User information determination systems and methods
View Patent ↗This disclosure relates to systems and methods for determining information associated with a user. In certain embodiments, various attributes, interest, and/or other demographic information related to a user may be determined based on information may be obtained from personal electronic devices associated with the user. User attribute models used to predict user attributes and/or demographics may be generated and trained based on inferred user interests and available label data. In some embodiments, information reflecting which applications and/or types of applications are installed and/or frequently used by a user on their personal electronic devices may be used in connection with determining interests associated with the user. Further embodiments of the systems and methods disclosed herein relate to determining various information associated with a user based on location information indicative of a user's location obtained from one or more personal electronic devices associated with the user.
1. A method performed by a system comprising a processor and a non-transitory computer-readable storage medium storing instructions that, when executed, cause the system to perform the method, the method comprising:
receiving, at an interface of the system, application usage information from an electronic device associated with a user, the application usage information being associated with an application installed on the electronic device;
mapping the application usage information to one or more interest taxonomies to identify one or more interests associated with the user;
determining one or more relative adjusted weights associated with the identified one or more interests based on the application usage information, wherein determining the one or more relative adjusted weights comprises:
determining one or more decay rates based an indication of a momentum associated with the application, the momentum being determined based on a first density of use of the application over a first time period and a second density of use of the application over a second time period, the second time period being longer than the first time period; and
adjusting one or more initial relative weights based on the one or more decay rates to generate the one or more one or more relative adjusted rates;
associating the one or more relative adjusted weights with the identified one or more interests to generate one or more weighted interests;
identifying one or more content items based on the one or more weighted interests; and
transmitting the one or more content items to the electronic device.
2. The method of claim 1 , wherein mapping the application usage information to the one or more interest taxonomies comprises:
identifying at least one application category associated with the application usage information; and
mapping the at least one application category to the one or more interest taxonomies to identify the one or more interests associated with the user.
3. The method of claim 2 , wherein identifying the at least one application category comprises deriving the at least one application category based on an available description of the application.
4. The method of claim 1 , wherein the application usage information comprises at one or more of information indicating usage of the application on the electronic device, information indicating installation of the application on the electronic device, information indicating usage of particular application features on the electronic device, information indicating a time of usage of the application on the electronic device, information indicating a frequency of usage of the application on the electronic device, and information indicating a duration of usage of the application on the electronic device.
5. The method of claim 4 , wherein the application usage information comprises time-series information.
6. The method of claim 5 , wherein determining the one or more relative weights associated with the identified one or more interests comprises identifying usage events of the application within a defined time window based on the application usage information.
7. The method of claim 1 , wherein the second time period comprises an installed lifetime of the application.
8. The method of claim 1 , wherein the indication of the momentum associated with the application is based on a regularity of use of the application.
9. The method of claim 1 , wherein the indication of the momentum associated with the application is based on a recency of use of the application.
10. The method of claim 1 , wherein the indication of the momentum associated with the application is based on a pattern of use of the application.