# US 2021/0342550 A1 · Systems and methods for natural language processing in gaming environments

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

- Number: US 2021/0342550 A1
- Document ID: US20210342550A1
- Status: Published application
- Kind: A1
- Application number: 17/376,683
- Filed: 2021-07-15
- Published: 2021-11-04
- Earliest claimed: 2018-04-19
- Assignees: SG Gaming, Inc.
- Applicants: SG Gaming, Inc.
- Inventors: Muthukumaran PALANICHAMY; Mahesh SUNDARAMURTHY
- Web page: https://patentgrep.com/patent/US20210342550A1

## CPC

- G06F 40/30: Semantic analysis
- G06F 40/284: Lexical analysis, e.g. tokenisation or collocates
- G07F 17/3209: {Input means, e.g. buttons, touch screen}
- G07F 17/3288: {Betting, e.g. on live events, bookmaking}
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- G07F: COIN-FREED OR LIKE APPARATUS (coin sorting G07D3/00; coin testing G07D5/00; {handling coins or paper currencies apart from payment activated apparatus G07D; payment architectures, schemes or protocols G06Q20/00})

## Abstract

A natural language processing system includes logic circuitry that receives, from a first user channel, a user query including natural language (NL) input in a colloquial format from an end user, selects a first NLP application from a plurality of NLP applications, transforms, using the first NLP application, the NL input into a plurality of language tokens, extracts a command trigger from the language tokens, determines that the command trigger is linked to a command executable by a first wagering game channel, transmits a command query incorporating the command to the first wagering game channel, receives a command reply including wagering game data associated with the command from the first wagering game channel, generates, using the first NLP application, an NL response that includes the wagering game data and is structured in a colloquial format, and transmits the NL response to the end user via the first user channel.

## Claims

### Claim 1

A natural language processing (NLP) system for a wagering game environment, the NLP system comprising:
a first network interface communicatively coupled to a plurality of user channels configured to receive natural language (NL) input in a colloquial format from end users for requesting wagering game data;
a second network interface communicatively coupled to a plurality of wagering game channels, each wagering game channel associated with at least one type of wagering game data; and
logic circuitry communicatively coupled to the first network interface and the second network interface, the logic circuitry configured to:
receive, via the first network interface, a user query including the NL input associated with an end user from a first user channel of the plurality of user channels;
in response to the user query and determining the first user channel is associated with the user query, select a first NLP application from a plurality of NLP applications based on the first user channel, the first NLP application configured to process user queries from the first user channel;
transform, using the first NLP application, the NL input of the user query into a plurality of language tokens;
formulate a command trigger based on the plurality of language tokens;
determine that the command trigger is linked to a command executable by a first wagering game channel of the plurality of wagering game channels by comparing the command trigger to a plurality of available commands of the plurality of wagering game channels;
generate a command query incorporating the command, the command query structured in a predetermined format compatible with the first wagering game channel to execute the command;
receive, from the first wagering game channel, a command reply including wagering game data associated with the command;
generate an NL response including the wagering game data; and
transmit the NL response to the end user via the first network interface.

### Claim 2 (depends on 1)

The NLP system of claim 1, wherein the user query includes a user channel identifier associated with the first user channel, and wherein the first NLP application is retrieved in response to identifying the first user channel as the origin of the user query based on the user channel identifier.

### Claim 3 (depends on 1)

The NLP system of claim 1, wherein the NL input is a text-based input or an audible input provided by the end user to the first user channel.

### Claim 4 (depends on 1)

The NLP system of claim 1, wherein the command includes a command input to provide data associated with the NL input to the first wagering game channel, and wherein at least a portion of the command trigger is extracted to be incorporated in the command input.

### Claim 5 (depends on 1)

The NLP system of claim 1, wherein the logic circuitry is further configured to establish a session associated with the end user in response to the user query, wherein a subsequent command query of the session is generated based at least partially on at least one of the original user query or the wagering game data associated with the original user query.

### Claim 6 (depends on 1)

The NLP system of claim 1, wherein the command trigger is formulated using an NLP profile associated with the first NLP application that maps language tokens to command triggers based on historical command triggers, wherein the NLP profile is updated based at least partially upon the plurality of language tokens.

### Claim 7 (depends on 1)

The NLP system of claim 1, wherein the NL response is generated based at least partially on historical NL responses stored by an NLP profile associated with the first NLP application.

### Claim 8

A wagering game data system for providing wagering game data to end users, the wagering game data system comprising:
a first wagering game channel configured to store at least one type of wagering game data, the first wagering game channel associated with a set of available commands;
a natural language processing (NLP) system communicatively coupled to the first wagering game channel and a first user channel that receives natural language (NL) input from end users to request wagering game data, the NLP system comprising logic circuitry configured to:
receive a user query requesting wagering game data from an end user via the first user channel, the user query comprising NL input including a plurality of phrases in a colloquial format from the end user;
in response to the user query and determining the first user channel is associated with the user query, select a first NLP application from a plurality of NLP applications based on the first user channel, the first NLP application configured to process user queries of the first user channel;
transform, using the first NLP application, the NL input of the user query into a plurality of language tokens;
extract a command trigger from the plurality of language tokens;
identify a command linked to the command trigger by comparing the command trigger to the set of available commands of the first wagering game channel;
generate a command query incorporating the command, the command query structured in a predetermined format compatible with the first wagering game channel to execute the command;
transmit the command query to the first wagering game channel, wherein the first wagering game channel is configured to execute the command of the command query;
receive, from the first wagering game channel, a command reply including wagering game data associated with the command;
generate, by the first NLP application, an NL response including the wagering game data; and
transmit the NL response to the end user via the first user channel.

