# US 2020/0394592 A1 · Generating a machine-learned model for scoring skills based on feedback from job posters

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

- Number: US 2020/0394592 A1
- Document ID: US20200394592A1
- Status: Published application
- Kind: A1
- Application number: 16/443,608
- Filed: 2019-06-17
- Published: 2020-12-17
- Earliest claimed: 2019-06-17
- Assignees: Microsoft Technology Licensing, LLC
- Applicants: Microsoft Technology Licensing, LLC
- Inventors: Baoxu Shi; Feng Guo; Jaewon Yang; Qi He
- Web page: https://patentgrep.com/patent/US20200394592A1

## CPC

- G06Q 10/063112: {Skill-based matching of a person or a group to a task}
- G06N 20/00: Machine learning
- G06Q 10/06393: {Score-carding, benchmarking or key performance indicator [KPI] analysis}
- G06Q 10/06398: {Performance of employee with respect to a job function}
- G06Q 10/1053: Employment or hiring
- 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

Techniques for scoring data items using a machine-learned model are provided. In one technique, multiple skills are identifying based on a job posting. Multiple attribute values of the job posting are identified. For each identified skill, multiple probabilities are identified, each probability corresponding to a different attribute value of the identified attribute values. The probabilities are input into a machine-learned model to generate multiple scores. Multiple skills of a candidate user are identified. An affinity score between the job posting and the candidate user is generated based the scores and the skills of the candidate user.

## Claims

### Claim 1

A method comprising:
identifying, based on a job posting, a first plurality of skills;
identifying a plurality of attribute values of the job posting;
for each skill of the first plurality of skills, identifying a plurality of probabilities, each probability corresponding to a different attribute value of the plurality of attribute values;
inputting the plurality of probabilities into a machine-learned model to generate a plurality of scores;
identifying a second plurality of skills of a candidate user;
generating an affinity score between the job posting and the candidate user based the plurality of scores and the second plurality of skills;
wherein the method is performed by one or more computing devices.

### Claim 2 (depends on 1)

The method of claim 1, further comprising:
receiving, from a job poster, input that comprises a plurality of data items for a second job posting;
identifying, based on the input, a third plurality of skills;
identifying a second plurality of attribute values of the second job posting;
for each skill of the third plurality of skills, identifying a second plurality of probabilities, each probability corresponding to a different attribute value of the second plurality of attribute values;
inputting the second plurality of probabilities into the machine-learned model to generate a second plurality of scores;
based on the second plurality of scores, selecting a subset of the third plurality of skills;
causing the subset of the third plurality of skills to be presented on a screen of a computing device of the job poster;
receiving, from the job poster, feedback that indicates selection or rejection of one or more skills in the subset of the third plurality of skills;
generating one or more training instances based on the feedback;
using one or more machine learning techniques to train a second machine-learned model based on the one or more training instances.

### Claim 3 (depends on 1)

The method of claim 1, wherein:
the feedback also indicates a particular skill that is not included in the subset of the third plurality of skills;
the one or more training instances includes a training instance that indicates the particular skill and includes a label that indicates that the particular skill was selected.

### Claim 4 (depends on 1)

The method of claim 1, wherein:
the feedback also indicates a rejection of a particular skill that is included in the subset of the third plurality of skills;
the one or more training instances includes a training instance that indicates the particular skill and includes a label that indicates that the particular skill was rejected.

### Claim 5 (depends on 1)

The method of claim 1, wherein the plurality of attribute values correspond, respectively, to a plurality of attributes that comprise one or more of job title, job function, or industry.

### Claim 6 (depends on 1)

The method of claim 1, wherein the plurality of probabilities include:
a first probability that said each skill is found in first job postings that include a first attribute value of the plurality of attribute values;
a second probability that said each skill is found in second job postings that include a second attribute value of the plurality of attribute values.

### Claim 7 (depends on 1)

The method of claim 1, wherein the plurality of probabilities include:
a first probability that said each skill is found in first user profiles that include a first attribute value of the plurality of attribute values;
a second probability that said each skill is found in second user profiles that include a second attribute value of the plurality of attribute values.

### Claim 8 (depends on 1)

The method of claim 1, wherein the plurality of probabilities include:
a first probability that said each skill was selected by job posting authors that is associated with a first attribute value of the plurality of attribute values;
a second probability that said each skill was selected by job posting authors that are associated with a second attribute value of the plurality of attribute values.

