Generating user profiles
AssigneeGoogle Inc.
InventorsXuefu Wang, Sarah Sirajuddin, Oren E. Zamir, Aitan Weinberg
Life of this patent
Estimating the term…
Abstract
Methods, systems, and apparatus, including computer program products, for generating user identifier profiles. A user identifier profile characterizes a user based on events that occurred during past user sessions for a user (e.g., past online activities). An event is an action that occurs during a user session, such as a web page view, an advertisement click-through, and a conversion. A user identifier profile can be used, for example, to select advertisements targeted to the user. A user identifier profile includes information about inferred user interests and inferred user demographics.
Claims20
A computer-implemented method, comprising:
- identifying event data for a user identifier based on past user sessions associated with the user identifier that occurred during a first time period, the event data specifying events that occurred during the past user sessions;
- categorizing, by one or more processors, each of the events as belonging to one or more vertical categories;
- for each vertical category, determining a user identifier interest weight for the user identifier based on the events that belong to the vertical category;
- generating long-term user identifier profile data for the user identifier based on the user identifier interest weights from the event data that are based on past user sessions that occurred during the first time period;
- generating short-term user identifier profile data for the user identifier based on the user identifier interest weights from the event data that are based on past user sessions that occurred during a second time period that is a most recent sub-period of the first time period;
- selecting a candidate advertisement set based on the long-term user identifier profile data;
- and
- adjusting a ranking order of the candidate advertisement set based on the short-term user-identifier profile data.
A system, comprising:
- one or more processors;
- and
- a computer-readable medium coupled to the one or more processors having instructions stored thereon which, when executed by the one or more processors, cause the one or more processors to perform operations comprising:
- identifying event data for a user identifier based on past user sessions associated with the user identifier that occurred during a first time period, the event data specifying events that occurred during the past user sessions;
- categorizing, by one or more processors, each of the events as belonging to one or more vertical categories;
- for each vertical category, determining a user identifier interest weight for the user identifier based on the events that belong to the vertical category;
- generating long-term user identifier profile data for the user identifier based on the user identifier interest weights from the event data that are based on past user sessions that occurred during the first time period;
- generating short-term user identifier profile data for the user identifier based on the user identifier interest weights from the event data that are based on past user sessions that occurred during a second time period that is a most recent sub-period of the first time period;
- selecting a candidate advertisement set based on the long-term user identifier profile data;
- and
- adjusting a ranking order of the candidate advertisement set based on the short-term user identifier profile data.
Software stored in a computer readable medium and comprising instructions executable by a processing system and upon such execution cause the processing system to perform operations comprising:
- identifying event data for a user identifier based on past user sessions associated with the user identifier that occurred during a first time period, the event data specifying events that occurred during the past user sessions;
- categorizing, by one or more processors, each of the events as belonging to one or more vertical categories;
- for each vertical category, determining a user identifier interest weight for the user identifier based on the events that belong to the vertical category;
- generating long-term user identifier profile data for the user identifier based on the user identifier interest weights from the event data that are based on past user sessions that occurred during the first time period;
- generating short-term user identifier profile data for the user identifier based on the user identifier interest weights from the event data that are based on past user sessions that occurred during a second time period that is a most recent sub-period of the first time period;
- selecting a candidate advertisement set based on the long-term user identifier profile data;
- and
- adjusting a ranking order of the candidate advertisement set based on the short-term user identifier profile data.
Dependencies are inferred from the claim text.