Code modification using a virtual assistant

Abstract

A system can receive an input via a chatbot interface. The input can be indicative of an update to a control parameter. The system can generate, using a large language model, a search query based on the input. The system can filter, using the search query, data related to control parameters. The system can detect, using the large language model, updates to the control parameters based on the filtered data. The system can identify, based on the updates to the control parameters and product metadata, products that fail to comply with the updates. The system can determine code segments within a source code corresponding to the products. The code segments can be related to the control parameters. The system can modify the code segments to comply with the updates to the control parameters. The system can execute the modified code segments to perform an operation.

Description

CODE MODIFICATION USING A VIRTUAL ASSISTANTCROSS-REFERENCE TO RELATED APPLICATIONS

[0001] This application claims the benefit of, and priority to, Indian Provisional Patent Application No. 202311075603, filed November 6, 2023, the entirety of which is hereby incorporated by reference herein.TECHNICAL FIELD

[0002] This application is generally related to improvements to computing technology, particularly natural language processing, and, more particularly, to code modification using a virtual assistant based on machine learning models and control parameters.BACKGROUND

[0003] Natural language processing can be used for efficient information storage, retrieval, and analysis across various fields. For example, natural language processing systems are used to process user prompts and provide contextually relevant responses. Natural language processing operations can be executed to improve query resolution and information management. However, due to the variety of types of constraints, frequent changes to constraints, and the complexity of applications configured to execute operations in accordance with such constraints, it is challenging to efficiently, reliably, or accurately detect such changes to constraints, let alone update application code to execute operations in accordance to updated constraints.SUMMARY

[0004] Aspects of the technical solutions described herein can be directed to a computing architecture configured for managing compliance operations across data integration, automated monitoring, and system scalability. In this regard, challenges arise in continuously monitoring compliance status and maintaining data interoperability between various regulatory databases and formats, particularly when dealing with differing data standards, protocols, and control parameters.Additionally, there are several technical complexities, such as coordinating different system components, including compliance monitoring systems, data security frameworks, and regulatory intelligence tools, to facilitate seamless operation and data exchange across interconnected systems while maintaining scalability and flexibility to adapt to evolving control parameters.

[0005] The technical solutions described herein address such technical challenges by implementing a computing architecture configured to manage compliance operations with precision and efficiency. By employing a multi-stage processing pipeline, the computing architecture improves the accuracy of compliance assessments. The computing architecture can integrate data from multiple sources to facilitate interoperability across various regulatory databases and formats, even when standards and protocols differ. When a user submits a query related to a regulatory change, the computing architecture can leverage natural language processing techniques to extract core intent and keywords. Based on the core intent and keywords extracted from the user query, the large language model can generate a search query to retrieve relevant information from a knowledge base or regulatory database, thereby facilitating accurate identification of specific control parameters. The computing architecture can utilize Al and machine learning algorithms to evaluate the data and generate responses that include explanations, actionable recommendations, or relevant code modifications aligned with compliance requirements.

[0006] Additionally, the technical solutions include automated code modification processes to maintain ongoing compliance with evolving regulations, achieving scalability and flexibility. By leveraging machine learning models trained on datasets of code, control parameters, and regulatory requirements, the computing architecture accurately identifies which code segments in the source code are to be updated based on changes to the control parameters. The computing architecture can generate targeted modifications, such as adjusting configuration settings to align with compliance standards. The technical solutions implement a security framework to protect sensitive data in compliance with international standards.[0007| Furthermore, the technical solutions support localization and customization based on specific regulatory requirements in different regions, including language support and customization of workflows to address local regulatory nuances. The inclusion of Al-powered regulatory tools can provide proactive compliance management by monitoring regulatory changes, identifying potential risks, and generating compliance reports. Additionally, automated testing and validation processes can confirm the accuracy and compliance of the modified code, reducing risk and accelerating time-to-market for compliant products and features.

[0008] An aspect of the technical solutions described herein is directed to a system. The system includes one or more processors coupled with memory. The system can receive, from a client device, a textual input via a chatbot interface. The textual input can be indicative of an update to a control parameter. The system can generate, using a large language model, a search query based on the textual input. The system can filter, using the search query, data related to control parameters stored in a database. The system can detect, using the large language model, updates to the control parameters based on the filtered data. The system can identify, based on the updates to the control parameters and product metadata, one or more products that fail to comply with the updates. The system can determine code segments within a source code corresponding to the one or more products. The code segments can be related to the control parameters. The system can modify the code segments to comply with the updates to the control parameters. The system can execute the modified code segments to perform an operation.|0009| The control parameters can correspond to a set of rules defining operational standards for the one or more products. The system can input the textual input into the large language model to cause the large language model to classify the textual input into one or more search terms. The system can generate, using the large language model, the search query based on the one or more search terms. The large language model can be trained on a dataset including at least one of regulatory requirements, industry standards, or legislative directives. The system can identify, responsive to comparing the product metadata of the one or more products to compliance thresholds defined by the updates to the control parameters, that the one or more products fail to comply with the updates to the control parameters. The product metadata caninclude information related to product configurations, operational settings, usage patterns, or deployment environments. The system can determine the code segments within the source code responsive to mapping the control parameters to the code segments implementing functionalities related to the control parameters. The system can execute the large language model, trained on a dataset of control parameters, to determine the code segments within the source code. The system can generate code modifications to the code segments based on the updates to the control parameters. The system can execute the generated code modifications to comply with the updates to the control parameters. The system can input the modified code segments and the updates to the control parameters into the large language model to cause the large language model to generate a data structure. The system can parse the data structure to display, via the client device, an electronic report indicating compliance of the one or more products with the updates to the control parameters.

[0010] An aspect of the technical solutions described herein is directed to a method. The method can include generating, using a large language model, a search query based on an input received via a chatbot interface indicative of an update to a control parameter. The method can include filtering, using the search query, data related to control parameters stored in a database. The method can include detecting, using the large language model, updates to the control parameters based on the filtered data. The method can include identifying, based on the updates to the control parameters and product metadata, one or more products that fail to comply with the updates. The method can include determining code segments within a source code corresponding to the one or more products. The code segments can be related to the control parameters. The method can include modifying the code segments to comply with the updates to the control parameters. The method can include executing the modified code segments to perform an operation.

