# US 2021/0240742 A1 · Systems and methods for keyword categorization

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

- Number: US 2021/0240742 A1
- Document ID: US20210240742A1
- Status: Published application
- Kind: A1
- Application number: 16/777,085
- Filed: 2020-01-30
- Published: 2021-08-05
- Earliest claimed: 2020-01-30
- Assignees: Walmart Apollo, LLC
- Applicants: Walmart Apollo, LLC
- Inventors: Zigeng Wang; Cheng Jie; Wei Shen
- Web page: https://patentgrep.com/patent/US20210240742A1

## CPC

- G06F 16/358: Browsing; Visualisation therefor
- G06F 16/355: Creation or modification of classes or clusters
- G06F 16/953: Querying, e.g. by the use of web search engines
- G06N 20/00: Machine learning
- G06Q 30/0283: Price estimation or determination
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- 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

## Abstract

Systems and methods including one or more processors and one or more non-transitory storage devices storing computing instructions configured to run on the one or more processors and perform: receiving a set of keywords from a graphical user interface of an electronic device of a user; pre-processing at least one keyword of the set of keywords; receiving a hierarchical categorization; pre-processing at least one category of the hierarchical categorization; determining a respective similarity between each of the at least one keyword of the set of keywords and each of the at least one category of the hierarchical categorization; determining a respective confidence level of a most likely category in the hierarchical categorization for each of the at least one keyword of the set of keywords using the respective similarity between each of the at least one keyword of the set of keywords and each of the at least one category of the hierarchical categorization; ranking each of the at least one keyword of the set of keywords using a respective importance of each keyword of the set of keywords and the respective confidence level of the most likely category in the hierarchical categorization; and altering a respective arrangement of at least one of the at least one keyword on the graphical user interface on the electronic device of the user based on the ranking. Other embodiments are disclosed herein.

## Claims

### Claim 1

A system comprising:
one or more processors; and
one or more non-transitory computer-readable storage devices storing computing instructions configured to run on the one or more processors and perform:
receiving a set of keywords from a graphical user interface of an electronic device of a user;
pre-processing at least one keyword of the set of keywords;
receiving a hierarchical categorization;
pre-processing at least one category of the hierarchical categorization;
determining a respective similarity between each of the at least one keyword of the set of keywords and each of the at least one category of the hierarchical categorization;
determining a respective confidence level of a most likely category in the hierarchical categorization for each of the at least one keyword of the set of keywords using the respective similarity between each of the at least one keyword of the set of keywords and each of the at least one category of the hierarchical categorization;
ranking each of the at least one keyword of the set of keywords using a respective importance of each keyword of the set of keywords and the respective confidence level of the most likely category in the hierarchical categorization; and
altering a respective arrangement of at least one of the at least one keyword on the graphical user interface on the electronic device of the user based on the ranking.

### Claim 2 (depends on 1)

The system of claim 1, wherein pre-processing the at least one keyword of the set of keywords comprises at least one of:
removing a stop word from the at least one keyword;
removing a retailer name from the at least one keyword;
removing a variation of the retailer name from the at least one keyword; or
removing an unnecessary symbol from the at least one keyword.

### Claim 3 (depends on 1)

The system of claim 1, wherein pre-processing the at least one category of the hierarchical categorization comprises at least one of:
renaming a partial directory of the at least one category;
creating a subcategory within a terminal category of the at least one category; or
inserting a category between two other categories in the hierarchical categorization.

### Claim 4 (depends on 1)

The system of claim 1, wherein determining the respective similarity between each of the at least one keyword of the set of keywords and each of the at least one category of the hierarchical categorization comprises:
using a sentence embedding algorithm to generate:
a respective first vector representation of each of the at least one keyword of the set of keywords; and
a respective second vector representation of each of the at least one category of the hierarchical categorization.

### Claim 5 (depends on 4)

The system of claim 4, wherein determining the respective similarity between each of the at least one keyword of the set of keywords and each of the at least one category of the hierarchical categorization further comprises:
determining a respective pairwise cosine similarity between the respective first vector representation of each of the at least one keyword of the set of keywords and the respective second vector representation of each of the at least one category of the hierarchical categorization; and
when a particular one of the respective pairwise cosine similarity between a particular one of the respective first vector representation of a particular one of the at least one keyword of the set of keywords and a particular one of the respective second vector representation of a particular one of the at least one category of the hierarchical categorization is a highest one of the respective pairwise cosine similarity, determining that the particular one of the at least one category of the hierarchical categorization is the most likely category for the particular one of the at least one keyword of the set of keywords.

### Claim 6 (depends on 1)

The system of claim 1, wherein the respective importance of each keyword of the set of keywords comprises at least one of:
revenue per click within a predefined time period for the keyword;
cost per click f within a predefined time period or the keyword; or
total number of clicks within a predefined time period for the keyword.

### Claim 7 (depends on 1)

The system of claim 1, wherein ranking each of the at least one keyword of the set of keywords comprises:
grouping each of the at least one keyword of the set of keywords with semantically similar keywords to create one or more clusters of keywords; and
ranking the one or more clusters of keywords.

### Claim 8 (depends on 7)

The system of claim 7, wherein grouping each of the at least one keyword of the set of keywords comprises:
using a distance based agglomerative clustering algorithm.

