Methods, systems, and non-temporary computer-readable media for assisting company employees with an AI engine.

Abstract

To provide methods, systems, and non-temporary computer-readable media for assisting company employees with an AI engine. [Solution] The AI-based configuration tool allows users to interact with a conversational AI that guides them through various configuration steps, eliminating the need to navigate through cumbersome menus and settings. The AI ​​also provides visual feedback and can drive UI configuration based on user input, ensuring a seamless and intuitive setup process.

Description

TECHNICAL FIELD

[0001] Embodiments of the present invention relate to a method, system, and non-transitory computer-readable medium for supporting company employees with an AI engine.

BACKGROUND ART

[0002]

[0001] In the retail industry, the configuration (or setting), setup and management of point-of-sale device systems ("POS") are complex, and it is difficult to understand the impact of all user configurations. Furthermore, customer business rules add to this complexity.

SUMMARY OF THE INVENTION

PROBLEMS TO BE SOLVED BY THE INVENTION

[0003] The problem to be solved by the present invention is to provide a method, system, and non-transitory computer-readable medium for supporting company employees with an AI engine.

MEANS FOR SOLVING THE PROBLEMS

[0004] The embodiment of the method is a method for assisting a company employee with an AI engine, comprising: a processor of a generative AI-based system receiving a request from the employee via a chat interface of the POS device, either by voice or text, to configure and / or set up the POS device; the processor using the data contained in the request to determine at least one action to be performed by at least one backend source system of the company to configure and / or set up the POS device; the processor generating a response for the at least one backend source system to perform when configuring and / or setting up the POS device; and the processor transmitting the response to the at least one backend source system so that the at least one backend source system directly configures and / or sets up the POS device. [Brief explanation of the drawing]

[0005]

[0002] The purposes and features of this disclosure can be better understood by referring to the drawings and claims described below. The drawings are not necessarily to scale and instead are generally intended to illustrate the principles of the embodiments of this disclosure. In the drawings, similar reference numerals are used to indicate similar parts throughout the drawings.

[0006] [Figure 1] This figure shows an exemplary high-level architecture based on the implementation of this disclosure. [Figure 2] This figure shows an example architecture based on the implementation of this disclosure. [Figure 3] This figure shows an example of a layered architecture and environment for a corporate artificial intelligence system implemented in this disclosure. [Figure 4] This figure shows an exemplary process that can be performed according to the implementation of this disclosure. [Figure 5]This figure shows another exemplary process that can be performed according to the implementation of this disclosure. [Modes for carrying out the invention]

[0007]

[0009] Generally, this disclosure provides AI-based tools used by point-of-sale devices ("POS"). Rather than navigating various configurations, each POS should be configured via a prompt-based chat interface that navigates and configures the system for the POS user based on user prompts. The resulting actions may drive a UI configuration interface or visualize the configuration by returning it to the user in the chat session. AI-based POS system configuration tools can also be used to augment various configuration sections of a POS back-office application with targeted prompt entries.

[0008]

[0010] The proposed POS system configuration tool leverages a chat-based interface to simplify the often complex process of setting up and managing POS systems. This tool features a conversational AI that guides the user through various configuration steps, enabling interaction and eliminating the need to navigate through cumbersome menus and settings. The AI ​​also provides visual feedback and can drive UI configuration based on user input, ensuring a seamless and intuitive setup process.

[0009]

[0011] A brief overview of how this might work in some common embodiments is provided below.

[0010]

[0012] 1. User interaction via chat interface and enhanced prompts

[0011]

[0013] Users interact with the POS system configuration tool via a chat interface. This interface can be accessed via a web application, a mobile application, or directly within the POS system. The chat interface uses natural language processing to understand and respond to user prompts.

[0012]

[0014] 2. AI-driven guidance

[0013]

[0015] The AI ​​engine is integrated with a chat interface to provide real-time guidance and responses. When a user asks a question or requests a specific configuration, the AI ​​tool interprets the query and provides step-by-step instructions, or, if necessary, directly configures the system and presents the output.

[0014]

[0016] 3. Configuration Actions

[0015]

[0017] The AI ​​tool can perform configuration actions such as setting up payment methods, adding new products, configuring tax rates, business rules, receipt configurations, and managing user roles. Users can also request visual feedback, and the AI-driven system can generate a UI preview of the configuration changes.

[0016]

[0018] 4. Visualization and Feedback

[0017]

[0019] The system can return and visualize configuration settings to the user within the chat interface. For example, if a user is configuring a new product, the chat interface can display product details for confirmation.

[0018]

[0020] AI can also provide recommendations based on best practices or common configurations.

[0019]

[0021] 5. Implementation Form

[0020]

[0022] Architecture Overview: Frontend: The chat interface functions as the primary user interaction point. Backend: The backend system processes chat interactions, integrates with the AI engine, and executes configuration actions.

[0021]

[0023] The AI engine may be an AI engine driven by a model or a custom-trained NLP model that understands user prompts and generates appropriate responses and actions.

[0022]

[0024] Database: A database (e.g., MongoDB) that stores configuration settings, user data, and interaction logs. The backend provides the AI engine with appropriate context and examples to drive the AI engine. Prompt flows can be used to guide the results to practical outcomes.

[0023]

[0025] Security and Permissions:

[0024]

[0026] User authentication and authorization to ensure only authorized personnel can perform specific configurations. Logging and monitoring of all configuration actions for audit purposes.

[0025]

[0027] 6. Solved Problems

[0026]

[0028] The proposed solution addresses the complex and time-consuming problem of POS system configuration by providing an intuitive chat-based interface that simplifies the configuration process, utilizing AI to guide users through setup steps and perform configurations based on natural language prompts, providing real-time visualization and feedback to ensure users understand the changes being made, and reducing the need for extensive training and manual navigation through traditional configuration menus. This approach not only improves the user experience, but also enhances efficiency and accuracy when configuring the POS system.

[0027]

[0029] Detailed description of several embodiments

[0028]

[0030] The following describes various examples and further details of this disclosure. The following descriptions provide specific details for the explanation that will enable a full understanding and implementation of these examples. However, those skilled in the art will understand that this disclosure can be implemented without many of these details. Furthermore, some well-known structures or functions may not be illustrated or described in detail to avoid unnecessarily obscuring the relevant explanations.

[0029]

[0031] The terms used in the following descriptions are intended to be interpreted in their broadest and most reasonable manner, even though they are used in conjunction with the detailed descriptions of the specific examples of this disclosure. Certain terms may even be emphasized below. However, any term intended to be interpreted in any limited form is so expressly and specifically defined in the section of the detailed description.

