Granted Patent B2

US 12,568,253 · App. 18/441,003 · Granted Mar 3, 2026

Inventors: Jayneel Pawar (Tokyo, JP); Hemant Mehta (Tokyo, JP); Uday Kumar Edudula (Tokyo, JP); Ajay Prakash Mangale (Tokyo, JP)

Assignee: 17LIVE JAPAN INC.

H04N21/2187

Abstract

Claims (20)

1 . A server of a livestreaming platform comprising a circuitry including a processor and a storage medium, wherein the circuitry is configured to:

hold, in the storage medium, a user ID for identifying a user on the livestreaming platform and livestreamer-of-interest IDs for identifying livestreamers of interest registered by the user as livestreamers that the user is interested in, the user ID and the livestreamer-of-interest IDs being associated with each other;

provide a plurality of machine learning models for outputting an interest score;

select, by the processor, one of the plurality of machine learning models based on an amount of money spent by the user on the livestreaming platform;

divide, by the processor, the livestreamers of interest into notification-allowed livestreamers of interest and notification-disallowed livestreamers of interest, based on the interest score outputted by the selected one of the plurality of machine learning models;

transmit, by the processor, a notification to the user when at least one of the notification- allowed livestreamers of interest performs a predetermined activity; and

not transmit the notification to the user when any one of the notification-disallowed livestreamers of interest performs the predetermined activity.

2 . The server of claim 1 , wherein the processor selects the notification-allowed livestreamers of interest from among the livestreamers of interest of the user, and registers a result of the selection in the storage medium in association with the user,

wherein the circuitry is further configured to send, by the processor functioning as a notification control unit, upon detection of the predetermined activity by a livestreamer on the livestreaming platform, a notification to a user for which the livestreamer is registered as the notification-allowed livestreamer of interest, and not send the notification to a user for which the livestreamer is registered as the livestreamer of interest but not as the notification-allowed livestreamer of interest.

3 . The server of claim 1 , wherein the processor divides the livestreamers of interest of the user into the notification-allowed livestreamers of interest and the other livestreamers of interest according to a predetermined criterion,

wherein the predetermined criterion is set so that a degree of interest of the user in the notification-allowed livestreamers of interest is higher than a degree of interest of the user in the other livestreamers of interest.

4 . The server of claim 1 , wherein the predetermined activity is start of a livestream.

5 . The server of claim 1 , wherein the circuitry is further configured to, when a non-registered livestreamer who is not registered as the livestreamer of interest of the user satisfies a predetermined recommendation criterion, send, by the processor functioning as a recommendation control unit, a notification to the user recommending that the user register the livestreamer as the livestreamer of interest.

6 . The server of claim 1 , wherein the circuitry is further configured to resend, by the processor functioning as a resending control unit, the notification to the user when a predetermined period of time elapses after the notification-allowed livestreamer of interest of the user starts a livestream without the user participating in the livestream.

7 . A method performed in a server of a livestreaming platform, comprising: holding, in a storage medium, a user ID for identifying a user on the livestreaming platform and livestreamer-of-interest IDs for identifying livestreamers registered by the user as livestreamers that the user is interested in, the user ID and the livestreamer-of-interest IDs being associated with each other;

provide a plurality of machine learning models for outputting an interest score;

select one of the plurality of machine learning models based on an amount of money spent by the user on the livestreaming platform;

dividing the livestreamers of interest into notification-allowed livestreamers of interest and notification-disallowed livestreamers of interest, based on the interest score outputted by the selected one of the plurality of machine learning models;

transmitting a notification to the user when at least one of the notification-allowed livestreamers of interest performs a predetermined activity; and

not transmitting the notification to the user when any one of the notification-disallowed livestreamers of interest performs the predetermined activity.

Assignments (1)

Priority Claims (1)

Continuity (1)

References Cited (37)