# US 12,003,817 B2 · Method for dynamically training a system to determine an age rating of media content

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 12,003,817 B2
- Document ID: US12003817B2
- Status: Granted
- Kind: B2
- Application number: 18/309,287
- Filed: 2023-04-28
- Published: 2023-08-24
- Granted: 2024-06-04
- Earliest claimed: 2021-11-05
- Assignees: British Board of Film Classification
- Applicants: British Board of Film Classification
- Inventors: Matthew James Blakemore; Hristo Vladimirov Tomov; Georgi Stefanov Yanev
- Web page: https://patentgrep.com/patent/US12003817B2

## CPC

- H04N 21/4542: {Blocking scenes or portions of the received content, e.g. censoring scenes}
- H04N 21/4756: {for rating content, e.g. scoring a recommended movie}
- H04N 21/2353: {specifically adapted to content descriptors, e.g. coding, compressing or processing of metadata}
- H04N 21/251: {Learning process for intelligent management, e.g. learning user preferences for recommending movies (details of learning user preferences for the retrieval of video data in a video database G06F16/739; computer systems using learning methods G06N3/08)}
- H04N 21/25841: {involving the geographical location of the client (retrieval from the Internet by querying based on geographical locations G06F16/9537; arrangements for identifying locations of receiving stations in broadcast systems H04H60/51; location of the user terminal in data switching networks H04L67/52; services making use of the location of users or terminals in wireless networks H04W4/02; locating users or terminals in wireless networks H04W64/00)}
- H04N 21/25883: {being end-user demographical data, e.g. age, family status or address (arrangements for identifying locations of users in broadcast systems H04H60/52)}
- H04N 21/4524: {involving the geographical location of the client (retrieval from the Internet by querying based on geographical locations G06F16/9537; systems specially adapted for using geographical information in broadcast systems H04H60/70; protocols in which the network application is adapted for the location of the user terminal in communication control or processing H04L67/52; services making use of the location of users or terminals in wireless networks H04W4/02; locating users or terminals in wireless networks H04W64/00)}
- H04N 21/4532: {involving end-user characteristics, e.g. viewer profile, preferences (monitoring of user activities for profile generation for accessing a video database G06F16/739; user profiles in network data switching protocols H04L67/306; processing of user preferences or user profiles in wireless networks H04W8/18)}
- H04N 21/4662: {characterized by learning algorithms}
- H04N 21/4665: {involving classification methods, e.g. Decision trees}
- H04N 21/84: Generation or processing of descriptive data, e.g. content descriptors {(systems specially adapted for using meta-information in broadcast systems H04H60/73)}
- H04N: PICTORIAL COMMUNICATION, e.g. TELEVISION

## Abstract

A system and method for dynamically training a system to determine an age rating for media content. An exemplary method includes obtaining age rating data for a plurality of territories; determining, based on the age rating data, a similarity vector relating to the target territory; determining, for the similarity vector, a territory associated with a highest prediction score; in response to determining that the territory associated with the highest prediction score is not the source territory, generating a training dataset comprising the age rating data for the target territory, the source territory, and the territory associated with the highest prediction score; and executing a machine learning model, trained by the training dataset, to output an age rating for a content item in the target territory based on an age rating for the content item in the source territory.

## Claims

### Claim 1

A computer-implemented method for dynamically training a system to determine an age rating for media content, the computer-implemented method comprising:
receiving a plurality of age ratings of a plurality of content items for a source territory, a target territory, and at least one other territory;
determining, based on the plurality of age ratings, a prediction score indicative of a similarity between age ratings of the source territory and the target territory and at least one other prediction score indicative of a similarity between age ratings of the at least one other territory and the target territory;
determining, from the source territory and the at least one other territory, a territory associated with a highest prediction score;
executing, on an input content item with a first age rating in the source territory, a machine learning model that outputs a second age rating of the input content item in the target territory, wherein the machine learning model is trained using age ratings of at least the target territory, the territory associated with the highest prediction score and the source territory when the territory associated with the highest prediction score is not the source territory; and
in response to determining that the second age rating exceeds a parental control threshold, blocking access the input content item.

