# US 2021/0280175 A1 · Systems and methods for interpreting natural language search queries using training data

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

- Number: US 2021/0280175 A1
- Document ID: US20210280175A1
- Status: Published application
- Kind: A1
- Application number: 16/807,419
- Filed: 2020-03-03
- Published: 2021-09-09
- Earliest claimed: 2020-03-03
- Assignees: Rovi Guides, Inc.
- Applicants: Rovi Guides, Inc.
- Inventors: Jeffry Copps Robert Jose; Ajay Kumar Mishra
- Web page: https://patentgrep.com/patent/US20210280175A1

## CPC

- G06F 16/2228: {Indexing structures}
- G06F 16/243: {Natural language query formulation}
- G06F 16/2455: Query execution
- G06N 3/0499: Feedforward networks
- G06N 3/08: Learning methods
- G06N 3/09: Supervised learning
- G06N 7/01: Probabilistic graphical models, e.g. probabilistic networks
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS

## Abstract

A frequency of occurrence for each term in a training data set is determined in relation to the entire training data set. A relational data structure is generated that associates each term in the training data with its respective frequency. Any term that has a frequency below a threshold frequency is then added to a list of relevant words. When a natural language search query is received, a plurality of terms in the natural language search query are identified and compared with the list of relevant words. If any term of the natural language search query is included in the relevant words list, that term is identified as a keyword. The natural language search query is then interpreted based on any identified keywords.

## Claims

### Claim 1

A computer-implemented method for interpreting a natural language search query based on training data comprising a first plurality of terms, the method comprising, using processing circuitry for:
determining a respective frequency with which each of the first plurality of terms occurs in the training data;
generating a relational data structure associating each term of the first plurality of terms with its respective frequency;
adding a term of the first plurality of terms to a list of relevant terms when its frequency is below a threshold;
receiving the natural language search query;
identifying a second plurality of terms in the natural language search query;
determining, for each term of the second plurality of terms, whether the term is included in the list of relevant terms;
in response to determining that a term of the second plurality of terms is included in the list of relevant terms, identifying the term as a keyword; and
interpreting the natural language search query based on the identified keywords.

### Claim 2 (depends on 1)

The method of claim 1, wherein determining a respective frequency with which each of the first plurality of terms occurs in the training data comprises:
counting the total number of words contained in the training data;
determining the total number of occurrences of each respective term of the plurality of terms in the training data; and
calculating, for each term of the plurality of terms, a percentage of the total number of words contained in the training data corresponding to the total number of occurrences of each respective term of the plurality of terms in the training data.

### Claim 3 (depends on 1)

The method of claim 1, wherein generating a relational data structure associating each term of the first plurality of terms with its respective frequency comprises:
creating a data structure comprising at least a token field and a corresponding value field;
adding to the data structure, for each term of the first plurality of terms, the respective term as a token; and
setting the value corresponding to the token to the frequency of the respective term.

### Claim 4 (depends on 1)

The method of claim 1, wherein adding a term of the plurality of terms to a list of relevant terms when its frequency is below a threshold comprises:
determining whether the frequency of the term meets or exceeds a threshold frequency;
in response to determining that the frequency meets or exceeds the threshold frequency, determining that the respective term is not relevant; and
in response to determining that the frequency is below the threshold frequency, adding the term to the list of relevant terms.

### Claim 5 (depends on 1)

The method of claim 1, wherein identifying a second plurality of terms in the natural language search query comprises:
splitting the natural language search query into a plurality of words;
analyzing a first word of the plurality of words;
determining, based on analyzing the first word, whether the first word can be part of a phrase;
in response to determining that the first word can be part of a phrase, analyzing the first word together with a second word that immediately follows the first word;
determining, based on analyzing the first word together with the second word, whether the first word and the second word form a phrase together;
in response to determining that the first word and the second word form a phrase together, identifying the first and second word as a single term; and
in response to determining that the first word does not form a phrase with the second word, identifying the first word as single term.

### Claim 6

A system for interpreting a natural language search query based on training data comprising a first plurality of terms, the system comprising control circuitry configured to:
determine a respective frequency with which each of the first plurality of terms occurs in the training data;
generate a relational data structure associating each term of the first plurality of terms with its respective frequency;
add a term of the first plurality of terms to a list of relevant terms when its frequency is below a threshold;
receive the natural language search query;
identify a second plurality of terms in the natural language search query;
determine, for each term of the second plurality of terms, whether the term is included in the list of relevant terms;
in response to determining that a term of the second plurality of terms is included in the list of relevant terms, identify the term as a keyword; and
interpret the natural language search query based on the identified keywords.