[0030]

[0032] Referring here to the drawings, this disclosure provides novel intelligent systems for various applications, such as retail applications. In some embodiments, the intelligent systems employ large-scale language models (LLMs) and generative AI, technologies that can understand and respond to human language in a meaningful and contextual manner to provide an automated method for company employees to set up and configure POS systems.

[0031]

[0033] In this regard, the implementations of this disclosure generally relate to a computer implementation platform for an artificial intelligence (AI)-based digital POS setup / configuration agent (also referred to herein as the AI ​​framework and / or AI assistant). The implementations of this disclosure may also relate to a hybrid platform that provides on-premise components and / or leverages cloud-hosted components. Generally, as will be described in more detail herein, the AI ​​framework of this disclosure reduces the burden of enterprise tasks, enables interaction across multiple channels, and maintains the security and confidentiality of information (e.g., customer and / or company sensitive information (SI)) while leveraging cloud-hosted components to perform services. It can assist employees in configuring / setting up POS systems, managing inventory, making data-driven decisions, and any other tasks relating to the management of POS systems.

[0032]

[0034] As will be described in more detail herein, an implementation of the present disclosure includes an action to receive communication data from a device, wherein the communication data includes data entered by the user of the device; an action to determine the context based on an extended finite state machine that defines contexts and transitions between contexts; an action to send a service request to at least one cloud-hosted service, wherein the service request is provided at least in part on masking sensitive information contained in the communication data; an action to receive a service response from at least one cloud-hosted service, wherein the service response includes one or more intents and entities; an action to determine at least one action to be performed by at least one backend source system based on the service response; an action to provide a response at least in part on the action result received from at least one backend source system; and an action to send result data to the device.

[0033]

[0035] This disclosure of an AI-based system represents a significant advance in retail technology, providing retail workers with an easy-to-use tool for performing a variety of complex tasks for POS systems. The use of advanced AI technology enables a more contextually aware and user-friendly system, improving the efficiency and effectiveness of retail workers.

[0034]

[0036] Figure 1 shows an exemplary high-level architecture 100 according to an implementation of the present disclosure. The exemplary architecture 100 includes a POS device 102 (and / or mobile device), backend systems 108, 110, and a network 112. In some examples, the network 112 includes a local area network (LAN), a wide area network (WAN), the Internet, a cellular network, a public switched telephone network (PSTN), a private branch exchange (PBX), or any appropriate combination thereof, connecting the website, the device (e.g., POS device 102), and the backend systems (e.g., backend systems 108, 110). In some examples, the network 112 may be accessible via wired and / or wireless links.

[0035]

[0037] In the illustrated examples, each of the backend systems 108, 110 includes at least one server system 114 and a data store 116 (e.g., a database). In some examples, one or more backend systems host one or more computer implementation services that a user can interact with using a device. For example, as described in more detail herein, backend systems 108, 110 may host an AI framework according to an implementation of this disclosure. The backend systems handle chat interactions, integrate with the AI ​​engine, and perform configuration actions directly in the POS (which may be a chat session).

[0036]

[0038] The backend provides the AI ​​engine with appropriate context and examples to drive the AI ​​engine, which may be obtained using a company's private database / model, open-source models, public AI models, or any other AI models or databases for training the POS training AI model described herein.

[0037]

[0039] In some examples, the POS device 102 is a point-of-sale (POS) device, such as a self-checkout system, configured to identify products being presented for purchase by a customer. The POS device 102 has imaging technology to scan product codes or the products themselves, communicates with a database and backend system to determine the price of the products, and also communicates with a payment system.

[0038]

[0040] It should be noted that each POS device 102 may include a desktop computing system, a laptop computing system, a handheld computer, a tablet computer, a personal digital assistant (PDA), a mobile phone, network equipment, a camera, a smartphone, a telephone, a mobile phone, a media player, a navigation device, an email device, a game console, or any suitable combination of two or more of these devices, or any suitable type of computing device such as other data processing devices.

[0039]

[0041] In the illustrated example, the POS device 102 is used by user 120. According to this disclosure, user 120 interacts with the AI ​​framework of this disclosure using the POS device 102. In some examples, user 120 may include a customer of the company providing the AI ​​framework. For example, user 120 may include a customer who communicates with the company through one or more channels using device 102. Depending on the implementation of this disclosure, as described in further detail herein, user 120 may provide the platform with oral input (e.g., voice), text input, and / or visual input (e.g., images, videos), and the platform may process the input to perform one or more actions and / or provide one or more responses.

[0040]

[0042] Implementations of the Disclosure are described in further detail herein with reference to exemplary contexts. An example of such contexts includes a retail store (e.g., a grocery store) as an enterprise implementing the AI ​​framework of the Disclosure. In the first exemplary context, an employee (e.g., a cashier) can interact with the hybrid AI framework to perform tasks related to POS configuration (e.g., POS training, POS setup, inventory setup / maintenance, etc.). However, it is believed that implementations of the Disclosure can be realized in any appropriate context.

[0041]

[0043] Figure 2 shows an exemplary architecture 200 of an AI framework according to an implementation of the present disclosure. In some examples, components of the exemplary architecture 200 can be hosted on one or more backend systems (e.g., backend systems 108, 110 in Figure 1). In some examples, each component of the exemplary architecture 200 is provided as one or more computer executable programs that run on one or more computing devices.

[0042]

[0044] In the illustrated example, the exemplary architecture 200 includes an on-premises unit 202 and a cloud unit 204 (e.g., the AI ​​assistant of this disclosure). In some examples, the on-premises unit 202 is hosted on a backend system of a company providing the AI ​​framework (e.g., a retail store), i.e., within the company's intranet. For example, the on-premises unit 202 can be hosted on backend system 108 in Figure 1. In some examples, the cloud unit 204 is hosted on one or more backend systems of a respective cloud-based service provider. For example, at least a portion of the cloud unit 204 can be hosted on backend system 110 in Figure 1.

[0043]

[0045] In the illustrated example, the exemplary architecture 200 includes a channel 206 through which a user (e.g., a retail store employee) can communicate with the on-premises unit 202. The exemplary channel may include voice, chat, text, etc. For example, a user can use a device (e.g., a POS device 102 in Figure 1) to provide speech-to-text to the on-premises unit 202 (e.g., via the network 112 in Figure 1).