[0011] The control parameters can correspond to a set of rules defining operational standards for the one or more products. The method can include inputting the input into the large language model to cause the large language model to classify the input into one or more search terms. The method can include generating, using the large language model, the search query basedon the one or more search terms. The large language model can be trained on a dataset including at least one of regulatory requirements, industry standards, or legislative directives. The method can include identifying, responsive to comparing the product metadata of the one or more products to compliance thresholds defined by the updates to the control parameters, that the one or more products fail to comply with the updates to the control parameters. The product metadata can include information related to product configurations, operational settings, usage patterns, or deployment environments. The method can include determining the code segments within the source code responsive to mapping the control parameters to the code segments implementing functionalities related to the control parameters. The method can include executing the large language model, trained on a dataset of control parameters, to determine the code segments within the source code. The method can include generating code modifications to the code segments based on the updates to the control parameters. The method can include executing the generated code modifications to comply with the updates to the control parameters.10012] An aspect of the technical solutions described herein is directed to a non-transitory computer readable medium, including one or more instructions stored thereon and executable by a processor. The processor can filter, using a large language model, data related to control parameters used by one or more products, the control parameters indicated via a chatbot interface. The processor can detect, using the large language model, updates to the control parameters based on the filtered data. The processor can identify, based on the updates to the control parameters and product metadata, that the one or more products fail to comply with the updates to the control parameters. The processor can determine code segments within a source code corresponding to the one or more products, the code segments related to the control parameters. The processor can modify the code segments to comply with the updates to the control parameters.BRIEF DESCRIPTION OF THE DRAWINGS

[0013] These and other aspects and features of the present implementations are depicted by way of example in the figures discussed herein. Present implementations can be directed to, but are not limited to, examples depicted in the figures discussed herein. Thus, this disclosure isnot limited to any figure or portion thereof depicted or referenced herein, or any aspect described herein with respect to any figures depicted or referenced herein.

[0014] FIG. 1 depicts an example system, in accordance with some implementations.

[0015] FIG. 2 depicts an example method of code modification for regulatory compliance using a virtual assistant based on machine learning models and control parameters, in accordance with some implementations.10016] FIG. 3 depicts an example compliance monitoring system, in accordance with some implementations.|0017| FIG. 4 depicts an example legislation bank architecture, in accordance with some implementations.

[0018] FIG. 5 depicts an example authority connector, in accordance with some implementations.

[0019] FIG. 6 depicts an example compliance meter, in accordance with some implementations.

[0020] FIG. 7 depicts an example LM chat, in accordance with some implementations.

[0021] FIG. 8 depicts an example cloud-based language platform, in accordance with some implementations.

[0022] FIGS. 9-10 depict an example system, in accordance with some implementations.

[0023] FIG. 11 depicts an example operational system, in accordance with some implementations.10024] FIG. 12 depicts a block diagram of an example computing system for implementing the embodiments of the present solution, including, for example, the systems depicted in FIGS. 1 and 9-11, and the method depicted in FIG. 2.DETAILED DESCRIPTION

[0025] Aspects of the technical solutions are described herein with reference to the figures, which are illustrative examples of this technical solution. The figures and examples below are not meant to limit the scope of the technical solutions to the present implementations or to a single implementation. Several other implementations in accordance with present implementations are possible, for example, by way of interchange of some or all of the described or illustrated elements. Where certain elements of the present implementations can be partially or fully implemented using known components, only those portions of such known components that are necessary for an understanding of the present implementations are described, and detailed descriptions of other portions of such known components are omitted to not obscure the present implementations. Terms in the specification and claims are to be ascribed no uncommon or special meaning unless explicitly set forth herein. Further, the technical solutions and the present implementations encompass present and future known equivalents to the known components referred to herein by way of description, illustration, or example.

[0026] The technical solutions described herein implement a multi-stage natural language processing pipeline to facilitate compliance management through a chatbot interface. When a user provides an input indicative of an update to a control parameter, the large language model (LLM) can generate a search query based on the input, allowing for targeted data retrieval from a database of control parameters. This query output can be used to filter stored data, identifying relevant control parameters associated with regulatory or compliance updates. Based on the filtered data, the system can execute the LLM to detect specific updates to control parameters, allowing the identification of products that fail to meet the updated compliance standards by comparing against product metadata. The system can identify corresponding code segments within the source code related to these products and modify the code segments to align with the updated compliance requirements. Once modified, the system can execute these code segments to apply the compliance updates across affected products, supporting efficient compliance management, accurate code adaptation, and real-time or near real-time updates to facilitate ongoing regulatory alignment. Thecomputing architecture, thus, enhances the accuracy of compliance monitoring and provides dynamic code modification to stay current with evolving regulatory standards.

[0027] FIG. 1 depicts an example system 100 according to one or more aspects of the technical solutions described herein. As illustrated by way of example in FIG. 1, a system 100 can include one or more of a data processing system 105, a client system 115, and a machine learning system 120. One or more components of the system 100 can communicate via network 110.100281 The data processing system 105 can include a physical computer system operatively coupled or couplable with one or more components of the system 100. The data processing system 105 can include, host, or be hosted by or on a cloud system, a server, a distributed remote system, or any combination thereof. The data processing system 105 can include a virtual computing system, an operating system, and a communication bus to effect communication and processing. The data processing system 105 can include physical infrastructure, such as physical servers, storage devices, and network equipment housed in data centers. The data processing system 105 can include a virtual computing system, which can include cloud-based virtual machines or containers for running applications and services. The data processing system 105 can include an operating system that can function as the core manager, allocating resources, configuring processes, and maintaining seamless interaction between hardware and applications. The data processing system 105 can include a communication bus that can facilitate communication between different components within the system. The data processing system 105 can be configured to connect with external systems to allow for data exchange and service delivery to end users.

[0029] The network 110 can include any type or form of network. The geographical scope of the network 110 can vary widely and the network 110 can include a body area network (BAN), a personal area network (PAN), a local-area network (LAN), e.g., Intranet, a metropolitan area network (MAN), a wide area network (WAN), or the Internet. The topology of the network 110 can be of any form and can include, e.g., any of the following: point-to-point, bus, star, ring, mesh, or tree. The network 110 can include an overlay network which is virtual and sits on top of one ormore layers of other networks 110. The network 110 can be of any such network topology as known to those ordinarily skilled in the art capable of supporting the operations described herein. For example, the network 110 can be any form of computer network that can relay information between the data processing system 105, the client system 115, and the machine learning system 120. The network 110 can utilize different techniques and layers or stacks of protocols, including, e.g., the Ethernet protocol, the Internet protocol suite (TCP or IP), the ATM (Asynchronous Transfer Mode) technique, the SONET (Synchronous Optical Networking) protocol, or the SD (Synchronous Digital Hierarchy) protocol. The TCP or IP Internet protocol suite can include application layer, transport layer, Internet layer (including, e.g., IPv6), or the link layer. The network 110 can include a type of a broadcast network, a telecommunications network, a data communication network, or a computer network.