### Claim 9 (depends on 8)

The wagering game data system of claim 8, wherein the NL input is a text-based input or audible input provided by the end user to the first user channel.

### Claim 10 (depends on 8)

The wagering game data system of claim 8, wherein the logic circuitry is further configured to:
transmit an authentication challenge to the end user in response to the user query;
receive an authentication response including authentication data from the end user;
in response to the end user being successfully authenticated based on the authentication data, transmit the NL response to the first user channel; and
in response to the end user being unsuccessfully authenticated based on the authentication data, prevent the NL response from being generated.

### Claim 11 (depends on 10)

The wagering game data system of claim 10, wherein the first wagering game channel is configured to request authentication from the end user via the NLP system in response to the command query.

### Claim 12 (depends on 10)

The wagering game data system of claim 10, wherein each command of the set of available commands has a respective authentication requirement, and wherein the logic circuitry transmits the authentication challenge in response to the command linked to the command trigger having an authentication requirement indicating authentication is required.

### Claim 13 (depends on 8)

The wagering game data system of claim 8, wherein the NL response is generated using an NLP profile associated with the first NLP application, the NLP profile configured to map colloquial formatting to the wagering game data based on historical NL responses to generate the NL response in a colloquial format.

### Claim 14 (depends on 13)

The wagering game data system of claim 13, wherein the logic circuitry is further configured to receive a feedback message including NL input from the end user via the first user channel in response to the NL response, wherein the NLP profile is updated based on the feedback message.

### Claim 15

A method for providing wagering game data from wagering game environments in response to natural language (NL) input from end users using a natural language processing (NLP) system, the method comprising:
receiving, by logic circuitry of the NLP system, a user query associated with an end user requesting wagering game data from a first user channel of a plurality of user channels, the user query comprising NL input in a colloquial format;
in response to user query and determining the first user channel is associated with the user query, selecting, by the logic circuitry, a first NLP application from a plurality of NLP applications based on the first user channel, the first NLP application configured to process user queries from the first user channel;
transforming, by the logic circuitry using the first NLP application, the NL input of the user query into a plurality of language tokens;
extracting, by the logic circuitry, a command trigger from the plurality of language tokens;
determining, by the logic circuity, that the command trigger is linked to a command executable by a first wagering game channel of a plurality of wagering game channels by comparing the command trigger to a plurality of available commands executable by the plurality of wagering game channels;
generating, by the logic circuitry, a command query incorporating the command, the command query structured in a predetermined format compatible with the first wagering game channel to execute the command;
in response to transmitting the command query to the first wagering game channel, receiving, by the logic circuitry, a command reply including wagering game data associated with the command from the first wagering game channel;
generating, by the logic circuitry using the first NLP application, an NL response including the wagering game data; and
transmitting, by the logic circuitry, the NL response to the end user via the first user channel.

### Claim 16 (depends on 15)

The method of claim 15, wherein the NL input is a text-based input or an audible input provided by the end user to the first user channel.

### Claim 17 (depends on 15)

The method of claim 15, wherein the command includes a command input to provide the first wagering game channel with data associated with the NL input, and wherein a language token of the plurality of language tokens is extracted to be included in the command input of the command.

### Claim 18 (depends on 15)

The method of claim 15, wherein the command trigger is formulated using an NLP profile associated with the first NLP application that maps language tokens to command triggers based on historical command triggers, wherein the NLP profile is updated based at least partially upon the plurality of language tokens.

### Claim 19 (depends on 15)

The method of claim 15 further comprising:
transmitting, by the logic circuitry, an authentication challenge to the end user in response to the user query;
receiving, by the logic circuitry, an authentication response including authentication data from the end user; and
in response to the end user being unsuccessfully authenticated based on the authentication data, preventing, by the logic circuitry, the NL response from being generated.

### Claim 20 (depends on 15)

The method of claim 15 further comprising:
identifying, by the logic circuitry, a second command trigger from the plurality of language tokens;
determining, by the logic circuitry, that the second command trigger is linked to a second command executable by a second wagering game channel of the plurality of wagering game channels;
transmitting, by the logic circuitry, a second command query incorporating the second command to the second wagering game channel;
receiving, by the logic circuitry, a second command reply including additional wagering game data associated with the second command from the second wagering game channel; and
incorporating, by the logic circuitry, the additional wagering game data with the wagering game data from the first wagering game channel in the NL response.