[0044]

[0046] As another example, a user can provide input to the on-premises unit 202 via one or more secure networks (e.g., a first secure network (SNX), a second secure network (SNY)) (e.g., via network 112 in Figure 1). As yet another example, a user can provide input via a messaging application (e.g., chat) associated with the POS device 102. While various methods can be used, the primary method used as an example herein is for the user to converse with the POS device 102, which then uses a Large Language Model (LLM) to converse and process the user's voice input. In this regard, the user of the POS device 102 interacts with the POS system configuration tool via a chat interface presented to the user on the POS device 102. It should be noted, however, that this interface can be accessed via a web application, a mobile application, or directly within the POS system. The chat interface uses natural language processing to understand and respond to user prompts.

[0045]

[0047] It should be understood that various channels 206 can be on-premises 202 or off-premises (as shown in Figure 2). In this regard, this disclosure is not limited to employees or company representatives of a retail company that will be on-premises to run the systems herein.

[0046]

[0048] In some implementations, the on-premise unit 202 includes a channel connector 208, a data integration layer 210, authentication middleware 212, artificial intelligence (AI) middleware 214, a connector 216, a web service publisher 218, a backend source system 220, a data management layer 222, and one or more databases 224. Generally, as will be described in more detail herein, user communications are received via one or more of the connectors 208 and provided to the AI ​​middleware 214 via the data integration layer 210. Users providing communications (e.g., requests, messages, etc.) are authenticated by the authentication middleware 212 to ensure that the user is an employee or company representative. In this regard, the authentication middleware 212 compares the employee number associated with the user with one in a database containing pre-stored employee numbers and corresponding passwords. The authentication middleware 212 then determines whether the password provided by the user matches a password in the database. Other authentication methods are possible, and it should be understood that this application should not be limited to the authentication methods described above.

[0047]

[0049] Next, once the user is authenticated by the authentication middleware 212, a session is established between the device the user is using (for example, device 102 in Figure 1) and the AI ​​middleware 214. Authentication can be performed using any appropriate method (e.g., credentials (username / password), token), as described above. During the session, the AI ​​middleware 214 can utilize cloud-hosted intelligent services provided by the cloud unit 204 via the connector 216.

[0048]

[0050] In some examples, the data integration layer 210 included in the on-premises unit 202 converts the format of incoming messages to a standard format that can be processed by the components of the on-premises unit 202. For example, a message may be received in a channel-specific format (e.g., a first format if received via web chat or chatbot, a second format if received via a secure network), and the data integration layer 210 converts the message to a standard format (e.g., text). In some examples, the data integration layer 210 converts outgoing messages from a standard format to a channel-specific format. In some examples, the channel connector 208 abstracts the details of integrating with various channels 206 of user communication from the rest of the on-premises unit 202. This includes abstracting communication details and data representations. In some examples, the connector 216 abstracts the details of integrating with cloud-hosted services from the rest of the AI ​​assistant. This includes communication details and data representations.

[0049]

[0051] In some implementations, the AI ​​middleware 214 includes a session manager 230, a confidential information masking component 232, an action handler 234, a communication orchestrator 236, an exception handler 238, a response generator 240, and one or more data validation modules 242. Each of these components is described below.

[0050]

[0052] First, the session manager 230 maintains the state and context of the interaction between the user and the AI ​​engine. In some examples, a new session is instantiated each time the user initiates communication with the AI ​​engine, and the context is maintained throughout based on mapped intent, which will be described in more detail herein. In this way, the AI ​​engine can provide context-relevant responses. In fact, the AI ​​engine is integrated with the chat interface to provide real-time guidance and responses. To do this, when a user asks a question or requests a specific configuration from the POS device 102, the AI ​​tool interprets the query and provides step-by-step instructions, or, if necessary, directly configures the POS device 102 and presents the output. The AI ​​tool can perform a variety of configuration actions, such as setting up payment methods, adding new products, configuring tax rates, business rules, receipt configurations, and managing user roles. The AI ​​engine can also provide recommendations based on best practices or common configurations.

[0051]

[0053] Next, in some examples, the sensitive information masking component 232 masks data that has been determined to be sensitive, such as payment information. In this way, sensitive information is not exposed on the public internet or elsewhere. In some examples, masking involves removing data (sensitive information) from a dataset that is communicated over a network. In some examples, the occurrence of sensitive information can be determined based on rules and / or regular expressions (regex). In some examples, masking can include character substitution, word substitution, shuffling, number / date variance, encryption, truncation, character masking, and / or partial masking.

[0052]

[0054] It can be determined that information from cloud-hosted services is necessary (for example, the intent / entities of a message received from a user need to be determined). This allows for the construction of a message containing a dataset that the cloud-hosted service should process. Any entities within a dataset that are determined to be sensitive can be removed from the dataset. For example, entities containing sensitive information do not need to receive an accurate response from one or more cloud-hosted services (for example, a user's account number is not required for the cloud-hosted service to determine the intent of the message).

[0053]

[0055] In some examples, the action handler 234 coordinates the execution of one or more actions. For example, the action handler 234 sends a request to one or more backend source systems to perform an action. Symbolic actions may include, but are not limited to, retrieving stored information, recording information, and performing calculations. The action handler 234 may also perform any of the configuration actions, such as setting up payment methods, adding new products, configuring tax rates, business rules, receipt configurations, and managing user roles. For example, it may determine that the action to be performed includes configuring pricing for grocery store inventory (e.g., further described herein), and to perform this action regarding grocery store inventory, the action handler 234 may send a request containing the item number of interest and other information (e.g., Brand X potato chips, $12.99, snacks). The action handler 234 then sends the request to a backend source system (e.g., a system managing the inventory database) and can receive a response from the backend source system containing the result of the action (e.g., a string value indicating the item and related details has been added to the current inventory database). The action handler 234 can communicate with the communication orchestrator 236 to communicate with the appropriate database or AI training model in order to complete the above actions. As mentioned above, the AI ​​engine may be an AI engine driven by a model that understands user prompts and generates appropriate responses and actions or a custom-trained NLP model, and thus the action handler 234 and the communication orchestrator 236 facilitate the connection between the POS-specific database and the AI ​​model to complete a specific task directly associated with the POS device 102.

[0054]

[0056] In some examples, the communications orchestrator 236 orchestrates communication between components both inside and outside the AI ​​middleware 214. For example, the communications orchestrator 236 receives communication from the data integration layer 210 (e.g., requests provided by user devices) and provides communication back to the data integration layer 210 (e.g., responses to user devices). In another example, the communications orchestrator 236 communicates with the data management layer 222 to access databases 224 such as a database for establishing payment methods (potentially external databases or systems associated with banking entities, entities trained for payment processing, etc.), a new product database managed by the company so that the POS can automatically configure itself with newly added products when new product entries are added, a tax rate database (so that the POS can configure the sales tax rate for shoppers), a business rules database (so that companies can add or modify business rules), and a database defining user roles within the company / store / etc.