[9030] The client system 115 (also referred to herein as a client device 115) can include a computing system that can be used to access the functionality of the data processing system 105. The client system 115 can include a smart phone, mobile device, laptop computer, desktop computer, one or more servers, or any other type of computing device. The client system 115 can include at least one processor and a memory, e.g., a processing circuit. The memory can store processor-executable instructions that, when executed by the processor, cause the processor to perform one or more of the operations described herein. The processor can include a microprocessor, an ASIC, an FPGA, etc., or combinations thereof. The memory can include, but is not limited to, electronic, optical, magnetic, or any other storage or transmission device capable of providing the processor with program instructions. The memory can further include a floppy disk, CD-ROM, DVD, magnetic disk, memory chip, ASIC, FPGA, ROM, RAM, EEPROM, EPROM, flash memory, optical media, or any other suitable memory from which the processor can read instructions. The instructions can include code from any suitable computer programming language.[00311 The client system 115 can include one or more devices to receive input from a user or to provide output to a user. For example, the output capabilities of the client system 115 can be presented through a display device that provides visual feedback to the user. The display devicecan enhance the user experience with electronic displays, such as liquid crystal displays (LCD), light-emitting diode (LED) displays, or organic light-emitting diode (OLED) displays. The electronic displays can implement interactive features, including capacitive or resistive touch input, allowing for multi-touch functionality. The input functionalities can include a keyboard, mouse, or an integrated touch-sensitive panel on the display device, but are not limited thereto.|0032| Each client device 115 can be associated with an identifier used to identify devices or user profiles operating the client devices 115. The identifier can be of one or more forms, such as a device ID, which can be a code assigned to the client device 115 by the manufacturer or operating system, a MAC address, which can be a hardware address assigned to the client device’s network interface, or an IP address, which can identify the client device 115 on a network. The identifier can be a user ID associated with the user profile operating the client device 115, or a session ID, which can be a temporary identifier assigned to a specific session. Other identifiers, such as a serial number, can be used depending on the system and device configuration.(0033] The client system 1 15 can execute an application that communicates with the data processing system 105. The application can present one or more chatbot interfaces 175. The chatbot interface 175 can include a set of rules or protocols that allow different software programs or systems to communicate with each other. The chatbot interface 175 can provide user interfaces to facilitate interaction. For example, the chatbot interface 175 can support text-based interactions or voice-based interactions. Users can input information, view content, or initiate actions through the chatbot interface 175. In some implementations, the chatbot interface 175 can be associated with a particular client application that communicates with the data processing system 105 to process user prompts. The client application can include an application executing on each client system 115. The client application can include a web application, a server application, a resource, a desktop, or a fde. In some embodiments, the client application can include a local application (e g., local to a client system 115), a hosted application, a software-as-a-service (SaaS) application, a virtual application, a mobile application, and other forms of content. In some embodiments, the client application can include or correspond to applications provided by remote servers or third- party servers.[0034| The client system 115 can parse the data structure generated by the large language model 185 to present, via the chatbot interface 175, one or more electronic reports, such as compliance assessment reports, payroll summaries, employee earnings statements, or tax forms. The client system 115 can evaluate the structure of data, such as the format (e.g., JSON, XML, or CSV) and the organization of elements (e.g., classes, properties, products, or individuals). The client system 115 can extract relevant information from the data structure based on indicators that correspond to content objects or data elements, which the client system 115 displays through the chatbot interface 175. In some embodiments, the client system 115 can extract specific details from the data structure, such as product IDs, updated control parameters, modified code segments, compliance status, reasons for compliance status, timestamps, and other relevant metadata.

[0035] The chatbot interface 175 can provide customizable displays that adapt to different types of applications and devices. The chatbot interface 175 can be integrated into various environments. For example, the chatbot interface 175 can be embedded in a web application, allowing users to view electronic reports through a browser interface. The chatbot interface 175 can be integrated into a desktop application, displaying electronic reports locally on the client system 115. The chatbot interface 175 can be deployed within a mobile application, presenting electronic reports in a format suitable for smaller screens and touch interactions. The chatbot interface 175 can operate within a software-as-a-service (SaaS) platform, where electronic reports are hosted remotely but accessed and displayed through the client system 115. The chatbot interface 175 can integrate with third-party applications or systems, allowing electronic reports to be displayed from external sources (e.g., human resources management systems).

[0036] The client system 115 can include, interface with, communicate with, or otherwise utilize a client communicator 180. The client communicator 180 within the client system 115 can be similar to, and include any of the structure and functionality of, the interface controller 135 described in connection with the data processing system 105. For example, the client communicator 180 within the client system 115 can communicate with the data processing system 105 or the machine learning system 120 via the network 110 using one or more communication interfaces to carry out the various operations described herein. The client communicator 180 canbe compatible with particular content delivery systems corresponding to particular content objects, structures of data, types of data, or any combination thereof.

[0037] The data processing system 105 can include, interface with, communicate with, or otherwise utilize a database 125. The database 125 can be a computer-readable memory that can store or maintain any of the information described herein. The database 125 can store data associated with the system 100. The database 125 can include one or more hardware memory devices to store binary data, digital data, or the like. The database 125 can include one or more electrical components, electronic components, programmable electronic components, reprogrammable electronic components, integrated circuits, semiconductor devices, flip flops, arithmetic units, or the like. The database 125 can include at least one of a non-volatile memory device, a solid-state memory device, a flash memory device, or a NAND memory device. The database 125 can include one or more addressable memory regions disposed on one or more physical memory arrays. A physical memory array can include a NAND gate array disposed on, for example, at least one of a particular semiconductor device, an integrated circuit device, or a printed circuit board device. In an aspect, the database 125 can correspond to a non-transitory computer readable medium. In an aspect, the non-transitory computer readable medium can include one or more instructions executable by a system processor 130.

[0038] The database 125 can store or maintain one or more data structures, which can include containers, indices, or otherwise store each of the values, pluralities, sets, variables, vectors, numbers, or thresholds described herein. The database 125 can be accessed using one or more memory addresses, index values, or identifiers of any item, structure, or region maintained in the database 125. The database 125 can be accessed by the components of the data processing system 105, the client system 115, the machine learning system 120, or any other computing device described herein, via the network 110. The database 125 can be internal to the data processing system 105. The database 125 can exist external to the data processing system 105 and can be accessed via the network 110. For example, the database 125 can be distributed across many different computer systems (e.g., a cloud computing system) or storage elements and can be accessed via the network 110 or a suitable computer bus interface.[0039| The database 125 can store or maintain various products 165, each configured to address specific operational functions within a computing infrastructure. The products 165 can include payment scheduling tools that manage payment structures across multiple regions, accommodating timelines or regulatory requirements for each jurisdiction. The products 165 can include human capital management (HCM) solutions that support employee management, spanning recruitment, onboarding, workforce analytics, or compliance tracking. The product 165 can provide payroll processing capabilities to accurately execute payments in adherence to country-specific payroll regulations, wage policies, or reporting standards. The products 165 can incorporate benefits administration tools to manage benefits packages in alignment with regional standards and optimize tax-related deductions for accurate compliance reporting. The products 165 can include tax compliance solutions that assist in the accurate calculation, deduction, and remittance of taxes according to region-specific tax laws, supporting the maintenance of financial integrity and regulatory adherence across computing environments. The aforementioned list of products 165 is non-exhausting and can include other software solutions tailored to specific operational needs or regulatory compliance requirements.

[0040] Each product 165 can be modeled by a data structure, such as a relational table or a NoSQL document, that can organize product-related attributes with specific datatypes, including strings, integers, dates, and Booleans. For example, a product data structure can include attributes such as product name (string), product ID (key), version (string), release date (date), supported languages (string array), and compliance standards (string array). The products 165 can be associated with other data structures through identifiers, such as product ID, allowing the database 125 to associate products 165 with relevant information, including client configurations, compliance records, system updates, employee training records, software licenses, hardware inventory, or regulatory certifications. The attributes can facilitate efficiently query and retrieve product data for analysis, reporting, and decision-making across human resources (HR), compliance, or other relevant domains.

