Granted Patent B2

US 12,610,097 · App. 18/130,822 · Granted Apr 21, 2026

Inventors: Peter Docherty (Glasgow, GB); Christopher McGuire (Glasgow, GB); Edward Young (Glasgow, GB)

Assignee: THINKANALYTICS LTD.

H04N21/25891G06Q30/0269H04N21/252H04N21/25841H04N21/2668H04N21/44222H04N21/4532H04N21/4661H04N21/4665H04N21/4667H04N21/4668H04N21/482H04N21/8126

Abstract

Claims (39)

1 . A computer-implemented method of determining affinity profiles, comprising:

for each of a plurality of user devices, monitoring user activity including identifying items of content selected for viewing by the user of the user device;

obtaining metadata concerning the selected items of content, the metadata representing at least some properties of the selected items of content;

generating or updating a user record for the user, the user record comprising or representing the user activity and comprising the associated content metadata; and

processing the user record to generate an affinity profile for the user based on the user record, the affinity profile comprising affinity categories and an affinity category score for each of the affinity categories, wherein each affinity category score represents a level of interest in the subject area for the respective affinity category,

wherein the affinity categories are selected from a stored set of affinity categories, each representing a user's affinity for a respective subject area,

wherein the processing of the user record to generate an affinity profile comprises applying a process to convert the metadata in the user record into the affinity categories and the affinity category score for each of the affinity categories,

wherein the process includes algorithms applying weightings or confidence scores for converting the metadata in the user record to obtain the affinity categories and the affinity category score for each of the affinity categories, and

wherein applying the process for converting the metadata in the user record into the affinity categories and the affinity category score for each of the affinity categories comprises applying a machine learning model to the user record, the machine learning model being configured to output the affinity profile based on the metadata in the user record.

2 . The method of claim 1 , performed in conjunction with a method of providing television content or other content to each of the plurality of user devices using a content distribution system, wherein each user device displays an electronic programme guide (EPG) or other user interface that is operable by a user to select one or more items of television or other content, and in response to the selections the distribution system distributes the selected items of television content or other content to the user devices for viewing or other consumption by the users during content viewing sessions.

3 . The method according to claim 2 , further comprising outputting the determined affinity profiles for the plurality of users via an API or operator interface thereby making the affinity profiles available to a third party external to the content distribution system.

4 . The method of claim 2 , wherein the user interface is operable by a user to select one or more items of content of content types other than television content,

the obtaining of metadata concerning the selected items of content, the generating or updating the user record for the user, and the processing the user record to generate the affinity profile for the user are performed using the content metadata obtained for both selected items of television content and selected items of content of the other content types, and

the other content types comprises at least one of computer games, books, music, spoken word content, other audio content, newspapers, or magazines.

5 . The method of claim 1 , wherein the machine learning model comprises or uses a text classifier that is operable to receive text input relating to television content, or other content, as part of the user record and to output the affinity profile.

6 . The method of claim 5 , wherein the text classifier comprises a zero shot classifier.

7 . The method of claim 1 , wherein the generating of the affinity profile is independent of, or is performed without taking into account, data representing socio-economic or demographic status of the user.

8 . The method of claim 7 , wherein at least one of a), b) or c):

a) the user record for each user does not include any data representing socio-economic or demographic status of the user;

b) the user record for each user is anonymous; or

c) the obtaining of metadata, the generating or updating of user records, and the processing of the user records to generate affinity profiles are performed by processing circuitry that does not have access to at least one of or all of data representing socio-economic or demographic status of the users or identity of the users.

9 . The method of claim 1 , wherein the stored set of affinity categories comprises at least 100 affinity categories or sub-categories.

10 . The method of claim 1 , wherein the metadata concerning the selected items of content is obtained from a stored ontology that includes at least 10,000 features that can be used as meta data to represent items of content.

11 . The method of claim 10 , wherein the stored ontology includes enriched versions of metadata obtained for items of content.

12 . The method of claim 1 , wherein the user record comprises or represent user activity for different time windows during a day or week, and the method comprises generating different affinity profiles for the user for the different time windows.

13 . The method of claim 1 , further comprising categorizing a user into one of a plurality of categories based on the affinity profile.

14 . The method of claim 1 , wherein at least one of: the affinity profile for the user comprises a set of scores, each score being for a respective one of the affinity categories, wherein the user is a user account.

15 . The method according to claim 14 , wherein a plurality of individuals has access to the user account or wherein a plurality of user devices are associated with the user account, and wherein the affinity profile for the user account is based on selection of content by at least one of the plurality of individuals or using the plurality of user devices.

16 . The method according to claim 1 , further comprising selecting additional television content or other content to push to the user based on the determined affinity profile for the user.

17 . A system comprising processing circuitry configured to:

for each of a plurality of user devices, monitor user activity including identifying items of television content or other content selected for viewing by the user;

obtain metadata concerning the selected items of television content or other content, the metadata representing at least some properties of the selected items of television content or other content;

generate or update a user record for the user, the user record comprising or representing at least one of the user activity and comprising the associated content metadata; and

process the user record to generate an affinity profile for the user based on the user record, the affinity profile comprising affinity categories and an affinity category score for each of the affinity categories, wherein each affinity category score represents a level of interest in the subject area for the respective affinity category,

wherein the affinity categories are selected from a stored set of affinity categories, each representing a user's affinity for a respective subject area,

wherein processor circuitry is configured to process the user record to generate an affinity profile that is configured to apply a process to convert the metadata in the user record into the affinity categories and the affinity category score for each of the affinity categories,

wherein the process includes algorithms applying weightings or confidence scores for converting the metadata in the user record to obtain the affinity categories and the affinity category score for each of the affinity categories, and

wherein the processor circuitry is further configured to apply the process to convert the metadata in the user record into the affinity categories and the affinity category score for each of the affinity categories comprises being configured to apply a machine learning model to the user record, the machine learning model being configured to output the affinity profile based on the metadata in the user record.

18 . A non-transitory computer-readable medium that comprises computer-readable instructions that are executable to perform a method according to claim 1 .

Assignments (1)

Priority Claims (1)

Continuity (1)

References Cited (165)