Generating snippets on online social networks

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

accessing a plurality of content objects of an online social network, each content object being associated with an entity of the online social network, wherein each content object comprises content of the content object and is associated with metadata;

generating a set of n-grams by extracting one or more n-grams from the contents of the content objects;

calculating, for each extracted n-gram, a quality score for the n-gram based on an occurrence count of the n-gram with respect to one or more map tiles of a geographical map, wherein the occurrence count of the n-gram comprises a count of entities associated with the n-gram that are geographically located in the one or more map tiles;

generating a snippet-module comprising one or more of the extracted n-grams from the set of n-grams having quality-scores greater than a threshold quality-score; and

sending, to a client system of a user of the online social network, instructions for presenting the snippet-module to the user.

2. The method of claim 1 , wherein calculating the quality score for the n-gram comprises:

calculating a similarity score that indicates a degree of similarity between the n-gram and an n-gram type.

3. The method of claim 2 , wherein calculating the quality score for the n-gram comprises:

determining an n-gram type for the n-gram based on the similarity score;

determining whether the n-gram type is informative for each of the entities; and

increasing the quality score based on the similarity score when the n-gram type is informative for the entity.

4. The method of claim 3 , wherein determining an n-gram type for the n-gram based on the similarity score is based on whether the similarity score satisfies a threshold similarity condition.

5. The method of claim 3 , wherein determining the n-gram type for the n-gram comprises:

for each of one or more predetermined n-gram types, calculating a corresponding similarity score based on the one of the predetermined n-gram types and the n-gram; and

selecting one of the predetermined n-gram types for which the corresponding similarity score satisfies a threshold similarity condition, wherein the n-gram type corresponds to the selected one of the predetermined n-gram types.

6. The method of claim 5 , wherein the corresponding similarity score satisfies the threshold similarity condition when the corresponding similarity score satisfies a threshold similarity value and is a maximum of the one or more similarity scores.

7. The method of claim 3 , wherein calculating the quality score for the n-gram further comprises:

decreasing the quality score based on the similarity score when the n-gram type is not informative for each of the entity.

8. The method of claim 3 , wherein the n-gram type comprises one or more of a location type, an event type, a sentiment type, or a topic type.

9. The method of claim 2 , wherein calculating the similarity score based on the n-gram comprises:

determining a count of entities in a first map tile of the geographical map that are associated with the n-gram, wherein the entity of the online social network is in the first map tile; and

determining a count of entities in one or more second map tiles of the geographical map that are associated with the n-gram,

wherein the similarity score is based on a ratio of the count of entities in the first map tile to the count of entities in the second map tiles.

10. The method of claim 9 , wherein calculating the similarity score based on the n-gram further comprises:

determining a count of map tiles that include one or more entities that are associated with the n-gram,

wherein the similarity score is further based on a ratio of the count of map tiles that include one or more entities that are associated with the n-gram to a total number of map tiles in the geographical map.

11. The method of claim 2 , wherein calculating the similarity score based on the n-gram comprises:

increasing the similarity score when the n-gram comprises at least one word or phrase having a periodic meaning.

12. The method of claim 2 , wherein calculating the similarity score based on the n-gram comprises:

identifying a plurality of calendar events associated with a user of the online social network, wherein each of the calendar events is associated with a date or time and is further associated with text that comprises the n-gram;

calculating an average period of time between calendar events that are associated with consecutive dates or times; and

calculating a deviation from the average period of time, wherein the similarity score is based on the deviation.

13. The method of claim 12 , wherein calculating the similarity score based on the n-gram further comprises:

calculating a count of the calendar events that are associated with dates or times that occur within a range of time from a selected date or time and correspond to multiples of the average period, wherein the multiples of the average period are relative to a date or time at which one of the calendar events occurs,

wherein the similarity score is further based on a ratio of the count of the calendar events to a maximum number of time periods in the range of time.

14. The method of claim 2 , wherein calculating the similarity score based on the n-gram comprises:

determining whether the n-gram comprises a sentiment name; and

when the n-gram comprises a sentiment name, increasing the similarity score.

15. The method of claim 2 , wherein calculating the similarity score based on the n-gram comprises:

generating, using a topic tagger, one or more topics based on the n-gram, wherein the similarity score is based on a result of the topic tagger.

16. The method of claim 15 , wherein the similarity score is based on a confidence value generated by the topic tagger.

17. The method of claim 1 , wherein calculating the quality score for the n-gram comprises boosting the quality score when the n-gram comprises one or more specified words or matches a specified pattern.

18. The method of claim 1 , further comprising:

determining whether the extracted n-gram is a synonym for another word or phrase; and

using other word or phrase instead of the extracted n-gram when the extracted n-gram is a synonym for the other word or phrase.

19. The method of claim 1 , wherein the entities are associated with the n-gram when each of the entities is associated with a snippet-module comprising the n-gram.

20. 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 a first entity; and

a plurality of second nodes corresponding to a plurality of second entities located in a plurality of second map tiles, respectively.

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

access a plurality of content objects of an online social network, each content object being associated with an entity of the online social network, wherein each content object comprises content of the content object and is associated with metadata;

generate a set of n-grams by extracting one or more n-grams from the contents of the content objects;

calculate, for each extracted n-gram, a quality score for the n-gram based on an occurrence count of the n-gram with respect to one or more map tiles of a geographical map, wherein the occurrence count of the n-gram comprises a count of entities associated with the n-gram that are geographically located in one or more map tiles;

generate a snippet-module comprising one or more of the extracted n-grams from the set of n-grams having quality-scores greater than a threshold quality-score; and

send, to a client system of a user of the online social network, instructions for presenting the snippet-module to the user.

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

access a plurality of content objects of an online social network, each content object being associated with an entity of the online social network, wherein each content object comprises content of the content object and is associated with metadata;

generate a set of n-grams by extracting one or more n-grams from the contents of the content objects;

calculate, for each extracted n-gram, a quality score for the n-gram based on an occurrence count of the n-gram with respect to one or more map tiles of a geographical map, wherein the occurrence count of the n-gram comprises a count of the entities geographically located in the one or more map tiles;

generate a snippet-module comprising one or more of the extracted n-grams from the set of n-grams having quality-scores greater than a threshold quality-score; and

send, to a client system of a user of the online social network, instructions for presenting the snippet-module to the user.