# US 2024/0004732 A1 · Application program interface analyzer for a universal interaction platform

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

- Number: US 2024/0004732 A1
- Document ID: US20240004732A1
- Status: Published application
- Kind: A1
- Application number: 18/347,199
- Filed: 2023-07-05
- Published: 2024-01-04
- Earliest claimed: 2015-12-31
- Assignees: Entefy Inc.
- Applicants: Entefy Inc.
- Inventors: Alston Ghafourifar; Philip Nathan Greenberg; Neil Achtman
- Web page: https://patentgrep.com/patent/US20240004732A1

## CPC

- G06F 9/543: {User-generated data transfer, e.g. clipboards, dynamic data exchange [DDE], object linking and embedding [OLE]}
- G06F 9/5027: {the resource being a machine, e.g. CPUs, Servers, Terminals}
- G06F 9/541: {via adapters, e.g. between incompatible applications}
- G06F 40/40: Processing or translation of natural language (natural language analysis G06F40/20; semantic analysis G06F40/30)
- G06F 2209/5015: Service provider selection
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)

## Abstract

An application program interface (API) analyzer that determines protocols and formats to interact with a service provider or smart device. The API analyzer identifies an API endpoint or websites for the service provider or smart device, determines a service category or device category, selects a category-specific corpus, forms a service-specific or device-specific corpus by appending information regarding the service provider or smart device to the category-specific corpus, and parses API documentation or the websites.

## Claims

### Claim 1

(canceled)

### Claim 2

A computer-implemented method of determining at least one user intent for interacting with an automated computing system, the computer-implemented method comprising:
receiving a message from a user for the automated computing system, wherein the message comprises message objects in the message;
executing, for the automated computing system, a natural language classifier comprising a machine learning engine including at least one machine learning model trained using a corpora of natural language expressions associated with different user intents, wherein the corpora of natural language expressions comprises service description information for services provided by at least one service provider and previous messages sent by one or more users including the user;
generating, by the natural language classifier using the machine learning engine, weighted probabilities that words or phrases from the message objects in the message are related to words or phrases from the corpora of natural language expressions;
determining, using the natural language classifier from the weighted probabilities, a natural language expression from the corpora of natural language expressions;
associating, using the natural language classifier, the natural language expression with the at least one user intent; and
providing, via the automated computing system, a response to the message based on the associating.

### Claim 3 (depends on 1)

The computer-implemented method of claim 1, wherein the corpora of natural language expressions associated with the different user intents comprises a mapping for the automated computing system for different probabilistic determinations by the natural language classifier of the different user intents based on the service description information and the previous messages, and wherein the mapping comprises links of each of the different user intents to the natural language expressions appearing in one or more of the service description information and the previous messages based on similarity scores determined by the machine learning engine.

### Claim 4 (depends on 1)

The computer-implemented method of claim 1, wherein the generating the weighted probabilities comprises:
applying a plurality of rules developed for the natural language classifier by analyzing a plurality of real-world examples for the service description information and the previous messages using the machine learning engine.

### Claim 5 (depends on 1)

The computer-implemented method of claim 1, wherein the service description information comprises descriptions of services provided by the at least one service provider, documentation regarding the services, or documentation of one or more smart devices providing the services on behalf of the at least one service provider, and wherein the one or more smart devices are communicatively coupled to the natural language classifier.

### Claim 6 (depends on 1)

The computer-implemented method of claim 1, wherein the message comprises a natural language statement by the user for a query or an instruction requesting a service, wherein the associating comprises mapping the natural language statement to one or more functions or one or more parameters that the at least one service provider or at least one smart device for the at least one service provider is configured to perform for the service, and wherein, prior to the generating the weighted probabilities, the computer-implemented method further comprises:
determining any settings by the user related to requesting the service via user interface of a universal interaction platform for the automated computing system,
wherein the settings are used with the message for requesting the service via the query or the instruction.

### Claim 7 (depends on 5)

The computer-implemented method of claim 5, wherein the automated computing system utilizes one or more Internet of Things (IoT) handlers to interact with the at least one service provider or the at least one smart device, wherein the one or more IoT handlers are configured to format the query or the instruction for the at least one service provider or the at least one smart device using a data format or a data protocol available to the universal interaction platform.

### Claim 8 (depends on 6)

The computer-implemented method of claim 6, wherein the response comprises the query or the instruction to the at least one service provider or the at least one smart device based on the associating, and wherein prior to the providing the response, the computer-implemented method further comprises:
formatting the response based on the data format or the data protocol.

### Claim 9 (depends on 1)

The computer-implemented method of claim 1, wherein the associating the natural language expression with the at least one user intent is based on at least one of a location of the user or a time that the message was provided by the user.