### Claim 9 (depends on 1)

The method of claim 1, wherein the plurality of probabilities include a first probability that said each skill is found in job postings that include a first attribute value of the plurality of attribute values and a second attribute value of the plurality of attribute values.

### Claim 10

A method comprising:
for each job posting of a first plurality of job postings:
using a first model to select a first plurality of suggested skills for said each job posting;
presenting the first plurality of suggested skills to a user that posted said each job posting;
receiving, from the user, input that indicates first selection of one or more first suggested skills in the first plurality of suggested skills;
creating one or more training instances based on the first selection;
adding the one or more training instances to a set of training instances;
using one or more machine learning techniques train a second model based on the set of training instances;
for each job posting of a second plurality of job postings:
using the second model to select a second plurality of suggested skills for said each job posting;
presenting the second plurality of suggested skills to a second user that posted said each job posting;
wherein the method is performed by one or more computing devices.

### Claim 11 (depends on 10)

The method of claim 10, wherein the second model is based on a plurality of features that comprise one or more of job title, job function, or industry.

### Claim 12

One or more storage media storing instructions which, when executed by one or more processors, cause:
identifying, based on a job posting, a first plurality of skills;
identifying a plurality of attribute values of the job posting;
for each skill of the first plurality of skills, identifying a plurality of probabilities, each probability corresponding to a different attribute value of the plurality of attribute values;
inputting the plurality of probabilities into a machine-learned model to generate a plurality of scores;
identifying a second plurality of skills of a candidate user;
generating an affinity score between the job posting and the candidate user based the plurality of scores and the second plurality of skills.

### Claim 13 (depends on 12)

The one or more storage media of claim 12, wherein the instructions, when executed by the one or more processors, further cause:
receiving, from a job poster, input that comprises a plurality of data items for a second job posting;
identifying, based on the input, a third plurality of skills;
identifying a second plurality of attribute values of the second job posting;
for each skill of the third plurality of skills, identifying a second plurality of probabilities, each probability corresponding to a different attribute value of the second plurality of attribute values;
inputting the second plurality of probabilities into the machine-learned model to generate a second plurality of scores;
based on the second plurality of scores, selecting a subset of the third plurality of skills;
causing the subset of the third plurality of skills to be presented on a screen of a computing device of the job poster;
receiving, from the job poster, feedback that indicates selection or rejection of one or more skills in the subset of the third plurality of skills;
generating one or more training instances based on the feedback;
using one or more machine learning techniques to train a second machine-learned model based on the one or more training instances.

### Claim 14 (depends on 12)

The one or more storage media of claim 12, wherein:
the feedback also indicates a particular skill that is not included in the subset of the third plurality of skills;
the one or more training instances includes a training instance that indicates the particular skill and includes a label that indicates that the particular skill was selected.

### Claim 15 (depends on 12)

The one or more storage media of claim 12, wherein:
the feedback also indicates a rejection of a particular skill that is included in the subset of the third plurality of skills;
the one or more training instances includes a training instance that indicates the particular skill and includes a label that indicates that the particular skill was rejected.

### Claim 16 (depends on 12)

The one or more storage media of claim 12, wherein the plurality of attribute values correspond, respectively, to a plurality of attributes that comprise one or more of job title, job function, or industry.

### Claim 17 (depends on 12)

The one or more storage media of claim 12, wherein the plurality of probabilities include:
a first probability that said each skill is found in first job postings that include a first attribute value of the plurality of attribute values;
a second probability that said each skill is found in second job postings that include a second attribute value of the plurality of attribute values.

### Claim 18 (depends on 12)

The one or more storage media of claim 12, wherein the plurality of probabilities include:
a first probability that said each skill is found in first user profiles that include a first attribute value of the plurality of attribute values;
a second probability that said each skill is found in second user profiles that include a second attribute value of the plurality of attribute values.

### Claim 19 (depends on 12)

The one or more storage media of claim 12, wherein the plurality of probabilities include:
a first probability that said each skill was selected by job posting authors that is associated with a first attribute value of the plurality of attribute values;
a second probability that said each skill was selected by job posting authors that are associated with a second attribute value of the plurality of attribute values.

### Claim 20 (depends on 12)

The one or more storage media of claim 12, wherein the plurality of probabilities include a first probability that said each skill is found in job postings that include a first attribute value of the plurality of attribute values and a second attribute value of the plurality of attribute values.
