Autocomplete using social activity signals

A system comprising:

• a machine having a memory and at least one processor;
• and
• a memory storing instructions that, when executed by the at least one processor, causes the machine to perform operations comprising:
• accessing a plurality of social activity signals associated with a user and indicative of actions that are performed by the user and viewable by other users;
• receiving, from the user, user-entered text in a search field for a search engine;
• determining predicted queries based on the user-entered text and the plurality of social activity signal, each one of the predicted queries comprising predicted text and at least a portion of the user-entered text, the predicted text being absent from the user-entered text, the determining comprising:
• determining potential predicted queries based on the user-entered text;
• and
• assigning a corresponding predicted query value to each one of the potential predicted queries based on a determination for each potential predicted query of whether the potential predicted query corresponds to one of the social activity signals indicative of an action performed by the user that is viewable by other users, the assigning including:
• for each potential predicted query determined to correspond to one of the plurality of social activity signals indicative of an action performed by the user that is viewable by other users, determining a corresponding social activity type of the one of the plurality of social activity signals indicative of an action performed by the user that is viewable by other users;
• and
• for each potential predicted query determined to correspond to the one of the plurality of social activity signals, calculating the corresponding predicted query value based on a corresponding weight of the corresponding social activity type;
• and
• causing the predicted queries to be displayed, to the user, in an autocomplete user interface element of the search field.

A non-transitory machine-readable storage medium storing a set of instructions that, when executed by at least one processor, causes the at least one processor to perform a set of operations comprising:

• accessing a plurality of social activity signals associated with a user and indicative of actions that are performed by the user and viewable by other users;
• receiving, from the user, user-entered text in a search field for a search engine;
• determining predicted queries based on the user-entered text and the plurality of social activity signals, each one of the predicted queries comprising predicted text and at least a portion of the user-entered text, the predicted text being absent from the user-entered text the determining comprising:
• determining potential predicted queries based on the user-entered text;
• and
• assigning a corresponding predicted query value to each one of the potential predicted queries based on a determination for each potential predicted query of whether the potential predicted query corresponds to one of the social activity signals indicative of an action performed by the user that is viewable by other users, the assigning including:
• for each potential predicted query determined to correspond to one of the plurality of social activity signals indicative of an action performed by the user that is viewable by other users, determining a corresponding social activity type of the one of the plurality of social activity signals indicative of an action performed by the user that is viewable by other users;
• and
• for each potential predicted query determined to correspond to the one of the plurality of social activity signals, calculating the corresponding predicted query value based on a corresponding weight of the corresponding social activity type;
• and
• causing the predicted queries to be displayed, to the user, in an autocomplete user interface element of the search field.