[0041] Each product 165 can be associated with product metadata 166, which can provide information about the product’s attributes, characteristics, or relevant context. The productmetadata 166 can include details across various aspects, such as product identification (identifiers such as product codes, SKUs, or internal reference numbers) or product descriptions (detailed descriptions of the product’s features, functionality, or benefits). The product metadata 166 can specify product categories (categorizing the product into specific product categories or hierarchies) or product specifications, including technical details such as dimensions, weight, materials, or compatibility. In some embodiments, pricing information can be included in the product metadata166 to capture base price, discounts, or special offers. The product metadata 166 can include inventory information, such as stock levels, reorder points, or inventory management policies. The product metadata 166 can further include compliance information, listing relevant certifications, regulatory approvals, or safety standards, along with marketing information that provides marketing materials, promotional campaigns, or target audience segments. In some embodiments, the product metadata 166 can further include information related to product configurations (e.g., specific configurations or variations of the product), operational settings (e.g., details on setup, adjustments, or usage), usage patterns (e.g., insights into common use cases, frequencies, or behaviors), or deployment environments (e.g., target platforms, compatibility requirements, or deployment strategies). In some embodiments, the product metadata 166 can include product lifecycle information that provides context for the product’s various stages, from introduction to end-of-life, for example.100421 The database 125 can store or maintain control parameters 167, which can be specific settings or configurations that govern the operation of the products 165. The control parameters 167 can be defined based on regulatory requirements, industry standards, or legislative directives. The control parameters 167 can be used to manage various aspects of human resources, payroll, or compliance functions. In the context of regulatory requirements, the control parameters167 can be used to manage tax withholding, overtime calculation, benefits administration, leave management, wage garnishments, Equal Employment Opportunity (EEO) reporting, data privacy, or payroll reporting. In the context of industry standards, the control parameters 167 can be used for job classification and coding, establishing salary bands and grades, defining performance metrics and Key Performance Indicators (KPIs), implementing competency frameworks, settingtraining and certification requirements, defining time and attendance policies, maintaining data privacy and security, setting health and safety parameters, defining diversity and inclusion metrics, or structuring employee benefits packages. In the context of legislative directives, the control parameters 167 can be used to facilitate compliance with labor and employment laws such as the Fair Labor Standards Act (FLSA), Family and Medical Leave Act (FMLA), Affordable Care Act (ACA), General Data Protection Regulation (GDPR), EEO regulations, Occupational Safety and Health Administration (OSHA) standards, Health Insurance Portability and Accountability Act (HIPAA), wage garnishment laws, the Immigration and Nationality Act (INA), or state-specific employment laws. The regulatory requirements or the control parameters 167 can be of one or more jurisdictions (countries, states, counties, cities, unions, etc.).[0043 | The database 125 can store or maintain source code 168 corresponding to one or more products 165. The source code 168 can include instructions written in a programming language that define the functions or operations associated with a software product. The source code 168 can be used for maintaining product compliance by facilitating development, maintenance, or enhancement to satisfy control parameters 167. The source code 168 can be organized into several code segments 169, such as header comments (providing an overview of the purpose, creation date, or compliance requirements), import statements (listing libraries or dependencies), global variables and constants (defining accessible values across the code), function definitions (blocks for specific tasks), class definitions (encapsulating data or functions), main execution block (where main logic runs), comments and documentation (improving readability of compliance-related code), or error handling (managing exceptions or logging compliance-related errors).

[0044] The code segments 169 within products 165 can be directed to compliance-related functionalities, including payroll processing, benefits administration, time and attendance, performance management, or compliance and reporting, among others. For example, payroll processing can include algorithms for accurate salary calculations or managing payroll cycles in line with federal, state, and local tax laws. Benefits administration can include algorithms for enrollment processes and eligibility management to maintain adherence to legal requirements orrelevant regulations. Time and attendance segments can track attendance metrics and leave requests to maintain compliance with labor laws. Performance management can include algorithms for goal setting and appraisals to satisfy internal policies or industry best practices. Compliance and reporting segments can generate reports for regulations such as EEO, OSHA, and GDPR, as well as other industry-specific compliance requirements. The code segments 169 can facilitate the automation and management of compliance across various products 165.

[0045] The database 125 can store or maintain a training dataset 170 used to train one or more machine learning models 185. The training dataset 170 can include regulatory requirements, industry standards, or legislative directives to support compliance-related model training. The training dataset 170 can undergo preprocessing, such as cleaning, normalization, and tokenization, to prepare the data for learning. For supervised learning, the training dataset 170 can be labeled according to compliance categories to guide the machine learning models 185 in identifying relevant compliance topics. For unsupervised learning, the training dataset 170 can provide unlabeled data, allowing the machine learning models 185 to detect patterns or clusters related to compliance topics without predefined categories. The training dataset 170 can be formatted in standard formats such as CSV, JSON, or custom formats to facilitate compatibility with the machine learning models 185. The training dataset 170 can include curated text examples, annotated with specific compliance categories, functioning as benchmarks for training or evaluating the machine learning models 185 on compliance-based identification tasks. The sequence length in the training dataset 170 can be limited to a threshold to optimize the machine learning model’s performance. The training dataset 170 can include a filtered version of applicable compliance categories and be divided into training, test, and validation sets to facilitate thorough model evaluation.

[0046] The training dataset 170 can include examples of control parameters 167. Each control parameter 167 can correspond to a set of rules defining operational standards for one or more products 165. The control parameters 167 can correspond to compliance thresholds across various regulatory, legislative, or industry standards. The control parameters 167 can specify a range of requirements, including regulatory compliance (e.g., with laws, legislative directives, orindustry standards), security standards (such as for data privacy, access control, and encryption), performance metrics to satisfy benchmarks, or functional requirements that define product features and capabilities. The training dataset 170 can incorporate instances of control parameters 167 and associated metadata to facilitate the identification of updates to control parameters 167. For example, the metadata can provide contextual information, such as parameter descriptions, formal definitions, or historical records of control parameter values, allowing the machine learning models 185 to evaluate how control parameters change over time. The machine learning models 185 can utilize these examples to learn how to detect modifications in control parameters based on shifts in metadata attributes or other contextual indicators. For example, the training dataset 170 can include historical datasets of regulatory requirements, legislative directives, or industry standards, with each dataset capturing corresponding updates to control parameters over time. By evaluating these examples, the machine learning models 185 can identify update patterns, such as the introduction of new compliance thresholds or modifications to existing standards, and apply this knowledge to detect changes in future data.|0047| The training dataset 170 can include examples of control parameters 167 and corresponding code segments 169 to train the machine learning models 185 to identify or modify code segments 169 related to specific control parameters 167. For example, the training dataset 170 can include code segment examples that implement various functionalities, including payroll processing, benefits administration, time tracking, or employee management, each paired with associated control parameters 167 that define compliance requirements for these functions. These code segments 169 can be annotated with tags or comments to indicate the specific control parameters 167 they relate to. By training on a diverse range of code examples, written in different programming languages and coding styles, the machine learning model 185 can learn to identify patterns or relevant code segments 169 associated with specific control parameters 167.|0048| The data processing system 105 can include, interface with, communicate with, or otherwise utilize a system processor 130. The system processor 130 can be or include any script, file, program, application, set of instructions, or computer-executable code that can be configured to execute one or more instructions associated with the data processing system 105. The systemprocessor 130 can include an electronic processor, an integrated circuit, or the like, including one or more of digital logic, analog logic, digital sensors, analog sensors, communication buses, volatile memory, nonvolatile memory, and the like. The system processor 130 can include, but is not limited to, at least one microcontroller unit (MCU), microprocessor unit (MPU), central processing unit (CPU), graphics processing unit (GPU), physics processing unit (PPU), embedded controller (EC), or the like. The system processor 130 can include a memory operable to store one or more instructions for operating components of the system processor 130 and operating components operably coupled to the system processor 130. For example, the one or more instructions can include one or more of firmware, software, hardware, operating systems, or embedded operating systems. The system processor 130 or the data processing system 105 generally can include one or more communication bus controllers to effect communication between the system processor 130 and the other elements of the data processing system 105.

