# US 10,769,548 B2 · Value model for sending notifications

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

- Number: US 10,769,548 B2
- Document ID: US10769548B2
- Status: Granted
- Kind: B2
- Application number: 15/185,581
- Filed: 2016-06-17
- Published: 2017-12-21
- Granted: 2020-09-08
- Earliest claimed: 2016-06-17
- Assignees: Facebook, Inc.
- Applicants: Facebook, Inc.
- Inventors: Aleksandar Ilic; Ariel Benjamin Evnine; Ashwin Murthy; Yiyu Li; Konstantine Oleksiyovich Kolomoyskyy; Florin Ratiu
- Web page: https://patentgrep.com/patent/US10769548B2

## CPC

- G06N 20/00: Machine learning
- G06Q 10/40: Business processes related to social networking or social networking services
- H04L 51/224: providing notification on incoming messages, e.g. pushed notifications of received messages
- H04L 51/52: for supporting social networking services
- H04L 67/55: Push-based network services
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06Q: INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- H04L: TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION (arrangements common to telegraphic and telephonic communication H04M)

## Abstract

In one embodiment, a method includes sending, through a communications network, several volumes of notifications corresponding to a first notification type to multiple users and several volumes of notifications corresponding to a second notification type to multiple users. The method further determines visitation impacts of the volumes of notifications of the first and second notification types and trains a machine-learning model based on the visitation impacts. The machine-learning model generates an assessment of a likelihood of interaction by a recipient user with each of the notifications.

## Claims

### Claim 1

A system comprising: one or more processors, and a memory coupled to the processors comprising instructions executable by the processors, the processors being operable when executing the instructions to:
send, through a communications network, sets of first notifications corresponding to a first notification type, wherein each of the sets of first notifications has a corresponding volume and is sent to a plurality of users, and the volumes corresponding to the sets of first notifications are different;
send, through the communications network, sets of second notifications corresponding to a second notification type, wherein each of the sets of second notifications has a corresponding volume and is sent to a plurality of users, and the volumes corresponding to the sets of second notifications are different;
determine visitation impacts of the sets of first and second notifications of the first and second notification types, respectively, wherein each of the visitation impacts is associated with one of the sets of first and second notifications and the volume corresponding to that set;
train, based on the visitation impacts and the associated volumes, a machine-learning model to output assessments of the first and second notification types;
determine, based on the assessments, a desired volume of a set of third notifications and a desired notification type for the set of third notifications, wherein the desired notification type is selected from the first notification type or the second notification type; and
send, through the communications network, the set of third notifications.

### Claim 2 (depends on 1)

The system of claim 1, wherein the processors are further operable when executing the instructions to:
receive a request to send a plurality of third notifications to a plurality of target users, wherein the set of third notifications is a subset of the plurality of third notifications; and
select a set of target users from the plurality of target users based on the desired volume of the set of third notifications;
wherein the set of third notifications are sent to the selected set of target users.

### Claim 3 (depends on 2)

The system of claim 2, wherein the processors are further operable when executing the instructions to:
derive, based on the machine-learning model, a characteristic value associated with a desired user characteristic for the set of third notifications, wherein the set of target users are selected based on the desired user characteristic.

### Claim 4 (depends on 1)

The system of claim 1, wherein the processors are further operable when executing the instructions to:
derive, based on the visitation impacts, a curve of visitation impact versus volume for the first notification type or the second notification type, wherein at least one of the desired volume or the desired notification type is determined based on the curve.

### Claim 5 (depends on 4)

The system of claim 4, wherein the desired volume is determined based on an application of a threshold visitation impact against the curve, wherein the threshold visitation impact is based on a cost of sending notifications of the first notification type or the second notification type.

### Claim 6 (depends on 1)

The system of claim 1, wherein the visitation impacts comprise a user rate of conversion responsive to a notification, a user rate of activity responsive to the notification, or a number of users that have accessed a social-networking system in a time period responsive to the notification.

### Claim 7 (depends on 1)

The system of claim 1, wherein each of the first notification type and the second notification type corresponds to Simple Message Service (SMS) notifications, Multimedia Messaging Service (MMS) notifications, e-mails, or push notifications.

### Claim 8 (depends on 1)

The system of claim 1, wherein the first notification type comprises a social-network post notification, a social-network tag notification, a social-network comment notification, or a social-network photo notification.