[0055]

[0057] In another example, the communications orchestrator 236 communicates with a cloud-hosted service via the connector 216 (for example, by sending requests and receiving responses). In some examples, upon receiving a message, the communications orchestrator 236 first provides message information to one or more data validation modules 242, as described herein, to determine whether the data provided in the message is in the expected format.

[0056]

[0058] In some examples, the exception handler 238 handles any exceptions (e.g., errors) that may occur in the data. Exceptions include, but are not limited to, inability to access a cloud-hosted service, a backend system being unresponsive, and runtime exceptions (such as a null pointer exception). In some examples, the response generator 240 constructs a response message to send back to the user. In some examples, the response message may include the result of an action and one or more entities. Continuing with the examples provided herein, an exemplary response message may include "Added to database: Brand X potato chips, snacks, price $12.99" (e.g., a response to an employee request for inventory of "Brand X potato chips, $12.99, snacks category"). In some examples, the response message may include requests for clarification and / or corrected information.

[0057]

[0059] The system can make any setting item visible to the user within the chat interface. For example, if a user is configuring a new product, the chat interface can display product details for confirmation. The AI ​​can also provide recommendations based on best practices or common configurations by using public AI models, constantly searching for other cases and configurations, and training itself to notify the user of possible changes to the POS device 102 based on what the AI ​​tool finds.

[0058]

[0060] In either case, the chat interface serves as the primary point of user interaction, while the backend system processes the chat conversations, integrates with the AI ​​engine, and performs configuration actions.

[0059]

[0061] Next, in some examples, one or more data validation modules 242 validate information provided by the user through the channel. In some examples, the validation module 242 determines whether the received information is in the expected format. For example, the AI ​​assistant may request an item number to be assigned to an item (e.g., a chat message stating, "Please tell me the item number"). The user's response message is processed by the data validation module 242 to determine whether the response message contains an item number in the expected format (e.g., a six-digit number). For example, if the response message contains "ABC123", the data validation module 242 can determine that the item number is not in an incorrect format. In such a case, a message indicating that the item number does not conform to the expected format (e.g., "This is an invalid format. Please provide a number in the six-digit numeric format (######)") can be returned to the user. In another example, if the response message contains "123456", the data validation module 242 can determine that the item number is in an incorrect format. In such a case, the AI ​​assistant can proceed with the conversation (e.g., transitions between contexts, as will be described in more detail herein).

[0060]

[0062] In some implementations, the data validation module 242 determines whether a request should be sent to one or more cloud-hosted services. Continuing the example above, if the response message contains "123456", the data validation module 242 may determine that the item number is in an incorrect format and there is no need to retrieve the intent or entity from the cloud-hosted services.

[0061]

[0063] In some implementations, components of the AI ​​middleware 214 can be referred to as conversation managers. Exemplary components of a conversation manager may include the session manager 230 and sensitive information masking component 232 of the AI ​​middleware 214 in Figure 2. Other components of the conversation manager may include a conversation logger, a domain integration component, and a conversation rule base (not shown in Figure 2). In some examples, the conversation logger records all communication between the user and the AI ​​assistant. In some examples, the logged conversation details can be used by the enterprise to monitor the effectiveness of the AI ​​assistant in delivering the intended user experience. In some examples, the domain integration component abstracts the details of integration with the backend source system 220 from the rest of the AI ​​assistant components. This includes abstracting communication details and data representation. In some examples, conversation rules manage conversations with the user and support an extended finite state machine approach for transitioning between different contexts based on multiple factors (e.g., intent, entity, action). For example, a response to a user is a function of context, intent, entity, and action.

[0062]

[0064] As described above, the AI ​​middleware (the same as the “AI engine” or “AI tool” as referred to herein) can perform configuration actions such as setting up payment methods, adding new products, configuring tax rates, business rules, receipt configurations, and managing user roles. The AI ​​can also provide recommendations based on best practices or common configurations. All of this is done through a chat interface that serves as the primary point of user interaction, with the backend system processing the chat interactions, integrating with the AI ​​engine, and performing the above (and other) configuration actions.

[0063]

[0065] In some implementations, the cloud unit 204 includes one or more cloud-hosted services. Examples of cloud-hosted services include, but are not limited to, natural language processing (NLP) (e.g., entity extraction, intent extraction), sentiment analysis, speech-to-text, and translation. An example of a speech-to-text service is Google Cloud Speech, provided by Google, Inc., Mountain View, California. In some examples, Google Cloud Speech converts audio data into text data by processing the audio data through a neural network model. An example of an NLP service is TensorFlow, provided by Google, Inc., Mountain View, California. In some examples, TensorFlow can be described as an open-source software library for numerical computation using a dataflow graph. While this specification refers to exemplary cloud-hosted services, implementations of this disclosure can be realized using any suitable cloud-hosted service. As described above, the AI ​​engine may be driven by a model or a custom-trained NLP model that understands user prompts and generates appropriate responses. The AI ​​engine is integrated with a chat interface to provide real-time guidance and responses. When a user asks a question or requests a specific configuration, the AI ​​tool interprets the query and provides step-by-step instructions, or configures the system directly and presents the output if necessary.

[0064]

[0066] In some implementations, the AI ​​middleware 214 communicates with one or more of the backend source systems 220 via a web service exposure 218. Exemplary backend source systems 220 include, but are not limited to, employee training systems, employee payment systems, inventory systems, customer relationship management (CRM) systems, billing systems, and enterprise resource planning (ERP) systems. In some implementations, the AI ​​middleware 214 communicates with the data management layer 222 to access the database 224. In the illustrated example, the data management layer 222 includes (ETL) components, data preparation components, data transformation components, and data extraction components.

[0065]

[0067] Database 224 (which may also include database 116 from Figure 1) includes databases owned and operated by the company on-premise 202, such as a product information database, a customer business rules database, an inventory database, an employee database, an equipment operation manual and information database, a sales data database, a customer database, a corporate policy, procedure and / or rules database, an operation procedure database, or any other database or collection of data used when managing and operating POS devices. Database 224 may also include any information useful to employees to assist employees in setting up POS devices effectively, efficiently, and properly. In this regard, this disclosure is a means for employees to have an AI-based system that provides POS configuration and / or setup for employees automatically and quickly with all the training necessary to know how to do it, instead of employees attempting to perform this complex task manually, all of which can be time-consuming and complex for employees.