### Claim 2 (depends on 1)

The computer-implemented method of claim 1, wherein the parental control threshold and the first age rating are set in a first rating scale of the source territory, and wherein the second age rating is set in a second rating scale of the target territory and is incompatible for direct comparison with the parental control threshold.

### Claim 3 (depends on 1)

The computer-implemented method of claim 1, further comprising generating an alert indicative of a parental control when the input content item is requested for access.

### Claim 4 (depends on 3)

The computer-implemented method of claim 3, wherein the input content item is accessed on a media content platform that is accessible in the source territory and the target territory, and wherein the alert is generated on a graphical user interface of the media content platform.

### Claim 5 (depends on 1)

The computer-implemented method of claim 1, wherein the input content item comprises media content.

### Claim 6 (depends on 1)

The computer-implemented method of claim 1, wherein the plurality of content items comprises thousands of content items.

### Claim 7 (depends on 1)

The computer-implemented method of claim 1, wherein the at least one other territory comprises a plurality of territories.

### Claim 8 (depends on 1)

The computer-implemented method of claim 1, wherein determining the prediction score and the at least one other prediction score further comprises determining a similarity vector by generating a similarity matrix comprising a plurality of similarity vectors for a plurality of pairs of territories.

### Claim 9 (depends on 8)

The computer-implemented method of claim 8, further comprising clustering the plurality of similarity vectors into clusters.

### Claim 10 (depends on 1)

The computer-implemented method of claim 1, further comprising determining one or more highest prediction scores by: (a) selecting one or more prediction scores exceeding a threshold, or (b) selecting a predetermined number of prediction scores having a highest value, or (c) selecting prediction scores within a set number of standard deviations above a mean prediction score.

### Claim 11

A system for dynamically training a media classification system to determine an age rating for media content, the system comprising:
a memory; and
a processor configured to implement instructions on the memory to configure the media classification system to:
receive a plurality of age ratings of a plurality of content items for a source territory, a target territory, and at least one other territory;
determine, based on the plurality of age ratings, a prediction score indicative of a similarity between age ratings of the source territory and the target territory and at least one other prediction score indicative of a similarity between age ratings of the at least one other territory and the target territory;
determine, from the source territory and the at least one other territory, a territory associated with a highest prediction score;
execute, on an input content item with a first age rating in the source territory, a machine learning model that outputs a second age rating of the input content item in the target territory, wherein the machine learning model is trained using age ratings of at least the target territory, the territory associated with the highest prediction score and the source territory when the territory associated with the highest prediction score is not the source territory; and
in response to determining that the second age rating exceeds a parental control threshold, block access the input content item.

### Claim 12 (depends on 11)

The system of claim 11, wherein the processor is further configured to implement instructions on the memory to set the parental control threshold and the first age rating in a first rating scale of the source territory, and to set the second age rating in a second rating scale of the target territory and that is incompatible for direct comparison with the parental control threshold.

### Claim 13 (depends on 11)

The system of claim 11, wherein the processor is further configured to implement instructions on the memory to generate an alert indicative of a parental control when the input content item is requested for access.

### Claim 14 (depends on 13)

The system of claim 13, wherein the processor is further configured to implement instructions on the memory to access the input content item on a media content platform that is accessible in the source territory and the target territory, and to generate the alert on a graphical user interface of the media content platform.

### Claim 15 (depends on 11)

The system of claim 11, wherein the input content item comprises media content.

### Claim 16 (depends on 11)

The system of claim 11, wherein the processor is further configured to implement instructions on the memory to determine the prediction score and the at least one other prediction score by determining a similarity vector by generating a similarity matrix comprising a plurality of similarity vectors for a plurality of pairs of territories.

### Claim 17 (depends on 16)

The system of claim 16, wherein the processor is further configured to implement instructions on the memory to cluster the plurality of similarity vectors into clusters.

### Claim 18 (depends on 11)

The system of claim 11, wherein the processor is further configured to implement instructions on the memory to determine one or more highest prediction scores by: (a) selecting one or more prediction scores exceeding a threshold, or (b) selecting a predetermined number of prediction scores having a highest value, or (c) selecting prediction scores within a set number of standard deviations above a mean prediction score.