Using Related Mentions to Enhance Link Probability on Online Social Networks

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

• receive, from a client system associated with a first user of an online social network, a search query comprising one or more n-grams;
• access a mention table, wherein the mention table comprises a plurality of entries corresponding to a plurality of strings extracted from a corpus of documents, wherein each entry in the mention table comprises:
• (1) a string, (2) a count of a number of times the string is used in the corpus, and (3) a count of a number of times the string is linked to a particular document in the corpus of documents;
• search the mention table to identify one or more entries corresponding to one or more strings that match the one or more n-grams of the search query;
• calculate, for each n-gram of the search query matching an entry in the mention table, a link probability for the n-gram with respect to a particular entity linked in the matching entry in the mention table, the link probability representing a likelihood that the n-gram corresponds to the particular entity;
• identify a plurality of content objects associated with the online social network that match the search query;
• rank the content objects based at least in part on a determination that, for each identified content object, the content object comprises a link to the particular entity of the online social network where the calculated link probability with respect to that particular entity is above a threshold probability;
• and
• send, to the client system responsive to the search query, instructions for presenting one or more search results, the search results corresponding to one or more of the identified content objects, respectively, wherein the search results are displayed in ranked order based on the rankings of the corresponding content objects.

A system comprising:

• one or more processors;
• and a non-transitory memory coupled to the processors comprising instructions executable by the processors, the processors operable when executing the instructions to:
• receive, from a client system associated with a first user of an online social network, a search query comprising one or more n-grams;
• access a mention table, wherein the mention table comprises a plurality of entries corresponding to a plurality of strings extracted from a corpus of documents, wherein each entry in the mention table comprises:
• (1) a string, (2) a count of a number of times the string is used in the corpus, and (3) a count of a number of times the string is linked to a particular document in the corpus of documents;
• search the mention table to identify one or more entries corresponding to one or more strings that match the one or more n-grams of the search query;
• calculate, for each n-gram of the search query matching an entry in the mention table, a link probability for the n-gram with respect to a particular entity linked in the matching entry in the mention table, the link probability representing a likelihood that the n-gram corresponds to the particular entity;
• identify a plurality of content objects associated with the online social network that match the search query;
• rank the content objects based at least in part on a determination that, for each identified content object, the content object comprises a link to the particular entity of the online social network where the calculated link probability with respect to that particular entity is above a threshold probability;
• and
• send, to the client system responsive to the search query, instructions for presenting one or more search results, the search results corresponding to one or more of the identified content objects, respectively, wherein the search results are displayed in ranked order based on the rankings of the corresponding content objects.