# US 2021/0240700 A1 · Systems and methods for a title quality scoring framework

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

- Number: US 2021/0240700 A1
- Document ID: US20210240700A1
- Status: Published application
- Kind: A1
- Application number: 16/777,767
- Filed: 2020-01-30
- Published: 2021-08-05
- Earliest claimed: 2020-01-30
- Assignees: Walmart Apollo, LLC
- Applicants: Walmart Apollo, LLC
- Inventors: Bo Ling; Benjamin Matthew Breuner; Abon Chaudhuri; Krishna Chaitanya Yalamanchili; Abilash Amarthaluri
- Web page: https://patentgrep.com/patent/US20210240700A1

## CPC

- G06Q 30/0641: {utilising user interfaces specially adapted for shopping}
- G06F 16/243: {Natural language query formulation}
- G06F 16/24564: {Applying rules; Deductive queries}
- G06F 16/24578: {using ranking}
- G06F 16/248: Presentation of query results
- G06F 16/30: of unstructured textual data (document management systems G06F16/93)
- G06F 16/3344: {using natural language analysis}
- G06F 18/22: Matching criteria, e.g. proximity measures
- G06F 18/295: {Markov models or related models, e.g. semi-Markov models; Markov random fields; Networks embedding Markov models}
- G06F 40/226: Validation
- G06F 40/30: Semantic analysis
- G06N 3/0499: Feedforward networks
- G06N 3/08: Learning methods
- G06N 20/10: using kernel methods, e.g. support vector machines [SVM]
- G06Q 30/0625: {by formulating product or service queries, e.g. using keywords or predefined options}
- G06V 10/761: {Proximity, similarity or dissimilarity measures}
- G06V 10/85: {Markov-related models; Markov random fields}
- G06N 3/045: Combinations of networks
- 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
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING

## Abstract

A system including one or more processors and one or more non-transitory computer-readable media storing computing instructions configured to run on the one or more processors and perform receiving a title of an item associated with an online catalog; interpreting, using a natural language model, one or more attributes of the predetermined set of attributes; determining a first title quality score for the title based on a first rule; determining a second title quality score for the title based on a second rule; determining an aggregated title quality score for the title based on at least the first title quality score and the second title quality score; generating a content quality list for the title; and sending instructions to display, on a user interface of an electronic device, a content quality dashboard comprising the content quality list for the title of the item. Other embodiments are disclosed.

## Claims

### Claim 1

A system comprising:
one or more processors; and
one or more non-transitory computer-readable media storing computing instructions configured to run on the one or more processors and perform:
receiving a title of an item associated with an online catalog, wherein:
the item is associated with a product type; and
the product type comprises a predetermined set of attributes;
interpreting, using a natural language model, one or more attributes of the predetermined set of attributes within the title of the item;
determining a first title quality score for the title based on a first rule used to evaluate the one or more attributes of the title;
determining a second title quality score for the title based on a second rule used to evaluate the one or more attributes of the title;
determining an aggregated title quality score for the title based on at least the first title quality score and the second title quality score;
generating a content quality list for the title, wherein the content quality list identifies at least one or more content issues based on the aggregated title quality score for the title; and
sending instructions to display, on a user interface of an electronic device, a content quality dashboard comprising the content quality list for the title of the item.

### Claim 2 (depends on 1)

The system of claim 1, wherein:
the computing instructions are further configured to run on the one or more processors and perform:
pre-processing data for each respective attribute of the predetermined set of attributes of each respective product type in the online catalog; and
extracting respective key values for the each respective product type, wherein each of the respective key values comprise a respective product type, a respective attribute name, and a respective attribute value; and
interpreting, using a natural language model, the one or more attributes of the predetermined set of attributes within the title of the item comprises:
processing each respective title into one or more n-grams; and
searching a nested dictionary using each combination of the one or more n-grams as a search query to locate the one or more attributes in the title of the item.

### Claim 3 (depends on 2)

The system of claim 2, wherein the computing instructions are further configured to run on the one or more processors and perform:
building the nested dictionary based on the respective key values, as extracted, for the each respective product type, wherein the nested dictionary comprises attributes of each product type, by;
lemmatize each respective word in each of the respective attribute values of the key values; and
storing each respective word of the respective attribute values of the key values, as lemmatized, in the nested dictionary database.

### Claim 4 (depends on 1)

The system of claim 1, wherein determining the first title quality score for the title further comprises:
generating an aggregated attribute score for the item based on at least (i) a quantity of the one or more attributes in the title and (ii) an assigned weight of each of the one or more attributes used in the title; and
assign the first title quality score for the item based on the aggregated attribute score.

### Claim 5 (depends on 1)

The system of claim 1, wherein determining the second title quality score for the title further comprises:
embedding, using a machine learning model, as trained, a product type of the item into a first vector;
embedding, using the machine learning model, as trained, each respective word in the title into one or more second vectors;
determining a similarity score between the first vector and each respective second vector of the one or more second vectors of each respective word in the title; and
assigning the second title quality score for the item based on a maximum of the similarity score.

### Claim 6 (depends on 5)

The system of claim 5, wherein the machine learning model further comprises a word2vec neural network.