### Claim 10

An automated computing system including a universal interaction platform for determining at least one user intent for interacting with the automated computing system, the automated computing system comprising:
a non-transitory computer readable storage medium; and
a software module stored on the non-transitory computer readable storage medium, the software module configured to determining the at least one user intent from a message, wherein the non-transitory computer readable storage medium comprises computer executable instructions stored thereon to cause one or more processing units to perform operations comprising:
receiving the message from a user for the automated computing system, wherein the message comprises message objects in the message;
executing, for the automated computing system, a machine learning engine including at least one machine learning model trained using a corpora of natural language expressions associated with different user intents, wherein the corpora of natural language expressions comprises service description information for services provided by at least one service provider and previous messages sent by one or more users including the user;
generating, by the machine learning engine, weighted probabilities that words or phrases from the message objects in the message are related to words or phrases from the corpora of natural language expressions;
determining, using the weighted probabilities, a natural language expression from the corpora of natural language expressions;
associating the natural language expression with the at least one user intent; and
providing, via the automated computing system, a response to the message based on the associating.

### Claim 11 (depends on 9)

The automated computing system of claim 9, wherein the corpora of natural language expressions associated with the different user intents comprises a mapping for the automated computing system for different probabilistic determinations of the different user intents based on the service description information and the previous messages, and wherein the mapping comprises links of each of the different user intents to the natural language expressions appearing in one or more of the service description information and the previous messages based on similarity scores determined by the machine learning engine.

### Claim 12 (depends on 9)

The automated computing system of claim 9, wherein the generating the weighted probabilities comprises:
applying a plurality of rules developed for the natural language classifier by analyzing a plurality of real-world examples for the service description information and the previous messages using the machine learning engine.

### Claim 13 (depends on 9)

The automated computing system of claim 9, wherein the service description information comprises descriptions of services provided by the at least one service provider, documentation regarding the services, or documentation of one or more smart devices providing the services on behalf of the at least one service provider, and wherein the one or more smart devices are communicatively coupled to the machine learning engine.

### Claim 14 (depends on 9)

The automated computing system of claim 9, wherein the message comprises a natural language statement by the user for a query or an instruction requesting a service, wherein the associating comprises mapping the natural language statement to one or more functions or one or more parameters that the at least one service provider or at least one smart device for the at least one service provider is configured to perform for the service, and wherein, prior to the generating the weighted probabilities, the operations further comprise:
determining any settings by the user related to requesting the service via user interface of the universal interaction platform for the automated computing system,
wherein the settings are used with the message for requesting the service via the query or the instruction.

### Claim 15 (depends on 13)

The automated computing system of claim 13, wherein the automated computing system utilizes one or more Internet of Things (IoT) handlers to interact with the at least one service provider or the at least one smart device, wherein the one or more IoT handlers are configured to format the query or the instruction for the at least one service provider or the at least one smart device using a data format or a data protocol available to the universal interaction platform.

### Claim 16 (depends on 14)

The automated computing system of claim 14, wherein the response comprises the query or the instruction to the at least one service provider or the at least one smart device based on the associating, and wherein prior to the providing the response, the operations further comprise:
formatting the response based on the data format or the data protocol.

### Claim 17 (depends on 15)

The automated computing system of claim 15, wherein the associating the natural language expression with the at least one user intent is based on at least one of a location of the user or a time that the message was provided by the user.

### Claim 18

A non-transitory computer readable storage medium comprising computer executable instructions stored thereon to cause one or more processing units to perform operations comprising:
receiving a message from a user for an automated computing system, wherein the message comprises message objects in the message;
executing, for the automated computing system, a natural language classifier comprising a machine learning engine including at least one machine learning model trained using a corpora of natural language expressions associated with different user intents, wherein the corpora of natural language expressions comprises service description information for services provided by at least one service provider and previous messages sent by one or more users including the user;
generating, by the natural language classifier using the machine learning engine, weighted probabilities that words or phrases from the message objects in the message are related to words or phrases from the corpora of natural language expressions;
determining, using the natural language classifier from the weighted probabilities, a natural language expression from the corpora of natural language expressions;
associating, using the natural language classifier, the natural language expression with at least one user intent; and
providing, via the automated computing system, a response to the message based on the associating.

### Claim 19 (depends on 17)

The non-transitory computer readable storage medium of claim 17, wherein the corpora of natural language expressions associated with the different user intents comprises a mapping for the automated computing system for different probabilistic determinations by the natural language classifier of the different user intents based on the service description information and the previous messages, and wherein the mapping comprises links of each of the different user intents to the natural language expressions appearing in one or more of the service description information and the previous messages based on similarity scores determined by the machine learning engine.

### Claim 20 (depends on 17)

The non-transitory computer readable storage medium of claim 17, wherein the generating the weighted probabilities comprises:
applying a plurality of rules developed for the natural language classifier by analyzing a plurality of real-world examples for the service description information and the previous messages using the machine learning engine.

### Claim 21 (depends on 17)

The non-transitory computer readable storage medium of claim 17, wherein the service description information comprises descriptions of services provided by the at least one service provider, documentation regarding the services, or documentation of one or more smart devices providing the services on behalf of the at least one service provider, and wherein the one or more smart devices are communicatively coupled to the natural language classifier.
