# US 2021/0241314 A1 · Method and system for optimizing user grouping for advertisement

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

- Number: US 2021/0241314 A1
- Document ID: US20210241314A1
- Status: Published application
- Kind: A1
- Application number: 17/218,293
- Filed: 2021-03-31
- Published: 2021-08-05
- Earliest claimed: 2020-04-13
- Assignees: Alipay (Hangzhou) Information Technology Co., Ltd.
- Applicants: Alipay (Hangzhou) Information Technology Co., Ltd.
- Inventors: Jia YAN; Zhigang HUA; Feng QI; Yingqi LIU; Yingping CAO
- Web page: https://patentgrep.com/patent/US20210241314A1

## CPC

- G06Q 30/0251: Targeted advertisements
- G06Q 30/0277: {Online advertisement}
- G06N 20/00: Machine learning
- G06Q 30/0254: {based on statistics}
- 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
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS

## Abstract

Methods, systems, and apparatus, including computer programs encoded on computer storage media, for grouping users are provided. One embodiment of the methods includes: dividing a plurality of users targeted by the an advertisement candidate into a plurality of user buckets, wherein each of the plurality of user buckets is associated with a first conversion score; obtaining a trained prediction model corresponding to the advertisement, wherein the trained prediction model is able to predict a conversion score based at least on the first conversion score associated with a user bucket and a second conversion score associated with a group of user buckets comprising the user bucket; and constructing an optimization model using the trained prediction model, wherein an objective function of the optimization problem is to maximize a total conversion score with a grouping strategy determined by solving the optimization problem.

## Claims

### Claim 1

A computer-implemented method for grouping users, comprising:
dividing a plurality of users targeted by the an advertisement candidate into a plurality of user buckets, wherein each of the plurality of user buckets is associated with a first conversion score;
obtaining a trained prediction model corresponding to the advertisement, wherein the trained prediction model is able to predict a conversion score based at least on the first conversion score associated with a user bucket and a second conversion score associated with a group of user buckets comprising the user bucket;
constructing an optimization model using the trained prediction model, wherein an objective function of the optimization problem is to maximize a total conversion score with a grouping strategy represented by a plurality of decision variables;
solving the optimization problem to determine values of the plurality of decision variables; and
assigning the plurality of user buckets to one or more user groups for the advertisement candidate according to the grouping strategy.

### Claim 2 (depends on 1)

The method of claim 1, further comprising:
identifying a plurality of the advertisement candidates targeting a user;
for each of the plurality of the advertisement candidates, determining a score for delivering the each advertisement candidate to the user, wherein the score is determined based at least on the grouping strategy;
identifying one or more of the plurality of advertisement candidates based at least on one or more corresponding scores; and
automatically delivering the one or more advertisement candidates to a computing device associated with the user to cause the one or more advertisement candidates to be displayed on the computing device.

### Claim 3 (depends on 2)

The method of claim 2, wherein the determining a score for delivering the each advertisement candidate comprises:
determining the score based at least on the first conversion score associated with one of the plurality of user buckets comprising the user, and the second conversion score associated with one of the one or more user groups comprising the one user bucket.

### Claim 4 (depends on 2)

The method of claim 2, wherein the identifying one or more of the plurality of advertisements comprises:
identifying the one or more advertisement candidates with the highest scores.

### Claim 5 (depends on 1)

The method of claim 1, wherein:
each of the plurality of users is associated with a third conversion score corresponding to the advertisement;
each of the plurality of user buckets comprises one or more of the plurality of users; and
the first conversion score associated with the each of the plurality of user buckets comprises an average of one or more third conversion scores associated with the one or more of the plurality of users in the each of the plurality of user buckets.

### Claim 6 (depends on 5)

The method of claim 5, wherein:
the third conversion score is a click through rate in response to the advertisement.

### Claim 7 (depends on 1)

The method of claim 1, wherein:
the second conversion score associated with a group of one or more of the plurality of user buckets is determined based at least on the first conversion scores associated with the one or more of the plurality of user buckets in the group and a number of users in each of the one or more of the plurality of user buckets.

### Claim 8 (depends on 1)

The method of claim 1, wherein:
the prediction model is a tensor flow model that fits historical data of exposure rate to estimate a plurality of parameters of a logistic function.

### Claim 9 (depends on 1)

The method of claim 1, wherein:
the values of the plurality of decision variables are binary values; and
each of the plurality of decision variables represents whether one of the plurality of user buckets is assigned to one of the one or more user groups.

### Claim 10 (depends on 1)

The method of claim 1, wherein:
a quantity of the one or more user groups is a configurable value associated with the advertisement.

### Claim 11 (depends on 1)

The method of claim 1, wherein:
the optimization model comprises a constraint that one user bucket is assigned to one and only one user group.