### Claim 9 (depends on 3)

The system of claim 3, wherein the desired user characteristic comprises a user demographic, a user usage pattern, a user device type, a social-network friend count, or a social-network age.

### Claim 10

One or more computer-readable non-transitory storage media embodying software that is operable when executed to:
send, through a communications network, sets of first notifications corresponding to a first notification type, wherein each of the sets of first notifications has a corresponding volume and is sent to a plurality of users, and the volumes corresponding to the sets of first notifications are different;
send, through the communications network, sets of second notifications corresponding to a second notification type, wherein each of the sets of second notifications has a corresponding volume and is sent to a plurality of users, and the volumes corresponding to the sets of second notifications are different;
determine visitation impacts of the sets of first and second notifications of the first and second notification types, respectively, wherein each of the visitation impacts is associated with one of the sets of first and second notifications and the volume corresponding to that set;
train, based on the visitation impacts and the associated volumes, a machine-learning model to output assessments of the first and second notification types;
determine, based on the assessments, a desired volume of a set of third notifications and a desired notification type for the set of third notifications, wherein the desired notification type is selected from the first notification type or the second notification type; and
send, through the communications network, the set of third notifications.

### Claim 11 (depends on 10)

The media of claim 10, wherein the software is further operable when executed to:
derive, based on the machine-learning model, a characteristic value associated with a desired user characteristic for the set of third notifications, wherein the set of target users are selected based on the desired user characteristic.

### Claim 12 (depends on 10)

The media of claim 10, wherein the software is further operable when executed to:
derive, based on the visitation impacts, a curve of visitation impact versus volume for the first notification type or the second notification type, wherein at least one of the desired volume or the desired notification type is determined based on the curve.

### Claim 13 (depends on 12)

The media of claim 12, wherein the desired volume is determined based on an application of a threshold visitation impact against the curve, wherein the threshold visitation impact is based on a cost of sending notifications of the first notification type or the second notification type.

### Claim 14 (depends on 10)

The media of claim 10, wherein the visitation impacts comprise a user rate of conversion responsive to a notification, a user rate of activity responsive to the notification, or a number of users that have accessed a social-networking system in a time period responsive to the notification.

### Claim 15 (depends on 10)

The media of claim 10, wherein each of the first notification type and the second notification type corresponds to Simple Message Service (SMS) notifications, Multimedia Messaging Service (MMS) notifications, e-mails, or push notifications.

### Claim 16

A computer-implemented method comprising, by one or more computing devices:
sending, through a communications network, sets of first notifications corresponding to a first notification type, wherein each of the sets of first notifications has a corresponding volume and is sent to a plurality of users, and the volumes corresponding to the sets of first notifications are different;
sending, through the communications network, sets of second notifications corresponding to a second notification type, wherein each of the sets of second notifications has a corresponding volume and is sent to a plurality of users, and the volumes corresponding to the sets of second notifications are different;
determining visitation impacts of the sets of first and second notifications of the first and second notification types, respectively, wherein each of the visitation impacts is associated with one of the sets of first and second notifications and the volume corresponding to that set;
training, based on the visitation impacts and the associated volumes, a machine-learning model to output assessments of the first and second notification types;
determining, based on the assessments, a desired volume of a set of third notifications and a desired notification type for the set of third notifications, wherein the desired notification type is selected from the first notification type or the second notification type; and
sending, through the communications network, the set of third notifications.

### Claim 17 (depends on 16)

The method of claim 16, further comprising:
receiving a request to send a plurality of third notifications to a plurality of target users, wherein the set of third notifications is a subset of the plurality of third notifications; and
selecting a set of target users from the plurality of target users based on the desired volume of the set of third notifications;
wherein the set of third notifications are sent to the selected set of target users.

### Claim 18 (depends on 17)

The method of claim 17, further comprising:
deriving, based on the machine-learning model, a characteristic value associated with a desired user characteristic for the set of third notifications, wherein the set of target users are selected based on the desired user characteristic.

### Claim 19 (depends on 16)

The method of claim 16, further comprising:
deriving, based on the visitation impacts, a curve of visitation versus volume for the first notification type or the second notification type, wherein at least one of the desired volume or the desired notification type is determined based on the curve.

### Claim 20 (depends on 19)

The method of claim 19, wherein the desired volume is determined based on an application of a threshold visitation impact against the curve, wherein the threshold visitation impact is based on a cost of sending notifications of the first notification type or the second notification type.