[0066]

[0068] Database 224 (e.g., MongoDB) may store arbitrary data such as configuration settings for POS device 102, user data, and interaction logs. The backend system provides the AI ​​engine with appropriate context and examples for driving the AI ​​engine. A prompt flow may be used to guide the results to actionable outcomes.

[0067]

[0069] Continuing to refer to Figure 2, the on-premises unit 202 also includes a management component 250, a reporting and dashboard component 252, and a performance evaluation component 254.

[0068]

[0070] In some implementations, the management component 250 includes an intelligent service selector, an AI training component, and an AI model management component. In some examples, the intelligent service selector allows the administrator to select cloud-hosted services to be used by the AI ​​assistant (e.g., sentiment analysis and translation for NLP). For example, it is possible to select the same service for all functions, or to select different cloud-hosted services for each function (e.g., Cortana for NLP, Google ML for sentiment analysis). In some examples, the AI ​​training component runs bulk training data on the AI ​​model of the cloud-hosted service (e.g., the AI ​​model used for sentiment analysis), thereby allowing the administrator to access the API functionality of the cloud-hosted service from a familiar frontend. In some examples, the AI ​​management component allows the administrator to maintain intentions and entities that need to be processed by the cloud-hosted service, thereby allowing the administrator to access the API functionality of the cloud-hosted service from a familiar frontend.

[0069]

[0071] In some implementations, the reporting and dashboard component 252 includes an audit log manager, an audit log repository, a reporting dashboard, and a conversation log repository. In some examples, the audit log manager allows administrators to view all logs generated by various AI assistant components (for example, for operational monitoring and maintenance). In some examples, log messages are stored in the audit log repository. In some examples, the reporting dashboard allows administrators to view all conversations that have taken place between the user and the AI ​​assistant, and can show, for example, successfully completed conversations and discarded conversations.

[0070]

[0072] In some implementations, the performance evaluation component 254 includes an AI accuracy reporting component, an AI testing facility, and an AI model log repository. In some examples, the AI ​​accuracy reporting component measures the overall accuracy of the cloud-hosted service (for example, based on deriving trained intents from provided inputs). In some examples, accuracy reporting relies on an AI model log repository where each request and response made to the cloud-hosted service is recorded. In some examples, an AI testing facility allows administrators to directly interact with the AI ​​models of the cloud-hosted service to test the performance of each AI model (for example, when deriving correct intents based on user inputs).

[0071]

[0073] Figure 3 shows an exemplary layered architecture and environment of a generative artificial intelligence system according to several embodiments. The enterprise generative artificial intelligence system architecture and environment includes a layered structure. More specifically, the layered structure includes an input layer 302, a monitoring layer 310, an agent layer 320, an agent and tool layer 330, a tool and data model layer 350, and an external layer 380. These layers are shown as examples, and it will be understood that other examples may include any number of such layers (e.g., any number of layers 320 and 330).

[0072]

[0074] The input layer 302 represents a layer of the enterprise's generative artificial intelligence system architecture that receives input (e.g., queries, compound inputs, instruction sets, etc.) from users or systems in the retail store (e.g., employees, store workers, authorized company representatives, etc.). For example, an interface module of the enterprise's generative artificial intelligence system can receive input.

[0073]

[0075] The monitoring layer 310 represents a layer of an enterprise generative artificial intelligence system architecture that includes one or more large language models (e.g., orchestrator modules) capable of developing plans to respond to inputs received in the input layer 302. The plans may include a predetermined set of tasks (e.g., search tasks, API call tasks, etc.). In one example, the monitoring layer 310 can provide the pre-processing and post-processing functions described herein, as well as the functions of the orchestrator and understanding modules described herein. The monitoring layer 310 can work in conjunction with one or more of the subsequent layers 320-380 to execute a predetermined set of tasks.

[0074]

[0076] The agent layer 320 represents a layer of the enterprise's generative artificial intelligence system architecture that contains agents capable of performing a given set of tasks. The agent layer 320 includes a machine learning insight agent 322, an information retrieval agent 324, a dashboard agent 326, and an optimizer agent 328. Each of agents 324-328 may include a large-scale language model ("LLM") that provides inference capabilities to accomplish their assigned portion of a given set of tasks. More specifically, agents 324-328 can instruct agents and tools in subsequent layers (e.g., layer 330), of which there may be any number, to perform tasks. For example, the machine learning insight agent 322 may instruct the text processing tool 332 to perform a text processing task (e.g., converting the output of an artificial intelligence application into natural language), the image processing tool 334 to perform an image processing task (e.g., generating a natural language summary of an image output from an artificial intelligence application), the time series tool 336 to obtain summarized time series data (e.g., time series data output from an artificial intelligence application), and the API calling tool 338 to perform an API calling task (e.g., making an API call to trigger or access an artificial intelligence application). Other tasks described above may include tools for configuring the POS, i.e., setting up product information, prices, etc. for the POS, configuring procedures and rules for the POS, enabling training for the POS, or any other data used when configuring and running the POS device 102.

[0075]

[0077] The information retrieval agent 324 may collaborate and / or cooperate with several different agents to perform data retrieval tasks. For example, the information retrieval agent 324 may instruct the unstructured data retriever agent 340 to receive unstructured data records, the structured data retriever agent 342 to retrieve structured data records, and the type system retriever agent 344 to retrieve one or more data models (or subsets of data models) and / or types from the type system. The type system provides compatibility across different data formats, protocols, operating languages, heterogeneous systems, etc. A type can encapsulate some or all of the data formats of the different types or modalities described herein (e.g., multimodal, text, coded, language, statistics, audio, visual, audiovisual, etc.). For example, a data model may contain various different types (e.g., within a tree or graph structure), each of which may describe data fields, behaviors, functions, etc. Each type can represent a different object (e.g., a real-world object) or system (e.g., a computing cluster, a corporate database, a file system, etc.), and each type can include a large language model context that provides a context for the large language model to design or update plans. For example, the context may include a natural language summary or a description of the type (e.g., a description of the represented object, its relationships with other types or objects, associated methods and functions, etc.). Types can be defined in natural language format for efficient processing by the large language model. The type system retriever agent 344 may traverse the data model 354 to retrieve a subset of the data model 354 and / or types of the data model 354. The structured data retriever agent 342 can then use the retrieved information to efficiently retrieve structured data from a structured data source (e.g., a structured data source structured or modeled according to the data model 354).