### Claim 9 (depends on 1)

The system of claim 1, wherein determining the respective confidence level of the most likely category comprises:
determining a respective pairwise cosine similarity between a respective first vector representation of each of the at least one keyword of the set of keywords and a respective second vector representation of each of the at least one category of the hierarchical categorization; and
determining a divergence of (1) the most likely category in the hierarchical categorization for each of the at least one keyword of the set of keywords and (2) a second most likely category in the hierarchical categorization for each of the at least one keyword of the set of keywords.

### Claim 10 (depends on 1)

The system of claim 1, wherein the one or more non-transitory computer-readable storage devices storing the computing instructions are further configured to run on the one or more processors and perform:
determining that the respective confidence level of one or more keywords of the set of keywords is low and that the respective importance of the one or more keywords of the set of keywords is high;
generating one or more artificial queries using the one or more keywords of the set of keywords;
facilitating displaying the one or more artificial queries on the graphical user interface of the electronic device of the user;
receiving, from the graphical user interface of the electronic device of the user, an interaction with the one or more artificial queries; and
altering the hierarchical categorization based on the interaction with the one or more artificial queries, as received.

### Claim 11

A method implemented via execution of computing instructions configured to run at one or more processors and configured to be stored at non-transitory computer-readable media, the method comprising:
receiving a set of keywords from a graphical user interface of an electronic device of a user;
pre-processing at least one keyword of the set of keywords;
receiving a hierarchical categorization;
pre-processing at least one category of the hierarchical categorization;
determining a respective similarity between each of the at least one keyword of the set of keywords and each of the at least one category of the hierarchical categorization;
determining a respective confidence level of a most likely category in the hierarchical categorization for each of the at least one keyword of the set of keywords using the respective similarity between each of the at least one keyword of the set of keywords and each of the at least one category of the hierarchical categorization;
ranking each of the at least one keyword of the set of keywords using a respective importance of each keyword of the set of keywords and the respective confidence level of the most likely category in the hierarchical categorization; and
altering a respective arrangement of at least one of the at least one keyword on the graphical user interface on the electronic device of the user based on the ranking.

### Claim 12 (depends on 11)

The method of claim 11, wherein pre-processing the at least one keyword of the set of keywords comprises at least one of:
removing a stop word from the at least one keyword;
removing a retailer name from the at least one keyword;
removing a variation of the retailer name from the at least one keyword; or
removing an unnecessary symbol from the at least one keyword.

### Claim 13 (depends on 11)

The method of claim 11, wherein pre-processing the at least one category of the hierarchical categorization comprises at least one of:
renaming a partial directory of the at least one category;
creating a subcategory within a terminal category of the at least one category; or
inserting a category between two other categories in the hierarchical categorization.

### Claim 14 (depends on 11)

The method of claim 11, wherein determining the respective similarity between each of the at least one keyword of the set of keywords and each of the at least one category of the hierarchical categorization comprises:
using a sentence embedding algorithm to generate:
a respective first vector representation of each of the at least one keyword of the set of keywords; and
a respective second vector representation of each of the at least one category of the hierarchical categorization.

### Claim 15 (depends on 14)

The method of claim 14, wherein determining the respective similarity between each of the at least one keyword of the set of keywords and each of the at least one category of the hierarchical categorization further comprises:
determining a respective pairwise cosine similarity between the respective first vector representation of each of the at least one keyword of the set of keywords and the respective second vector representation of each of the at least one category of the hierarchical categorization; and
when a particular one of the respective pairwise cosine similarity between a particular one of the respective first vector representation of a particular one of the at least one keyword of the set of keywords and a particular one of the respective second vector representation of a particular one of the at least one category of the hierarchical categorization is a highest one of the respective pairwise cosine similarity, determining that the particular one of the at least one category of the hierarchical categorization is the most likely category for the particular one of the at least one keyword of the set of keywords.

### Claim 16 (depends on 11)

The method of claim 11, wherein the respective importance of each keyword of the set of keywords comprises at least one of:
revenue per click within a predefined time period for the keyword;
cost per click f within a predefined time period or the keyword; or
total number of clicks within a predefined time period for the keyword.

### Claim 17 (depends on 11)

The method of claim 11, wherein ranking each of the at least one keyword of the set of keywords comprises:
grouping each of the at least one keyword of the set of keywords with semantically similar keywords to create one or more clusters of keywords; and
ranking the one or more clusters of keywords.

### Claim 18 (depends on 17)

The method of claim 17, wherein grouping each of the at least one keyword of the set of keywords comprises:
using a distance based agglomerative clustering algorithm.

### Claim 19 (depends on 11)

The method of claim 11, wherein determining the respective confidence level of the most likely category comprises:
determining a respective pairwise cosine similarity between a respective first vector representation of each of the at least one keyword of the set of keywords and a respective second vector representation of each of the at least one category of the hierarchical categorization; and
determining a divergence of (1) the most likely category in the hierarchical categorization for each of the at least one keyword of the set of keywords and (2) a second most likely category in the hierarchical categorization for each of the at least one keyword of the set of keywords.

### Claim 20 (depends on 11)

The method of claim 11 further comprising:
determining that the respective confidence level of one or more keywords of the set of keywords is low and that the respective importance of the one or more keywords of the set of keywords is high;
generating one or more artificial queries using the one or more keywords of the set of keywords;
facilitating displaying the one or more artificial queries on the graphical user interface of the electronic device of the user;
receiving, from the graphical user interface of the electronic device of the user, an interaction with the one or more artificial queries; and
altering the hierarchical categorization based on the interaction with the one or more artificial queries, as received.