### Claim 7 (depends on 6)

The system of claim 6, wherein the control circuitry configured to determine a respective frequency with which each of the first plurality of terms occurs in the training data is further configured to:
count the total number of words contained in the training data;
determine the total number of occurrences of each respective term of the plurality of terms in the training data; and
calculate, for each term of the plurality of terms, a percentage of the total number of words contained in the training data corresponding to the total number of occurrences of each respective term of the plurality of terms in the training data.

### Claim 8 (depends on 6)

The system of claim 6, wherein the control circuitry configured to generate a relational data structure associating each term of the first plurality of terms with its respective frequency is further configured to:
create a data structure comprising at least a token field and a corresponding value field;
add to the data structure, for each term of the first plurality of terms, the respective term as a token; and
set the value corresponding to the token to the frequency of the respective term.

### Claim 9 (depends on 6)

The system of claim 6, wherein the control circuitry configured to add a term of the plurality of terms to a list of relevant terms when its frequency is below a threshold is further configured to:
determine whether the frequency of the term meets or exceeds a threshold frequency;
in response to determining that the frequency meets or exceeds the threshold frequency, determine that the respective term is not relevant; and
in response to determining that the frequency is below the threshold frequency, add the term to the list of relevant terms.

### Claim 10 (depends on 6)

The system of claim 6, wherein the control circuitry configured to identify a second plurality of terms in the natural language search query is further configured to:
split the natural language search query into a plurality of words;
analyze a first word of the plurality of words;
determine, based on analyzing the first word, whether the first word can be part of a phrase;
in response to determining that the first word can be part of a phrase, analyze the first word together with a second word that immediately follows the first word;
determine, based on analyzing the first word together with the second word, whether the first word and the second word form a phrase together;
in response to determining that the first word and the second word form a phrase together, identify the first and second word as a single term; and
in response to determining that the first word does not form a phrase with the second word, identify the first word as single term.

### Claim at position 11

-15. (canceled)

### Claim 16

A non-transitory computer-readable medium having non-transitory computer-readable instructions encoded thereon for interpreting a natural language search query based on training data comprising a first plurality of terms that, when executed by control circuitry, cause the control circuitry to:
determine a respective frequency with which each of the first plurality of terms occurs in the training data;
generate a relational data structure associating each term of the first plurality of terms with its respective frequency;
add a term of the first plurality of terms to a list of relevant terms when its frequency is below a threshold;
receive the natural language search query;
identify a second plurality of terms in the natural language search query;
determine, for each term of the second plurality of terms, whether the term is included in the list of relevant terms;
in response to determining that a term of the second plurality of terms is included in the list of relevant terms, identify the term as a keyword; and
interpret the natural language search query based on the identified keywords.

### Claim 17 (depends on 16)

The non-transitory computer-readable medium of claim 16, wherein execution of the instruction to determine a respective frequency with which each of the first plurality of terms occurs in the training data further causes the control circuitry to:
count the total number of words contained in the training data;
determine the total number of occurrences of each respective term of the plurality of terms in the training data; and
calculate, for each term of the plurality of terms, a percentage of the total number of words contained in the training data corresponding to the total number of occurrences of each respective term of the plurality of terms in the training data.

### Claim 18 (depends on 16)

The non-transitory computer-readable medium of claim 16, wherein execution of the instruction to generate a relational data structure associating each term of the first plurality of terms with its respective frequency further causes the control circuitry to:
create a data structure comprising at least a token field and a corresponding value field;
add to the data structure, for each term of the first plurality of terms, the respective term as a token; and
set the value corresponding to the token to the frequency of the respective term.

### Claim 19 (depends on 16)

The non-transitory computer-readable medium of claim 16, wherein execution of the instruction to add a term of the plurality of terms to a list of relevant terms when its frequency is below a threshold further causes the control circuitry to:
determine whether the frequency of the term meets or exceeds a threshold frequency;
in response to determining that the frequency meets or exceeds the threshold frequency, determine that the respective term is not relevant; and
in response to determining that the frequency is below the threshold frequency, add the term to the list of relevant terms.

### Claim 20 (depends on 16)

The non-transitory computer-readable medium of claim 16, wherein execution of the instruction to identify a second plurality of terms in the natural language search query further causes the control circuitry to:
split the natural language search query into a plurality of words;
analyze a first word of the plurality of words;
determine, based on analyzing the first word, whether the first word can be part of a phrase;
in response to determining that the first word can be part of a phrase, analyze the first word together with a second word that immediately follows the first word;
determine, based on analyzing the first word together with the second word, whether the first word and the second word form a phrase together;
in response to determining that the first word and the second word form a phrase together, identify the first and second word as a single term; and
in response to determining that the first word does not form a phrase with the second word, identify the first word as single term.

### Claim at position 17

-25. (canceled)