[0049] The data processing system 105 can include, interface with, communicate with, or otherwise utilize an interface controller 135. The interface controller 135 can be or include any script, file, program, application, set of instructions, or computer-executable code that can be configured to facilitate communication between the data processing system 105, the client system 115, and the machine learning system 120. The interface controller 135 can include hardware, software, or any combination. The interface controller 135 can facilitate communication between the data processing system 105, the network 110, the client system 115, or the machine learning system 120 via one or more communication interfaces. A communication interface can include, for example, an application programming interface (“API”) compatible with a particular component of the data processing system 105, the client system 115, or the machine learning system 120. The communication interface can provide a particular communication protocol compatible with a particular component of the data processing system 105, a particular component of the client system 115, or a particular component of the machine learning system 120. The interface controller 135 can be compatible with particular content objects and can be compatible with particular content delivery systems corresponding to particular content objects, structures of data, types of data, or any combination thereof. For example, the interface controller 135 can becompatible with the transmission of structured or unstructured data according to one or more metrics.

[0050] The data processing system 105 can include, interface with, communicate with, or otherwise utilize an input receiver 140. The input receiver 140 can be or include any script, file, program, application, set of instructions, or computer-executable code that can be configured to process input data in the form of a message, query, or prompt. The input receiver 140 can include hardware, software, or any combination thereof. The input data can include any user-provided command, request, or text data. The input receiver 140 can receive textual inputs from the client system 115 in natural language (e.g., a text string) and process textual inputs through user interactions with the chatbot interface 175. User interactions can include clicking buttons, entering text, or using voice commands within the chatbot interfaces 175. The input receiver 140 can expose an API endpoint, allowing other applications or systems to send prompts in structured formats such as JSON, facilitating tasks related to payroll, compliance, or other system interactions. The input receiver 140 can identify textual inputs related to an update to a control parameter 167 by extracting or receiving an associated keyword, phrase, or identifier. The input receiver 140 can identify keywords or phrases, such as update, modify, or change, to detect an update to a control parameter 167. In some embodiments, the input receiver 140 can leverage natural language processing techniques to interpret the semantic meaning of the textual input and determine the user’s intent related to an update to a control parameter 167. In some embodiments, the input receiver 140 can process structured input formats, such as JSON or XML, that specify the control parameter 167 to be updated.[00511 The data processing system 105 can include, interface with, communicate with, or otherwise utilize a product manager 145. The product manager 145 can be or include any script, file, program, application, set of instructions, or computer-executable code that can be configured to manage or update data associated with products 165. The product manager 145 can assign identifiers to each product 165. The product manager 145 can retrieve or update the products 165 based on the associated identifiers. When additional or updated data points are received, the product manager 145 can extract relevant attributes and match them to specific products 165. Theproduct manager 145 can maintain an index structure that maps product identifiers to their corresponding data. Upon receiving an additional data point, the product manager 145 can query the index using the extracted identifier to identify the target products 165. The product manager 145 can implement a hash table, where the identifier can be hashed to generate a key. The product manager 145 can use the key to access the corresponding products 165. In some embodiments, the product manager 145 can maintain access permissions associated with products 165 by assigning role-based permissions to fields within the product data structure. As user roles or organizational structures change, the product manager 145 can update the permissions.

[0052] The product manager 145 can identify a corresponding product 165 in response to a request from the client device 115. For example, if a client device 115 requests information about a specific product 165, the product manager 145 can receive an identifier from the client device 115, which may be a product ID, device ID, MAC address, IP address, or other identifying information. The product manager 145 can query the database 125 using the received identifier to identify the corresponding product 165. If a matching identifier is identified from the product data structure, the product manager 145 can retrieve information associated with the product 165, such as its specifications, pricing, availability, or compliance status.

[0053] The product manager 145 can identify, based on updates to the control parameters 167 and product metadata 166, one or more products 165 that fail to comply with these updates. The product manager 145 can retrieve relevant metadata for each product 165 from the database 125, including information about product attributes, configurations, or current compliance status. The product manager 145 can evaluate the updates to control parameters 167 to detect specific changes, such as new regulations, legislative directives, or security standards, and determine their implications for compliance. The product manager 145 can compare product metadata 166 with compliance thresholds defined by the updates to control parameters 167 to detect discrepancies, such as determining whether specific attributes, values, or configurations align with the newly established standards. The compliance thresholds can specify a limit of allowable values, required configurations, or specific conditions that a product is to satisfy to be deemed compliant. For example, a compliance threshold can desire certain security protocols to be active, a particularconfiguration setting to be configured, or performance parameters to remain within a defined range. Based on the comparison, the product manager 145 can evaluate each product’s compliance status to determine whether specific compliance criteria are satisfied. For example, if a new security regulation requires products to implement a specific security patch, the product manager 145 can retrieve metadata for all products, evaluate the security requirements, compare each product’s security configuration, identify those that lack the required patch, and generate a report listing non-compliant products with details on the specific issues.