[0076]

[0078] The dashboard agent 326 may be configured to generate one or more visualizations and / or graphical user interfaces, such as dashboards. For example, the dashboard agent 326 may run tools 352-5 and 352-6 to generate a dashboard based on information retrieved by other agents and / or output by other agents (e.g., natural language summaries of relevant tool outputs).

[0077]

[0079] The optimizer agent 328 may be configured to perform various different prescriptive analysis functions and mathematical optimizations 352-7, and may assist in the computation of answers to various problems. For example, the large language model 306 may use the optimizer agent 328 to generate a plan, determine a given set of tasks, determine whether further information is needed to produce a final result, and so on.

[0078]

[0080] The tool and data model layer 350 is intended to represent a layer of the enterprise's generative artificial intelligence system architecture, including the tool 352 and the data model 354. Agents 340-342 can run the tool 352 to retrieve information from various applications and databases 224 within the external layer 380 (e.g., outside the enterprise's generative artificial intelligence system). The tool 252 may include connectors that can connect to systems and data stores outside the enterprise's generative artificial intelligence system.

[0079]

[0081] Figure 4 shows an example of a process 400 that can be executed in an implementation of the present disclosure. In some examples, the exemplary process 400 is provided using one or more computer executable programs that are executed by one or more computing devices (e.g., the backend system 108 in Figure 1).

[0080]

[0082] In block 402, user 120 initiates a chat session with POS device 102. User 120 needs to authenticate themselves to POS device 102, such as by logging in, so that only authorized users can configure and set up POS device 102.

[0081]

[0083] In block 404, user 120 enters a configuration request into POS device 102 via the chat interface of POS device 102. In block 406, the request is processed and sent to backend server 108. In this regard, POS device 102 communicates with AI engine 214, which helps identify what user 120 has provided as a request, and at that point POS device 102 triggers the sending of the request to backend server 108.

[0082]

[0084] In block 408, the backend server 108 processes the request, enhances the data, and queries the AI ​​engine 214. In this regard, the request is parsed to determine the requested action, the data provided, etc. Next, the backend server 108 determines what additional data is needed based on the request reformatting by the AI ​​engine 214.

[0083]

[0085] In block 410, the AI ​​engine 214 interprets the request and generates a response to the backend server 108. The AI ​​engine 214 determines what the request is about, determines the set of data required to satisfy this request, and then compares the provided data with the required data. If the response from the AI ​​engine 214 contains missing or incorrectly formatted datasets, or if the data is not understood, the AI ​​engine 214 includes an optional request to the user for more information, which the user provides to the user 120 visually via the chat interface of the POS device 102.

[0084]

[0086] In block 410, the AI ​​engine 214 may also query the database 224 for any necessary information. Furthermore, once all the data is available and the AI ​​engine 214 has what it needs, the AI ​​engine 214 notifies the backend server 108 of the instructions it should take.

[0085]

[0087] In block 412, the response (including instructions for the backend server 108) is sent back to the backend server 108, which then verifies the response by stepping through the necessary configuration actions and presents such results to the user 120. In this regard, the backend server 108 then executes the instructions provided to the backend server 108 by the AI ​​engine 214. For example, the backend server 108 may update the database 224, update the software in the POS device 102, or modify the rules / procedures in the POS device 102, the database 224, or the cloud service.

[0086]

[0088] In block 414, the chat interface displays a response or any other visual feedback to user 120.

[0087]

[0089] Figure 5 shows an exemplary process that can be performed according to the implementation of this disclosure. Figure 5 shows a specific example of processing based on Figure 4. In this example, the employee is authenticated to POS device 102.

[0088]

[0090] First, in block 502, the user speaks to the POS device 102 saying, "I want to add a new product." The POS device 102 receives the speech, converts it to text, and then sends the text request to the AI ​​engine 214 via the chat interface of the POS device 102.

[0089]

[0091] In block 504, the AI ​​engine 214 uses LLM to interpret the user's request as the user wanting to update the product database 224. Based on its view of the product database 224, the AI ​​engine 214 determines what data is needed (product name, price, category) and responds with "Understood. Please provide the product name, price, and category" using the procedure described above for Figures 1 to 3.

[0090]

[0092] Next, in block 508, the user tells the POS device 102, "The product name is 'Coffee Mug,' the price is '$12.99,' and the category is 'Beverage Container.'"

[0091]

[0093] In block 510, a similar process to that in 504 takes place, but the AI ​​engine 214 determines whether all the requested data has been provided and formats the data as necessary based on the steps and system described in Figures 1 to 3. The AI ​​engine 214 determines that the data is complete and responds, as shown in block 512, "Okay, I'll add 'Coffee Mug' to the 'Beverage Containers' category for '$12.99'." The AI ​​engine 214 also sends a command to the backend server 108 to update the product database 224 using the procedure described above with respect to Figures 1 to 3.

[0092]

[0094] In block 514, the backend server 108 updates the product database with new product details based on instructions from the AI ​​engine.

[0093]

[0095] In block 516, visual feedback is provided to the user via POS device 102, such as "Product 'Coffee Mug' has been successfully added." In this regard, the system can return and visualize the configuration settings to the user within the chat interface. For example, if the user configures a new product, the chat interface can display product details for confirmation.

[0094]

[0096] The implementations and all functional operations described herein can be realized in digital electronic circuits, computer software, firmware, or hardware, or one or more combinations thereof, including the structures disclosed herein and their structural equivalents. The implementations can be realized as one or more computer program products, i.e., one or more modules of computer program instructions encoded on a computer-readable medium for execution by a data processing device or for controlling the operation of a data processing device. The computer-readable medium may be a machine-readable storage device, a machine-readable storage substrate, a memory device, a composite that yields a machine-readable propagated signal, or one or more combinations thereof. The term “computing system” encompasses all devices and machines for processing data, including, for example, a programmable processor, a computer, or multiple processors or computers. In addition to hardware, a device may include code that creates the execution environment for the computer program in question (e.g., code that constitutes processor firmware, a protocol stack, a database management system, an operating system, or any suitable combination of one or more thereof). A propagated signal is an artificially generated signal (e.g., a machine-generated electrical, optical, or electromagnetic signal) produced to encode information for transmission to a suitable receiving device.