### Claim 7 (depends on 5)

The system of claim 5, wherein determining the similarity score further comprises using a cosine similarity measurement.

### Claim 8 (depends on 1)

The system of claim 1, wherein:
the computing instructions are further configured to run on the one or more processors and perform:
determining a third title quality score for the title based on a third rule used to evaluate a length of the title based on the product type; and
determining a fourth title quality score for the title based on a fourth rule used to evaluate an occurrence of prohibited characters within the title based on a predetermined list of prohibited characters; and
determining the aggregated title quality score for the title further comprises:
determining the aggregated title quality score for the title based on additional factors of the third title quality score and the fourth title quality score.

### Claim 9 (depends on 1)

The system of claim 1, wherein:
the computing instructions are further configured to run on the one or more processors and perform:
determining a promotional title quality score for the title based on a promotional rule used to evaluate an occurrence of one or more promotional words within the title; and
determining the aggregated title quality score for the title further comprises:
determining the aggregated title quality score for the title based on an additional factor of the promotional title quality score.

### Claim 10 (depends on 1)

The system of claim 1, wherein:
the computing instructions are further configured to run on the one or more processors and perform:
determining a title capitalization quality score for the title based on a capitalization rule used to evaluate capitalization of each word in the title;
determining the aggregated title quality score for the title further comprises:
determining the aggregated title quality score for the title based on an additional factor of the title capitalization quality score.

### Claim 11

A method of claim being implemented via execution of computing instructions configured to run at one or more processors and stored at one or more non-transitory computer-readable media, the method comprising:
receiving a title of an item associated with an online catalog, wherein:
the item is associated with a product type; and
the product type comprises a predetermined set of attributes;
interpreting, using a natural language model, one or more attributes of the predetermined set of attributes within the title of the item;
determining a first title quality score for the title based on a first rule used to evaluate the one or more attributes of the title;
determining a second title quality score for the title based on a second rule used to evaluate the one or more attributes of the title;
determining an aggregated title quality score for the title based on at least the first title quality score and the second title quality score;
generating a content quality list for the title, wherein the content quality list identifies at least one or more content issues based on the aggregated title quality score for the title; and
sending instructions to display, on a user interface, a content quality dashboard comprising the content quality list for the item.

### Claim 12 (depends on 11)

The method of claim 11, further comprising:
pre-processing data for each respective attribute of the predetermined set of attributes of each respective product type in the online catalog; and
extracting respective key values for the each respective product type, wherein each of the respective key values comprise a respective product type, a respective attribute name, and a respective attribute value,
wherein:
interpreting, using a natural language model, the one or more attributes of the predetermined set of attributes within the title of the item comprises:
processing each respective title into one or more n-grams; and
searching a nested dictionary using each combination of the one or more n-grams as a search query to locate the one or more attributes in the title of the item.

### Claim 13 (depends on 12)

The method of claim 12, further comprising:
building the nested dictionary based on the respective key values, as extracted, for the each respective product type, wherein the nested dictionary comprises attributes of each product type, by;
lemmatize each respective word in each of the respective attribute values of the key values; and
storing each respective word of the respective attribute values of the key values, as lemmatized, in the nested dictionary database.

### Claim 14 (depends on 11)

The method of claim 11, wherein determining the first title quality score for the title further comprises:
generating an aggregated attribute score for the item based on at least (i) a quantity of the one or more attributes in the title and (ii) an assigned weight of each of the one or more attributes used in the title; and
assign the first title quality score for the item based on the aggregated attribute score.

### Claim 15 (depends on 11)

The method of claim 11, wherein:
determining the second title quality score for the title further comprises:
embedding, using a machine learning model, as trained, a product type of the item into a first vector;
embedding, using the machine learning model, as trained, each respective word in the title into one or more second vectors;
determining a similarity score between the first vector and each respective second vector of the one or more second vectors of each respective word in the title; and
assigning the second title quality score for the item based on a maximum of the similarity score.

### Claim 16 (depends on 15)

The method of claim 15, wherein the machine learning model further comprises a word2vec neural network.

### Claim 17 (depends on 15)

The method of claim 15, wherein determining the similarity score further comprises using a cosine similarity measurement.

### Claim 18 (depends on 11)

The method of claim 11, further comprising:
determining a third title quality score for the title based on a third rule used to evaluate a length of the title based on the product type; and
determining a fourth title quality score for the title based on a fourth rule used to evaluate an occurrence of prohibited characters within the title based on a predetermined list of prohibited characters,
wherein:
determining the aggregated title quality score for the title further comprises:
determining the aggregated title quality score for the title based on additional factors of the third title quality score and the fourth title quality score.

### Claim 19 (depends on 11)

The method of claim 11, further comprising:
determining a promotional title quality score for the title based on a promotional rule used to evaluate an occurrence of one or more promotional words within the title,
wherein:
determining the aggregated title quality score for the title further comprises:
determining the aggregated title quality score for the title based on an additional factor of the promotional title quality score.

### Claim 20 (depends on 11)

The method of claim 11, further comprising:
determining a title capitalization quality score for the title based on a capitalization rule used to evaluate capitalization of each word in the title,
wherein:
determining the aggregated title quality score for the title further comprises:
determining the aggregated title quality score for the title based on an additional factor of the title capitalization quality score.