### Claim 12 (depends on 1)

The method of claim 1, wherein:
the objective function of the optimization problem is max Σj=1J Σi=1I f(si, tj)*ni*xij;
where J is the quantity of the one or more user groups;
where j corresponds to jth user group;
where I is a number of the plurality of user buckets;
where i corresponds to ith user bucket;
where f(si, tj) corresponds to the trained prediction model;
where si corresponds to the first conversion score of the ith user bucket;
where tj corresponds to the second conversion score of the jth user group;
where ni corresponds to a number of users in the ith user bucket; and
where xij corresponds to a decision variable indicating whether the ith user bucket is assigned to the jth user group.

### Claim 13

A system for grouping users, comprising one or more processors and one or more non-transitory computer-readable memories coupled to the one or more processors and configured with instructions executable by the one or more processors to cause the system to perform operations comprising:
dividing a plurality of users targeted by the an advertisement candidate into a plurality of user buckets, wherein each of the plurality of user buckets is associated with a first conversion score;
obtaining a trained prediction model corresponding to the advertisement, wherein the trained prediction model is able to predict a conversion score based at least on the first conversion score associated with a user bucket and a second conversion score associated with a group of user buckets comprising the user bucket;
constructing an optimization model using the trained prediction model, wherein an objective function of the optimization problem is to maximize a total conversion score with a grouping strategy represented by a plurality of decision variables;
solving the optimization problem to determine values of the plurality of decision variables; and
assigning the plurality of user buckets to one or more user groups for the advertisement candidate according to the grouping strategy.

### Claim 14 (depends on 13)

The system of claim 13, wherein the operations further comprise:
identifying a plurality of the advertisement candidates targeting a user;
for each of the plurality of the advertisement candidates, determining a score for delivering the each advertisement candidate to the user, wherein the score is determined based at least on the grouping strategy;
identifying one or more of the plurality of advertisement candidates based at least on one or more corresponding scores; and
automatically delivering the one or more advertisement candidates to a computing device associated with the user to cause the one or more advertisement candidates to be displayed on the computing device.

### Claim 15 (depends on 13)

The system of claim 13, wherein the determining a score for delivering the each advertisement candidate comprises:
determining the score based at least on the first conversion score associated with one of the plurality of user buckets comprising the user, and the second conversion score associated with one of the one or more user groups comprising the one user bucket.

### Claim 16 (depends on 13)

The system of claim 13, wherein:
each of the plurality of users is associated with a third conversion score corresponding to the advertisement;
each of the plurality of user buckets comprises one or more of the plurality of users; and
the first conversion score associated with the each of the plurality of user buckets comprises an average of one or more third conversion scores associated with the one or more of the plurality of users in the each of the plurality of user buckets.

### Claim 17

A non-transitory computer-readable storage medium configured with instructions executable by one or more processors to cause the one or more processors to perform operations comprising:
dividing a plurality of users targeted by the an advertisement candidate into a plurality of user buckets, wherein each of the plurality of user buckets is associated with a first conversion score;
obtaining a trained prediction model corresponding to the advertisement, wherein the trained prediction model is able to predict a conversion score based at least on the first conversion score associated with a user bucket and a second conversion score associated with a group of user buckets comprising the user bucket;
constructing an optimization model using the trained prediction model, wherein an objective function of the optimization problem is to maximize a total conversion score with a grouping strategy represented by a plurality of decision variables;
solving the optimization problem to determine values of the plurality of decision variables; and
assigning the plurality of user buckets to one or more user groups for the advertisement candidate according to the grouping strategy.

### Claim 18 (depends on 17)

The non-transitory computer-readable storage medium of claim 17, the operations further comprising:
identifying a plurality of the advertisement candidates targeting a user;
for each of the plurality of the advertisement candidates, determining a score for delivering the each advertisement candidate to the user, wherein the score is determined based at least on the grouping strategy;
identifying one or more of the plurality of advertisement candidates based at least on one or more corresponding scores; and
automatically delivering the one or more advertisement candidates to a computing device associated with the user to cause the one or more advertisement candidates to be displayed on the computing device.

### Claim 19 (depends on 17)

The non-transitory computer-readable storage medium of claim 17, wherein the determining a score for delivering the each advertisement candidate comprises:
determining the score based at least on the first conversion score associated with one of the plurality of user buckets comprising the user, and the second conversion score associated with one of the one or more user groups comprising the one user bucket.

### Claim 20 (depends on 17)

The non-transitory computer-readable storage medium of claim 17, wherein:
each of the plurality of users is associated with a third conversion score corresponding to the advertisement;
each of the plurality of user buckets comprises one or more of the plurality of users; and
the first conversion score associated with the each of the plurality of user buckets comprises an average of one or more third conversion scores associated with the one or more of the plurality of users in the each of the plurality of user buckets.