[0054] The data processing system 105 can include, interface with, communicate with, or otherwise utilize a code modifier 150. The code modifier 150 can be or include any script, file, program, application, set of instructions, or computer-executable code that can be configured to manage or update code segments 169 associated with products 165. The code modifier 150 can automate updates to the source code 168, incorporating the code segments 169, to comply with additional or modified control parameters 167 for maintaining compliance across systems. The code modifier 150 can identify code segments 169 related to control parameters 167 in response to mapping control parameters 167 to the code segments 169 that implement functionalities related to those parameters. The functionalities can refer to specific operations or processing tasks that the code segments 169 perform to fulfill the specifications defined by the control parameters 167. These functionalities can include security protocols, data validation processes, configuration settings, or compliance checks, each of which can correspond to particular aspects of the control parameters 167. The code modifier 150 can utilize metadata associated with the source code 168, such as comments, tags, or keywords, to identify which code segments 169 correspond to specific control parameters 167. In some embodiments, the code modifier 150 can evaluate code syntax, semantics, or control flow to determine or isolate target code segments 169 that implement functionalities related to the control parameters 167.[0055| The code modifier 150 can implement template-based code modifications (e.g., automating repetitive updates such as configuration changes), rule-based code modifications (e.g., replacing values, adding or removing code blocks, or adjusting control flow), or Al-assisted code modifications, where machine learning or natural language processing techniques can interpretcode intent to generate or apply updates. In response to updates to control parameters 167, the code modifier 150 can identify relevant code segments 169 associated with the impacted functionalities. For example, a change in tax laws can trigger the identification of code segments 169 related to payroll calculations, tax deductions, or reporting. Once the relevant code segments are identified, the code modifier 150 can generate targeted modifications within the source code 168 to comply with the updated control parameters 167. These modifications can include updating tax formulas, adjusting withholding rates, or modifying reporting formats. The code modifier 150 can execute the modified code segments to perform various operations depending on the updates to the control parameters 167. These operations can include updating payroll calculations, adjusting benefit deductions, modifying employee records, generating compliance reports, or implementing new HR policies. The code modifier 150 can execute the modifications directly or through an automated deployment process such that the updated code segments 169 are integrated into the system and operational. 0056] The code modifier 150 can automatically generate or execute unit tests to verify the correctness of the modified code segments 169. The code modifier 150 can integrate the modified code segments 169 with other system components and test their interaction. The code modifier 150 can track changes or manage different versions of the code. The code modifier 150 can automate the deployment process, including building, packaging, or deploying the modified code to target environments. For example, in response to a new tax regulation affecting payroll calculations in an HR software product, the code modifier 150 can identify relevant code segments 169, update tax calculation logic to satisfy the requirements, perform unit testing to validate changes, or deploy the updated code to production to maintain timely compliance with changes or updates to control parameters 167.

[0057] The data processing system 105 can include, interface with, communicate with, or otherwise utilize a search engine 160. The search engine 160 can be or include any script, file, program, application, set of instructions, or computer-executable code that can be configured to process queries and retrieve relevant information from the database 125. A query can be a structured command or statement used to retrieve data stored in the database 125. The searchengine 160 can process queries by parsing them into individual tokens, normalizing the tokens (e.g., parts of the query) to a standard form, and applying stemming or lemmatization to reduce words to their root forms. The search engine 160 can search an index, which can be a structured database of crawled and indexed content. The search engine 160 can perform an index lookup to identify potential search results corresponding to the query terms. The search engine 160 can apply Boolean operations to refine search results based on query structure. The search engine 160 can rank the potential matches based on relevance, using factors such as keyword relevance, content quality, and user engagement, among others. The search engine 160 can generate snippets to summarize each result and format the results in a layout that includes related searches.|0058| The search engine 160 can utilize a search query generated by the large language model 185 based on keywords to filter data related to control parameters 167 stored in the database 125. The filtered data can include various information, such as parameter definitions, usage contexts, associated regulations, and historical data, among others. The database 125 can maintain an index of control parameters to facilitate efficient search and retrieval based on attributes such as keywords, categories, or fields. The search engine 160 can parse the search query to identify search terms and logical operators. In some embodiments, the search engine 160 can implement query rewriting or query expansion. The search engine 160 can apply the search query to generate a set of filtered data related to control parameters 167 by extracting relevant entries associated with the query based on keyword matching, phrase matching, or semantic search. The search engine 160 can apply criteria such as date range, category, or relevance score to rank the filtered data. For example, if the search query is “tax regulations for remote workers”, the search engine 160 can identify keywords such as “tax”, “regulation”, “remote work”, and “worker”, query the database to identify relevant control parameters, filter to include those control parameters related to tax regulations for remote work, and rank them by relevance to the query. In cases of ambiguous queries or no matching results, the search engine can provide appropriate error messages or suggestions.

[0059] The data processing system 105 can include, interface with, communicate with, or otherwise utilize an operation controller 162. The operation controller 162 can be or include anyscript, file, program, application, set of instructions, or computer-executable code that can be configured to manage and execute actions associated with one or more components of the data processing system 105, the client system 115, or the machine learning system 120. The operation controller 162 can define and manage workflows comprised of multiple interconnected tasks. The operation controller 162 can initiate, monitor, and control the execution of workflow steps. The operation controller 162 can implement conditional logic for dynamic workflow routing. The operation controller 162 can execute multiple tasks concurrently through parallel processing. The operation controller 162 can implement error handling and recovery mechanisms for workflow exceptions. The operation controller 162 can track workflow progress and provide status updates. For example, the operation controller 162 can include one or more interfaces to detect input at various portions of a workflow and can provide output responsive to specific portions of a workflow.

[0060] The machine learning system 120 can include, interface with, communicate with, or otherwise utilize one or more machine learning models 185 trained on datasets. The machine learning system 120 can include a cloud system, a server, a distributed remote system, or any combination thereof. The machine learning system 120 can include, but is not limited to, at least, central processing unit (CPU), graphics processing unit (GPU), tensor processing units (TPUs), or the like. The machine learning system 120 can include a memory operable to store one or more instructions for operating components of the machine learning system 120 or operating components operably coupled to the machine learning system 120. The machine learning system 120 can be internal to the data processing system 105. The machine learning system 120 can exist external to the data processing system 105 and can be accessed via the network 110. The machine learning system 120 implements or otherwise provides access to one or more application programming interfaces (APIs), via which the data processing system 105 or the client system 115 can access the one or more machine learning models 185 or other functionality of the machine learning system 120.

[0061] The machine learning system 120 can include, interface with, communicate with, or otherwise utilize various machine learning models 185. The machine learning model 185 canbe deployed within the data processing system 105 (as indicated by machine learning model 155) or externally as remote services. The machine learning model 155, residing within the data processing system 105, can be similar to, and include any of the structure and functionality of, the machine learning model 185. The machine learning models 185 can be trained with same or different types of machine learning techniques, trained with the same or different types of training data, or trained or configured to receive different types of input or provide different types of output. Example machine learning techniques can include neural networks, such as a generative adversarial network (e.g., a generator neural network and a discriminator neural network that are trained simultaneously through adversarial training), a variational autoencoder (e.g., an autoencoder neural network that learns to generate new data samples by modeling the underlying probability distribution of the data), an autoregressive model, or other types of neural networks (e.g., deep learning models, convolutional neural networks, recurrent neural networks, or transformers). Transformers can refer to or include a type of deep learning model architecture configured for natural language processing, including, for example, bidirectional encoder representations (“BERT”), generative pre-trained transformers, text-to-text transformer, transformer-XL, robustly optimized BERT, or distilled BERT. Other types of machine learning techniques can include supervised learning models, unsupervised learning models, semisupervised learning models, or reinforcement learning models. For example, a supervised machine learning technique can include a support vector machine used for classification and regression tasks. Given a set of labeled training data, a support vector machine can identify the hyperplane that separates the data into classes with the largest possible margin (e.g., distance between the hyperplane and nearest data points from each class).