[0095]

[0097] Computer programs (also known as programs, software, software applications, scripts, or code) can be written in any suitable programming language, including compiled or interpreted languages, and may be deployed in any suitable form, including as standalone programs or as modules, components, subroutines, or other units suitable for use in a computing environment. Computer programs do not necessarily correspond to files in a file system. A program may be stored in part of a file that holds other programs or data (e.g., one or more scripts stored in a markup language document), in a single file dedicated to the program in question, or in a group of interconnected files (e.g., a file that stores one or more modules, subprograms, or parts of code). Computer programs may be deployed to run on one computer, or on multiple computers located in one site, or distributed across multiple sites and interconnected by a communication network.

[0096]

[0098] The processes and logic flows described herein may be executed by one or more programmable processors that run one or more computer programs to perform a function by manipulating input data and generating outputs. The processes and logic flows may also be executed by dedicated logic circuits (e.g., FPGAs (field programmable gate arrays) or ASICs (application-specific integrated circuits)), and the devices may also be implemented as dedicated logic circuits.

[0097]

[0099] Processors suitable for executing computer programs include, for example, both general-purpose and dedicated microprocessors, as well as any one or more processors of any suitable type of digital computer. Generally, a processor receives instructions and data from read-only memory or random-access memory or both. The elements of a computer may include a processor for executing instructions and one or more memory devices for storing instructions and data. Generally, a computer also includes one or more mass storage devices (e.g., magnetic, magneto-optical disks, or optical disks) for storing data, or is operablely coupled to receive data from there, transfer data to there, or both. However, a computer does not need to have such devices. Also, a computer may be built into other devices (e.g., mobile phones, personal digital assistants (PDAs), mobile audio players, Global Positioning System (GPS) receivers). Computer-readable media suitable for storing computer program instructions and data include all forms of non-volatile memory, media, and memory devices, including, for example, semiconductor memory devices (e.g., EPROM, EEPROM, and flash memory devices), magnetic disks (e.g., internal hard disks or removable disks), magneto-optical disks, and CD-ROM and DVD-ROM disks. Processors and memory may be complemented by or incorporated into dedicated logic circuits.

[0098]

[0100] The implementation may be carried out on a computer having a display device for displaying information to the user (e.g., a CRT (cathode ray tube), an LCD (liquid crystal display) monitor) and a keyboard and pointing device (e.g., a mouse, trackball, touchpad) that allows the user to provide input to the computer, in order to provide interaction with the user. Interaction with the user may also be provided using other types of devices, for example, the feedback provided to the user may be any appropriate form of sensory feedback (e.g., visual feedback, auditory feedback, haptic feedback), and the input from the user may be received in any appropriate form, including acoustic, voice, or haptic input.

[0099]

[0101] The implementation may be implemented in a computing system that includes a backend component (e.g., as a data server), a middleware component (e.g., an application server), and / or a frontend component (e.g., a client computer having a graphical user interface or a web browser that allows a user to interact with the implementation), or any suitable combination of one or more such backend, middleware, or frontend components. The components of the system may be interconnected by digital data communication (e.g., a communication network) in any suitable form or medium. Examples of communication networks include local area networks ("LANs") and wide area networks ("WANs"), such as the Internet.

[0100]

[0102] A computing system may include clients and servers. Clients and servers are generally remote from each other and typically interact via a communication network. The client-server relationship arises from computer programs running on each computer that have a client-server relationship with each other.

[0101]

[0103] This specification contains many details, which should not be construed as limitations on the scope of this disclosure or what can be claimed, but rather as descriptions of features specific to particular implementations. Certain features described herein in the context of separate implementations can also be implemented in combination in a single implementation. Conversely, various features described in the context of a single implementation may also be implemented separately or in any suitable subcombination in multiple implementations. Furthermore, features are described above as acting in a particular combination and may be initially claimed as such, but one or more features from a claimed combination may, in some cases, be removed from the combination, and the claimed combination may be directed to a partial combination or a variation of a partial combination.

[0102]

[0104] Similarly, while the operations are shown in a specific order in the diagrams, this should not be understood as requiring that such operations be performed in the specific order shown, or in a sequential order, or that all shown operations be performed, in order to achieve the desired result. In certain situations, multitasking and parallel processing may be advantageous. Furthermore, the separation of various system components in the implementation forms described above should not be understood as requiring such separation in all implementation forms, and it should be understood that the described program components and systems may generally be integrated together in a single software product or packaged in multiple software products.

[0103]

[0105] Several implementations have been described. Nevertheless, it will be understood that various modifications can be made without departing from the spirit and scope of this disclosure. For example, various forms of the flow shown above may be used, and steps may be rearranged, added, or removed. Thus, other implementations are within the scope of the following claims.

[0104]

[0106] Unless the context clearly indicates otherwise, words such as “including,” “including,” and so on should be interpreted comprehensively throughout the specification and claims, as opposed to exclusive or exhaustive meanings; that is, “including, but not limited to.” Where used herein, the terms “connected,” “joined,” or any variation thereof mean any direct or indirect connection or joining between two or more elements, and the joining of connections between elements may be physical, logical, or a combination thereof. Furthermore, where used in this application, the words “here,” “above,” “below,” and words with similar meanings refer to the entire application and not to any particular part thereof. Where the context allows, the words in the above detailed descriptions, which use singular or plural forms, may also include plural or singular, respectively. The word “or” relating to a list of two or more items encompasses all of the following interpretations of the word, namely any of the items in the list, all of the items in the list, and any combination of the items in the list.

[0105]

[0107] The above detailed description of embodiments of the Disclosure is not intended to be exhaustive or to limit the Disclosure to the exact forms disclosed above. Specific embodiments and examples of the Disclosure are described above for illustrative purposes, but as those skilled in the art will recognize, various equivalent modifications are possible within the scope of the Disclosure. For example, while processes or blocks are presented in a given order, alternative embodiments may execute routines having steps in a different order, or use systems having blocks, and some processes or blocks may be deleted, moved, added, subdivided, combined, and / or modified to provide alternative or partial combinations. Each of these processes or blocks may be implemented in various different ways. Also, while processes or blocks may be shown as being executed in series, these processes or blocks may instead be executed in parallel or at different times. Furthermore, the specific numbers described herein are merely examples, and alternative implementations may use different values ​​or ranges.

[0106]

[0108] The teachings of this disclosure provided herein may be applied to systems other than those described above. Further embodiments can be provided by combining elements and operations of the various embodiments described above.

[0107]

[0109] Any of the above-mentioned patents and applications and other references, including those that may be listed in the attached application documents, are incorporated herein by reference. Aspects of this disclosure may be modified, as needed, to use the systems, functions, and concepts of the various references described above to provide yet another embodiment of this disclosure.

