Mixed-grained detection and analysis of user life events for context understanding

A method comprising:

• receiving a user query associated with a user account of a user;
• accessing a context graph linking user life events of the user with predicted activities of the user based on one or more relationships between the user life events and the predicted activities, wherein the predicted activities are inferred by an artificial intelligence (AI) from the user life events, wherein the context graph includes a plurality of nodes for the user life events and the predicted activities and a plurality of edges connecting the plurality of nodes based on the one or more relationships, and wherein each of the plurality of edges has a corresponding weight attributed to a corresponding one of the one or more relationships;
• determining a current context for the user query based on one or more categories of properties of the user query;
• determining one or more of the plurality of nodes associated with the current context;
• processing the user query based on the current context and the one or more of the plurality of nodes, wherein the processing includes connecting the one or more of the plurality of nodes in the context graph;
• and
• outputting a result of the user query based on the processing.

A non-transitory computer readable medium comprising instructions stored thereon to cause one or more processing units to perform operations comprising:

• receiving a user query associated with a user account of a user;
• accessing a context graph linking user life events of the user with predicted activities of the user based on one or more relationships between the user life events and the predicted activities, wherein the predicted activities are inferred by an artificial intelligence (AI) from the user life events, wherein the context graph includes a plurality of nodes for the user life events and the predicted activities and a plurality of edges connecting the plurality of nodes based on the one or more relationships, and wherein each of the plurality of edges has a corresponding weight attributed to a corresponding one of the one or more relationships;
• determining a current context for the user query based on a category associated with the user query;
• determining one or more of the plurality of nodes associated with the current context;
• processing the user query based on the current context and the one or more of the plurality of nodes, wherein the processing includes connecting the one or more of the plurality of nodes in the context graph;
• and
• outputting a result of the user query based on the processing.

A system comprising:

• a non-transitory memory;
• and
• one or more hardware processors coupled to the non-transitory memory and configured to read instructions from the non-transitory memory to cause the system to perform operations comprising:
• receiving a user query associated with a user account of a user;
• accessing a context graph linking user life events of the user with predicted activities of the user based on one or more relationships between the user life events and the predicted activities, wherein the predicted activities are inferred by an artificial intelligence (AI) from the user life events, wherein the context graph includes a plurality of nodes for the user life events and the predicted activities and a plurality of edges connecting the plurality of nodes based on the one or more relationships, and wherein each of the plurality of edges has a corresponding weight attributed to a corresponding one of the one or more relationships;
• determining a current context for the user query based on one or more categories of the user life events associated with the user query;
• determining one or more of the plurality of nodes associated with the current context;
• processing the user query based on the current context and the one or more of the plurality of nodes, wherein the processing includes connecting the one or more of the plurality of nodes in the context graph;
• and
• outputting a result of the user query based on the processing.