[0062] The machine learning model 185, such as the large language model 185, can perform a variety of text processing tasks, including, but not limited to, generating text, formatting instructions, comprehending and processing natural language input, and responding to prompts with contextually relevant information. The large language model 185 can include a transformer architecture, such as a generative pre-trained transformer (GPT) architecture. The transformer architecture can include an encoder that can process the input text and a decoder that can generatethe output text. The large language model 185 can include multiple layers that can operate to process and generate text. For example, embedding layers can convert words or tokens into dense vectors of fixed size, attention layers can use mechanisms such as self-attention to weigh the importance of different tokens in a sequence, and feedforward layers can apply transformations to the data to learn complex patterns. The large language model 185 can use a self-attention mechanism to weight different parts of the input sequence when generating predictions. The large language model 185 can determine one or more properties of an input prompt that correspond to natural language. Natural language input can have a syntactic structure in which individual words, collections of words (e.g., phrases), or relative positions of words (e.g., word order) can indicate specific meanings. The large language model 185 can parse sentences into their grammatical components to understand the structure and relationships between words. The large language model 185 can use phrasing structure rules that define how words combine to form phrases and sentences.

[0063] The large language model 185 can process a wide range of input formats, including but not limited to text, audio, images, video, or other modalities. Based on the input context, the large language model 185 can generate output, which can range from simple text responses to complex data structures. The large language model 185 can use the input prompt to iteratively predict the next token, word, or phrase to generate responses in response to prompts. In some embodiments, the textual input can be indicative of an update to a control parameter 167, where a prompt including information on regulatory, industry standard, or legislative changes can trigger adjustments in the product parameters. The large language model 185 can generate instructions or commands that automatically invoke tools or functions to perform specific tasks or operations. The large language model 185 can generate formatting instructions that define the visual attributes of graphical elements. In some embodiments, based on the input prompt, the large language model 185 can generate a data structure, including details such as text content, layout specifications, and visual or graphical elements.

[0064] The large language model 185 can generate a search query based on textual input. The large language model 185 can classify the textual input into one or more search terms orkeywords. The large language model 185 can combine the classified keywords into a unified query string and apply logical operators such as AND, OR, or NOT to establish relationships between keywords. The large language model 185 can utilize its understanding of natural language and context to generate a relevant search query that can retrieve information from the database 125 or other knowledge sources. The large language model 185 can construct search queries that can be in the form of SQL queries, regular expressions, or queries specific to the database 125 or search engine in use. For example, given the input “How does new US legislation impact data privacy standards?”, the large language model 185 can classify search terms such as “US legislation”, “data privacy standards”, and “impact” and generate a search query such as “US legislation AND data privacy impact” OR “US data privacy law changes” OR “new US data privacy regulations”. In cases of ambiguity or insufficient information, the large language model 185 can employ error handling mechanisms to provide informative feedback or suggest alternative search terms.

[0065] The large language model 185 can detect updates to control parameters 167 based on filtered data. The large language model 185 can process various data formats (e.g., structured, semi-structured, or unstructured) to identify relevant attributes and relationships. The large language model 185 can compare current data with historical data or previous versions to detect deviations from expected patterns. The large language model 185 can evaluate patterns over time to track trends or detect changes in control parameters 167. The large language model 185 can evaluate the semantic structure of filtered data to identify keywords, phrases, or entities related to control parameters 167. Based on the keywords, phrases, or entities, the large language model 185 can evaluate the context of the data to determine if updates are present. The large language model 185 can be trained on historical data or previous control parameter changes to learn patterns in the data or detect deviations. The large language model 185 can apply predefined rules based on keywords, regular expressions, or syntactic patterns to identify changes. For example, the large language model 185 can identify keywords indicating updates.[0066| The data processing system 105 can execute the machine learning model 155 or cause the machine learning system 120 to execute the machine learning model 185 to determine or identify the code segments 169 within the source code 168 corresponding to one or moreproducts 165. For example, the large language model 185, trained on a dataset of control parameters 167, can determine relevant code segments 169 by identifying updates to control parameters 167. The large language model 185 can parse the source code into smaller tokens, such as words, symbols, or numbers, and the large language model 185 can evaluate the syntactic structure to understand relationships between code elements 169. The large language model 185 can evaluate the semantic context of variables, functions, or control flow statements to determine the intended functionality of the code. The large language model 185 can detect keywords, phrases, or patterns associated with control parameters 167 to identify relevant control parameters 167. The large language model 185 can utilize knowledge graphs or ontologies that specify relationships between control parameters 167 and code segments 169 specific to each product 165. The large language model 185 can match the identified control parameters 167 with code segments 169 within a product’s source code 168 to determine relevant code sections based on various factors, such as code comments, documentation, or version history. For example, when a new tax regulation requires an update to payroll calculation logic, the large language model 185 can identify keywords such as “tax”, “income tax”, “payroll”, “deduction”, and “withholding”, evaluate code structure to identify functions relevant to payroll calculations, match the keywords with specific code segments, and recommend modifications, such as modifying one or more formulas, conditions, or data structures.

[0067] The large language model 185 can receive modified code segments 169 and updates to control parameters 167 from the data processing system 105 and can generate a data structure indicating compliance of the respective product 165 based on these updates. The large language model 185 can evaluate the modified code segments to interpret specific changes and determine their impact on the product’s functionality or adherence to updated control parameters. The large language model 185 can map the updated control parameters to the specific code segments that have been modified to establish an association between control parameters and corresponding code changes. The large language model 185 can evaluate the modified code segments against updated control parameters to determine whether the product satisfies the new requirements. For example, the large language model 185 can check for specific keywords, patterns, or code structuresindicative of compliance. The large language model 185 can generate a data structure including information such as the product ID, updated control parameters, modified code segments, compliance status (e.g., compliant, partially compliant, non-compliant), reason for compliance status (e.g., successful modification, pending testing, identified issues), timestamp of the update, and other relevant metadata such as version control details. The data structure can support various applications, including generating compliance reports, tracking change history, or identifying potential compliance risks.

[0068] The machine learning system 120 can include, interface with, communicate with, or otherwise utilize a machine learning model manager 190. The machine learning model manager 190 can train, fine-tune, update, re-train, or otherwise maintain one or more machine learning models 185. The machine learning model manager 190 can include hardware, software, or any combination thereof. The machine learning model manager 190 can reside within the data processing system 105, where the machine learning model manager 190 can manage and coordinate the training, fine-tuning, and updating of the machine learning models 155. In some embodiments, the model manager 190 can exist outside the data processing system 105, operating as a remote service that interacts with the data processing system 105 via APIs. The machine learning model manager 190 can facilitate the training of machine learning models 185 on training datasets 170 and can fine-tune the machine learning models 185 for specific tasks or domains. The machine learning model manager 190 can manage the deployment of machine learning models 185 in production environments and the integration of machine learning models 185 with other systems and services.

