# US 2021/0064682 A1 · Transfer model learning for relevance models

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

- Number: US 2021/0064682 A1
- Document ID: US20210064682A1
- Status: Published application
- Kind: A1
- Application number: 16/557,263
- Filed: 2019-08-30
- Published: 2021-03-04
- Earliest claimed: 2019-08-30
- Assignees: Microsoft Technology Licensing, LLC
- Applicants: Microsoft Technology Licensing LLC
- Inventors: Manas Haribhai Somaiya; Mohit Rajkumar Kothari; Ian Robert Ackerman; Yuan Shao
- Web page: https://patentgrep.com/patent/US20210064682A1

## CPC

- G06F 16/9535: Search customisation based on user profiles and personalisation
- G06F 16/9536: Search customisation based on social or collaborative filtering
- G06F 16/9538: Presentation of query results
- G06F 16/9577: {Optimising the visualization of content, e.g. distillation of HTML documents}
- G06N 3/08: Learning methods
- G06N 3/09: Supervised learning
- G06N 3/096: Transfer learning
- G06N 5/01: Dynamic search techniques; Heuristics; Dynamic trees; Branch-and-bound
- G06N 20/00: Machine learning
- G06N 20/20: Ensemble learning
- G06N 5/022: Knowledge engineering; Knowledge acquisition
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS

## Abstract

Systems and techniques for a transfer model learning for relevance models are described herein. In an example, a system for member relevance prediction is adapted to collect a first data set of member interactions with the online service that occur on a first platform and train a first model using the first data set. The system for member relevance prediction may collect a second data set of member interactions with the online service that occur on a second platform. The system for member relevance prediction may predict a third data set related to member interactions using the first model and aggregate the first data set, the second data set, and the third data set. The system for member relevance prediction may train a second model for the second platform using the aggregated platform data and predict for the second platform, using the second model, online service items for the member.

## Claims

### Claim 1

A system for member relevance prediction with an online service, comprising:
at least one processor; and
memory including instructions that, when executed by the at least one processor, cause the at least one processor to:
collect a first set of data related to interactions, of a member, with the online service, wherein the interactions occur on a first platform;
train a first model using the first set of data;
collect a second set of data related to interactions, of the member, with the online service, wherein the interactions occur on a second platform;
predict a third set of data related to member interactions using the first model;
aggregate, as aggregated platform data, the first set of data, the second set of data, and the third set of data;
train a second model for the second platform using the aggregated platform data;
predict for the second platform, using the second model, a set of online service items for the member; and
present, on the second platform, the set of online service items.

### Claim 2 (depends on 1)

The system of claim 1, wherein the first platform is a first device and the second platform is a second device.

### Claim 3 (depends on 1)

The system of claim 1, wherein the first platform is a first operating system and the second platform is a second operating system.

### Claim 4 (depends on 1)

The system of claim 1, wherein the first platform is a first geo-location and the second platform is a second geo-location, distinct from the first geo-location.

### Claim 5 (depends on 1)

The system of claim 1, wherein the set of online service items includes: member updates, articles, internet links, videos, images, audio, and job opportunities.

### Claim 6 (depends on 1)

The system of claim 1, wherein the interactions include clicking an internet link; commenting, sharing, or providing a positive acknowledgment.

### Claim 7 (depends on 1)

The system of claim 1, wherein the first platform is a first application and the second platform is a second application, and wherein applications include messaging, online service feed, notifications, and online service promotions.

### Claim 8

At least one non-transitory computer readable medium including instructions for member relevance prediction with an online service that when executed by at least one processor, cause the at least one processor to:
collect a first set of data related to interactions, of a member, with the online service, wherein the interactions occur on a first platform;
train a first model using the first set of data;
collect a second set of data related to interactions, of the member, with the online service, wherein the interactions occur on a second platform;
predict a third set of data related to member interactions using the first model;
aggregate, as aggregated platform data, the first set of data, the second set of data, and the third set of data;
train a second model for the second platform using the aggregated platform data;
predict for the second platform, using the second model, a set of online service items for the member; and
present, on the second platform, the set of online service items.

### Claim 9 (depends on 8)

The at least one computer readable medium of claim 8, wherein the first platform is a first device and the second platform is a second device.

### Claim 10 (depends on 8)

The at least one computer readable medium of claim 8, wherein the first platform is a first operating system and the second platform is a second operating system.

### Claim 11 (depends on 8)

The at least one computer readable medium of claim 8, wherein the first platform is a first geo-location and the second platform is a second geo-location, distinct from the first geo-location.

### Claim 12 (depends on 8)

The at least one computer readable medium of claim 8, wherein the set of online service items includes: member updates, articles, internet links, videos, images, audio, and job opportunities.

### Claim 13 (depends on 8)

The at least one computer readable medium of claim 8, wherein the interactions include clicking an internet link, commenting, sharing, or providing a positive acknowledgment.

### Claim 14 (depends on 8)

The at least one computer readable medium of claim 8, wherein the first platform is a first application and the second platform is a second application, and wherein applications include messaging, online service feed, notifications, and online service promotions.

### Claim 15

A method for member relevance prediction with an online service, the method comprising:
collecting a first set of data related to interactions, of a member, with the online service, wherein the interactions occur on a first platform;
training a first model using the first set of data;
collecting a second set of data related to interactions, of the member, with the online service, wherein the interactions occur on a second platform;
predicting a third set of data related to member interactions using the first model;
aggregating, as aggregated platform data, the first set of data, the second set of data, and the third set of data;
training a second model for the second platform using the aggregated platform data;
predicting for the second platform, using the second model, a set of online service items for the member; and
presenting, on the second platform, the set of online service items.

### Claim 16 (depends on 15)

The method of claim 15, wherein the first platform is a first device and the second platform is a second device.

### Claim 17 (depends on 15)

The method of claim 15, wherein the first platform is a first operating system and the second platform is a second operating system.

### Claim 18 (depends on 15)

The method of claim 15, wherein the first platform is a first geo-location and the second platform is a second geo-location, distinct from the first geo-location.

### Claim 19 (depends on 15)

The method of claim 15, wherein the set of online service items includes: member updates, articles, internet links, videos, images, audio, and job opportunities.

### Claim 20 (depends on 15)

The method of claim 15, wherein the interactions include clicking an internet link, commenting, sharing, or providing a positive acknowledgment.
