of objects associated with the online social network that substantially match the query. The method also includes calculating, for each identified object, multiple scores corresponding to multiple scoring axes, respectively, each scoring axis having a threshold score that is statistically determined for the scoring axis. The method further includes filtering one or more of the identified objects from the set of objects based on the calculated scores, where each filtered object is associated with one or more scores for one or more scoring axes, respectively, below the threshold score of the respective scoring axis. The method also includes generating one or more search results corresponding to one or more of the identified objects remaining in the set of objects, respectively.

Claims

1. A method comprising, by one or more computing devices:

receiving a query from a first user of an online social network;

identifying a set of objects associated with the online social network that substantially match the query;

calculating, for each identified object, a plurality of scores corresponding to a plurality of scoring axes, respectively, each scoring axis having a threshold score that is statistically determined for the scoring axis;

filtering one or more of the identified objects from the set of objects based on the calculated scores, wherein each filtered object is associated with one or more scores for one or more scoring axes, respectively, below the threshold score of the respective scoring axis;

generating one or more search results corresponding to one or more of the identified objects remaining in the set of objects, respectively, each search result comprising a reference to a corresponding identified object; and

sending, responsive to the query, one or more search results for display to the first user.

2. The method of claim 1, further comprising:

accessing 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 the first user associated with the online social network; and

a plurality of second nodes that each correspond to an object associated with the online social network.

3. The method of claim 2, wherein, for each identified object, the score corresponding to a scoring axis is based at least in part on a degree of separation between the first node and a second node corresponding to the identified object.

4. The method of claim 1, wherein, for each identified object, the score corresponding to a scoring axis is based at least in part on a social-graph affinity of the first user with respect to the identified object.

5. The method of claim 1, wherein, for each identified object, the score corresponding to a scoring axis is based at least in part on a recency value associated with the identified object.

6. The method of claim 1, wherein, for each identified object, the score corresponding to a scoring axis is based at least in part on a calculated topic relevance for the identified object with respect to the query.

7. The method of claim 1, wherein, for each identified object, the score corresponding to a scoring axis is based at least in part on a calculated author quality associated with the identified object.

8. The method of claim 1, wherein, for each identified object, the score corresponding to a scoring axis is based at least in part on a calculated text similarity between the identified object and the query.

9. The method of claim 1, wherein the threshold score associated with a scoring axis is a number of standard deviations below a mean of scores associated with the scoring axis.

10. The method of claim 1, wherein the threshold score associated with a scoring axis is greater than a percentage of scores associated with the scoring axis.

11. The method of claim 1, wherein the threshold score associated with a scoring axis is greater than a number of scores associated with the scoring axis.

12. The method of claim 1, wherein an identified object is filtered from the set of objects if it is associated with two or more scores for two or more scoring axes, respectively, below the threshold scores of the scoring axes.

13. The method of claim 1, wherein the threshold score associated with a scoring axis is based on a step function associated with the scores for the scoring axis.

14. The method of claim 1, further comprising retaining in the set of objects one or more of the identified objects whose scores associated with one or more scoring axes are above one or more respective upper-threshold scores.

15. The method of claim 14, wherein the upper-threshold score associated with a scoring axis is greater than a percentage of scores associated with the scoring axis.

16. The method of claim 14, wherein the upper-threshold score associated with a scoring axis is less than a number of scores associated with the scoring axis.

17. The method of claim 14, wherein an identified object is retained in the set of objects if it is associated with two or more scores for two or more scoring axes, respectively, above the upper-threshold scores of the scoring axes.

18. The method of claim 1, further comprising determining, for each scoring axis, a threshold score, the threshold score based at least in part on the calculated scores associated with the scoring axis.

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

receive a query from a first user of an online social network;

identify a set of objects associated with the online social network that substantially match the query;

calculate, for each identified object, a plurality of scores corresponding to a plurality of scoring axes, respectively, each scoring axis having a threshold score that is statistically determined for the scoring axis;

filter one or more of the identified objects from the set of objects based on the calculated scores, wherein each filtered object is associated with one or more scores for one or more scoring axes, respectively, below the threshold score of the respective scoring axis;

generate one or more search results corresponding to one or more of the identified objects remaining in the set of objects, respectively, each search result comprising a reference to a corresponding identified object; and

send, responsive to the query, one or more search results for display to the first user.

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 a query from a first user of an online social network;

identify a set of objects associated with the online social network that substantially match the query;

calculate, for each identified object, a plurality of scores corresponding to a plurality of scoring axes, respectively, each scoring axis having a threshold score that is statistically determined for the scoring axis;

filter one or more of the identified objects from the set of objects based on the calculated scores, wherein each filtered object is associated with one or more scores for one or more scoring axes, respectively, below the threshold score of the respective scoring axis;

generate one or more search results corresponding to one or more of the identified objects remaining in the set of objects, respectively, each search result comprising a reference to a corresponding identified object; and

send, responsive to the query, one or more search results for display to the first user.