Recommending content with assistant systems
In one embodiment, a method includes, by one or more computing devices, receiving, from a client system associated with a first user of an online social network, an indication of a trigger action by the first user, wherein the trigger action is associated with a user activity, determining a first user context based on the trigger action, accessing one or more recommended content objects associated with the first user context, calculating a recommendation score for each recommended content object, generating one or more content suggestions comprising one of the one or more recommended content objects, respectively, each content suggestion corresponding to a recommended content object having a recommendation score above a threshold recommendation score, and sending, to the client system in response to the trigger action, instructions for presenting one or more of the content suggestions to the first user.
1. A method comprising, by one or more computing systems:
receiving, from a client system associated with a first user of an online social network, an indication of a trigger action by the first user, wherein the trigger action is associated with a user activity;
determining, by the one or more computing systems, a first user context based on the trigger action;
accessing, by the one or more computing systems, one or more recommended content objects associated with the first user context;
calculating, by the one or more computing systems, a recommendation score for each recommended content object;
generating, by the one or more computing systems, one or more content suggestions comprising one of the one or more recommended content objects, respectively, for the first user to share on the online social network, wherein each content suggestion corresponds to a recommended content object having a recommendation score above a threshold recommendation score; and
sending, to the client system in response to the trigger action, instructions for presenting one or more of the content suggestions for the first user to share on the online social network, wherein each content suggestion comprises an activatable element selectable by the first user to share the corresponding recommended content object on the online social network.
2. The method of claim 1 , wherein the trigger action comprises one or more of:
clicking on a composer interface of the online social network;
browsing a content object on the online social network for more than a threshold amount of time;
uploading a content object to the online social network;
updating a user status on the online social network;
checking-in at an entity associated with the online social network;
capturing an image with the client system;
completing an achievement within an online game associated with the online social network;
being located within a threshold distance of a geographic location associated with a particular user context; or
receiving a content object from a second user associated with the online social network.
3. The method of claim 1 , wherein determining the user context based on the trigger action comprises:
identifying, from a context index, one or more user contexts that are associated with the trigger action, wherein each user context is associated with one or more parameters indicative of the user context;
calculating, for each identified user context, a context-probability score for the identified user context based on a comparison of the parameters associated with the identified user context and information associated with the trigger action; and
selecting, by the one or more computing systems, the first user context from the one or more identified contexts based on its calculated context-probability score.
4. The method of claim 3 , wherein the one or more parameters comprise one or more of:
one or more entities associated with the user context;
a geographic location of the one or more entities associated with the user context;
a date associated with the user context; or
a time associated with the user context.
5. The method of claim 1 , wherein the one or more recommended content objects associated with the first user context comprises one or more of:
a comment;
an audio clip;
an image;
a video;
a link;
a message; or
a review.
6. The method of claim 1 , wherein each content suggestion is operable to allow the first user to select a corresponding recommended content object.
7. The method of claim 6 , further comprising:
receiving, from the client system, a selection by the first user of a first content suggestion of the one or more content suggestions; and
generating a post based on the first content suggestion corresponding to a first recommended content object.
8. The method of claim 7 , further comprising:
monitoring one or more social signals associated with the post; and
updating a user profile associated with the first user based on the social signals associated with the post.
9. The method of claim 1 , further comprising generating a context index storing a plurality of contexts by:
accessing a plurality of content objects, wherein each content object comprises one or more social signals;
identifying one or more content objects based on their respective social signals, wherein each content object comprises content metadata;
determining, for each identified content object, one or more user contexts associated with the content object based on its content metadata;
aggregating the identified content objects into one or more aggregated sets based on the user contexts associated with the content objects, wherein each aggregated set corresponds to one or more user contexts, and wherein the content objects in a particular aggregated set have a similarity measure with respect to each other greater than a threshold similarity measure;
selecting one or more of the aggregated sets based on a number of content objects within the respective aggregated set;
generating, for each user context associated with one of the selected aggregated sets, a set of parameters indicative of the user context based on the content metadata of the content objects in the respect aggregated set; and
storing each set of parameters for each user context as a record in the context index.
10. The method of claim 9 , generating the context index storing the plurality of contexts further comprises:
analyzing, for each user context, the content objects of the aggregated set associated with the user context to extract content representative of the user context; and
generating, for each user context, a set of recommended content objects associated with the user context, wherein the set of recommended content objects are stored with the record for the user context in the context index.
11. The method of claim 9 , further comprising:
filtering one or more recommended content objects associated with a negative user sentiment.
12. The method of claim 1 , wherein calculating the recommendation score comprises calculating, for each recommended content object, a global-recommendation score and a personalized-recommendation score for the recommended content object.
13. The method of claim 12 , wherein the personalized-recommendation score for each recommended content object is calculated based on a comparison of content metadata of the recommended content object to content metadata of one or more content objects associated with the first user, wherein a first weight is assigned to the personalized-recommendation score in calculating the recommendation score.
14. The method of claim 12 , wherein the global-recommendation score for each recommended content object is calculated based on a comparison of content metadata of the recommended content object to content metadata of one or more content objects associated with a popular global user context, wherein a first weight is assigned to the global-recommendation score in calculating the recommendation score.
15. The method of claim 1 , wherein the one or more content suggestions to the first user comprise one or more suggestions to perform an action responsive to the trigger action.
16. The method of claim 15 , wherein the suggested action comprises:
capturing an image;
recording a video; or
interacting with a content object.
17. 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, an indication of a trigger action by the first user, wherein the trigger action is associated with a user activity;
determine, by the one or more computing systems, a first user context based on the trigger action;
access, by the one or more computing systems, one or more recommended content objects associated with the first user context;
calculate, by the one or more computing systems, a recommendation score for each recommended content object;
generate, by the one or more computing systems, one or more content suggestions comprising one of the one or more recommended content objects, respectively, for the first user to share on the online social network, wherein each content suggestion corresponds to a recommended content object having a recommendation score above a threshold recommendation score; and
send, to the client system in response to the trigger action, instructions for presenting one or more of the content suggestions for the first user to share on the online social network, wherein each content suggestion comprises an activatable element selectable by the first user to share the corresponding recommended content object on the online social network.
18. 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, an indication of a trigger action by the first user, wherein the trigger action is associated with a user activity;
determine, by the one or more computing systems, a first user context based on the trigger action;
access, by the one or more computing systems, one or more recommended content objects associated with the first user context;
calculate, by the one or more computing systems, a recommendation score for each recommended content object;
generate, by the one or more computing systems, one or more content suggestions comprising one of the one or more recommended content objects, respectively, for the first user to share on the online social network, wherein each content suggestion corresponds to a recommended content object having a recommendation score above a threshold recommendation score; and
send, to the client system in response to the trigger action, instructions for presenting one or more of the content suggestions for the first user to share on the online social network, wherein each content suggestion comprises an activatable element selectable by the first user to share the corresponding recommended content object on the online social network.