[0108]

[0110] In light of the above detailed description, these and other modifications may be made to this disclosure. While the above description describes specific embodiments of this disclosure and the intended best form, the disclosure can be implemented in many ways, no matter how detailed the above description may be. The details of the system may vary considerably in the details of its implementation, but are still encompassed within the disclosure disclosed herein. As stated above, any specific terms used when describing a particular feature or aspect of this disclosure should not be construed as meaning that the terms are redefined herein to be limited to any particular characteristic, feature, or aspect of the disclosure in which they relate. In general, the terms used in the following claims should not be construed as limiting this disclosure to the specific embodiments disclosed herein unless such terms are explicitly defined in the above detailed description section. Thus, the actual scope of this disclosure includes not only the disclosed embodiments but also all equivalent ways of implementing or carrying out the disclosure under the claims.

[0109]

[0111] Certain aspects of this disclosure are presented below in the form of specific claims, but the inventors have envisioned various aspects of this disclosure in the form of any number of claims. For example, only one aspect of this disclosure may be described as a means-plus-function claim under 35 U.S.C.sec., and other aspects may likewise be described as means-plus-function claims or in other forms such as those implemented in computer-readable media, as stipulated in 112(f). (Claims intended to be dealt with under 35 U.S.C. § 112(f) begin with the word “means.”) Accordingly, the inventors reserve the right to add additional claims after filing in order to pursue such additional claim forms for other aspects of this disclosure. [Explanation of Symbols]

[0110] 100 High-Level Architectures 102 POS devices 108 Backend systems, backend servers 110 Backend Systems 112 Network 114 Server Systems 116 Datastores, Databases 120 users 200 Architectures 202 On-Premise Department 204 Cloud Department 206 channels 208 Channel Connector, Connector 210 Data Integration Layer 212 Authentication Middleware 214 AI middleware, AI engines 216 Connector 218 Web service launched 220 Backend Source Systems 222 Data Management Layer 224 Databases 230 Session Managers 232 Confidential Information Masking Components 234 Action Handler 236 Communications Orchestrator 238 Exception handlers 240 Response Generator 242 Data Validation Module 250 Management Components 252 Reporting and Dashboard Components, Tools 254 Performance Evaluation Components 302 Input Layer 306 Large-scale language models 310 Monitoring layer 320 agent layers 322 Machine Learning Insight Agent 324 Information Retrieval Agent 326 Dashboard Agent 328 Optimizer Agent 330 Agent and Tool Layers 332 Text Processing Tools 334 Image Processing Tools 336 Time Series Tools 338 API Tools 340 Unstructured Data Retriever Agents 342 Structured Data Retriever Agent 344 Type System Retriever Agent 350 Tools and Data Model Layers 352 Tools 352-5 Tool 352-6 Tool 352-7 Numerical Optimization 354 Data Models 380 Outer Layer 400 processes

Claims

1. A method for supporting company employees with an AI engine, The processor of the generation AI-based system receives requests from the employee via the chat interface of the POS device, either by voice or text, to configure and / or set up the POS device. The processor uses the data included in the request to determine at least one action to be performed by at least one backend source system of the company to configure and / or set up the POS device. The processor generates a response for the at least one backend source system to execute when configuring and / or setting up the POS device, The processor transmits the response to the at least one backend source system so that the at least one backend source system directly configures and / or sets up the POS device. Methods that include...

2. If the dataset to be sent as the response contains confidential information, and the response is to be sent over a network, the confidential information in the dataset is to be masked before the response is sent. The method according to claim 1, further comprising:

3. Before sending the response, the sensitive information in the dataset is masked using an on-premises masking component. The method according to claim 2.

4. The method according to claim 1, wherein the requirements are for configuring one of the following: the POS device itself, product information, inventory, corporate policies, and operating procedures.

5. The method according to claim 1, wherein the requirement is that the employee requests the configuration and / or setup of the POS device before the POS device is put into operation at the company.

6. The method of claim 5, further comprising providing AI engine training data for training the AI ​​engine when configuring and / or setting up a system for the POS device and a system surrounding the POS device.

7. The method according to claim 1, wherein the generative AI-based system includes an intent classification model using natural language processing (NLP) and provides an intent set including one or more intents specified in the request.

8. The method according to claim 1, wherein the actions for configuring the POS device include at least one of setting up a payment method, adding a new product, configuring tax rates, business rules, receipt configuration, and managing user roles.

9. A generative AI-based system equipped with a processor, The aforementioned processor, Receiving requests from employees via the POS device, either by voice or text, to configure and / or set up the POS device, Using the data included in the request, determine at least one action to be performed by at least one backend source system of the company to configure and / or set up the POS device; To generate a response for the at least one backend source system to execute when configuring and / or setting up the POS device, Sending the response to the at least one backend source system so that the at least one backend source system configures and / or sets up the POS device, A system configured to perform the following actions.

10. The aforementioned processor, If the dataset to be sent as the response contains confidential information, and the response is to be sent over a network, the confidential information in the dataset is to be masked before the response is sent. The system according to claim 9, further configured to perform the following:

11. Before sending the response, the sensitive information in the dataset is masked using an on-premises masking component. The system according to claim 10.

12. The system according to claim 9, wherein the actions for configuring the POS device include at least one of setting up a payment method, adding a new product, configuring tax rates, business rules, receipt configuration, and managing user roles.

13. The system according to claim 9, wherein the requirement is that the employee requests the configuration and / or setup of the POS device before the POS device is put into operation at the company.

14. The system according to claim 9, further comprising providing AI engine training data for training the AI ​​engine when configuring and / or setting up the system for the POS device and the system surrounding the POS device.

15. The system according to claim 9, wherein the generative AI-based system includes an intent classification model using natural language processing (NLP) and provides an intent set including one or more intents specified in the request.

16. The system according to claim 9, wherein the POS device includes a self-checkout system configured for a customer to self-scan and identify products and pay for the products.

17. The system according to claim 9, wherein the request is from an employee authenticated as an administrator in the system.

18. A non-temporary computer-readable medium, when executed by the processor of a generative AI-based system, Receiving requests from employees via the POS device, either by voice or text, to configure and / or set up the POS device, Using the data included in the request, determine at least one action to be performed by at least one backend source system of the company to configure and / or set up the POS device; To generate a response for the at least one backend source system to execute when configuring and / or setting up the POS device, Sending the response to the at least one backend source system so that the at least one backend source system configures and / or sets up the POS device, A non-temporary computer-readable medium that performs a method including [this].

Images