[0069] The machine learning model manager 190 can continuously monitor the performance of deployed machine learning models 185, identify issues, and update machine learning models 185. The machine learning model manager 190 can use various machine learning algorithms, including supervised learning techniques, to train, fine-tune, or update one or more machine-learning models 185 using labeled data or training dataset 170 to improve the prediction accuracy, classification capabilities, and intent determination of the machine learning models 185. The machine learning model manager 190 can implement unsupervised learning techniques, suchas clustering and association rule mining, to identify patterns in unlabeled data, or to generate labels for said unlabeled data. The machine learning model manager 190 can implement reinforcement learning techniques to update the machine learning models 185 based on user- provided feedback or training dataset 170.

[0070] The machine learning model manager 190 can train the machine learning model 185 on a large dataset 170 that includes examples of control parameters 167 and corresponding code segments 169 within the source code 168, associated with specific products 165. Each code segment 169 in the training dataset 170 can be labeled or annotated to specify the control parameter 167 it relates to, such as regulatory requirements, legislative directives, or industry protocols, and the associated product 165 to which the code segment 169 belongs. The machine learning model manager 190 can balance the training dataset 170 by maintaining a representative number of examples for each control parameter category and product association to prevent model bias. After training, the machine learning model manager 190 can evaluate the performance of the machine learning model 185 using a separate validation dataset to confirm that the machine learning model 185 can accurately identify and classify code segments related to different control parameters or products. The machine learning model manager 190 can train the machine learning model 185 to identify or categorize control parameters as they apply to specific products 165. The machine learning model manager 190 can refine the machine learning model’s ability to detect specific patterns, such as frequent modifications to configuration settings or adjustments to compliance thresholds, based on the operational requirements of each product 165.

[0071] FIG. 2 depicts a method 200 of code modification for regulatory compliance using a virtual assistant based on machine learning models and control parameters. The method 200 can be implemented using a system 100, 900, 1000, and 1100, or any other features discussed in FIG. 1, FIGS. 9-11, or FIG. 12. The method can include Acts 202-216. The Acts 202-216 can be executed in any order or sequence.

[0072] At 202, the method 200 can receive an input from a client device indicative of an update to a control parameter. In an aspect, the method can include receiving the input via achatbot interface. The chatbot interface can receive natural language queries, text-based commands, or voice commands. The chatbot interface can allow users to provide updates regarding control parameters using natural language queries.

[0073] At 204, the method 200 can generate a search query based on the input. In an aspect, the method can include generating, using a large language model, the search query based on the input. The large language model can be trained on a dataset including at least one of regulatory requirements, industry standards, or legislative directives. In an aspect, the method can include inputting the input into the large language model to cause the large language model to classify the input into one or more search terms. In another aspect, the method can include generating, using the large language model, the search query based on the one or more search terms.

[0074] At 206, the method 200 can fdter data related to control parameters from a database. In an aspect, the method can include filtering, using the search query, data related to control parameters stored in the database. The control parameters can correspond to a set of rules defining operational standards for the one or more products.

[0075] At 208, the method 200 can detect updates to the control parameters. In an aspect, the method 200 can include detecting, using the large language model, updates to the control parameters based on the filtered data. The large language model can evaluate the filtered data for changes in keywords, phrases, or semantic meaning that indicate updates to control parameters.

[0076] At 210, the method 200 can identify one or more products that fail to comply with the updates to the control parameters. In an aspect, the method can include identifying, based on the updates to the control parameters and product metadata, one or more products that fail to comply with the updates. The product metadata can include information related to product configurations, operational settings, usage patterns, or deployment environments. In another aspect, the method can include identifying, responsive to comparing the product metadata of the one or more products to compliance thresholds defined by the updates to the control parameters, that the one or more products fail to comply with the updates to the control parameters.

[0077] At 212, the method 200 can determine code segments related to the non-compliant control parameters. In an aspect, the method can include determining code segments within a source code corresponding to the one or more products. The code segments can be related to the control parameters. In an aspect, the method can include determining the code segments within the source code responsive to mapping the control parameters to the code segments implementing functionalities related to the control parameters. In another aspect, the method can include executing the large language model, trained on a dataset of control parameters, to determine the code segments within the source code.

[0078] At 214, the method 200 can modify the code segments to comply with the updates to the control parameters. In an aspect, the method can include generating code modifications to the code segments based on the updates to the control parameters. The modifications can include adding, removing, or modifying code segments such that the code aligns with the updates to the control parameters.

[0079] At 216, the method 200 can execute the modified code segments to perform an operation. In an aspect, the method can include executing the generated code modifications to comply with the updates to the control parameters. In an aspect, the method can include inputting the modified code segments and the updates to the control parameters into the large language model to cause the large language model to generate a data structure. In an aspect, the method can include parsing the data structure to display, via the client device, an electronic report indicating compliance of the one or more products with the updates to the control parameters.

[0080] FIG. 3 depicts an example compliance monitoring system 300 according to one or more aspects of the technical solutions described herein. As illustrated by way of example in FIG. 3, a compliance monitoring system 300 can facilitate adherence to various regulations or standards across multiple jurisdictions. The compliance monitoring system 300 can include a range of components to support compliance with respect to control parameters.

[0081] The compliance monitoring system 300 can include compliance authorities 302 from different countries, where each authority may have its own set of reporting or compliancerequirements. These requirements can vary significantly across countries. Each compliance authority 302 can release new control parameters or issue changes to existing control parameters in line with updated legislation or regulatory standards. The compliance monitoring system 300 can include an authority connector 304 to facilitate data exchange, report submissions, and automated synchronization with the compliance authorities 302. For example, the authority connector 304 can provide API-based connectivity (e.g., RESTful APIs) or micro-front-end-based data rendering to exchange data with the compliance authorities 302. The authority connector 304 can implement robust security measures, such as encryption and authentication. The authority connector 304 can define a standardized input format to maintain connectivity across diverse compliance authorities 302 and products 306. The authority connector 304 can regulate health checks, audit trails, and other governance mechanisms to facilitate compliance with control parameters.

[0082] The compliance monitoring system 300 can include multiple products 306, such as a human capital management platform to manage employee data, payroll, or benefits, a global payroll platform to process payroll across countries, and a client service level to monitor customer interactions and service level agreements. The products 306 can push reporting data to the authority connector 304. The authority connector 304 can transmit the data to compliance authorities 302 automatically via APIs or manually through secure file transfer protocols.

[0083] The compliance monitoring system 300 can include a legislation bank 308 or knowledge bank 308 that stores or maintains control parameters, including laws, regulations, or industry standards. The legislation bank 308 or the knowledge bank 308 can function as a centralized compliance reference. An automated process within the compliance monitoring system 300 can detect and update the updates to control parameters in the legislation bank 308 or knowledge bank 308. In some embodiments, a legislation manager can manually add the updates to the control parameters to the bank.|0084| The compliance monitoring system 300 can include a compliance meter 310 to monitor the compliance status of each product 306 by identifying potential compliance gaps. Thecompliance meter 310 can determine updates to control parameters, which correspond to regulatory requirements, industry standards, or legislative directives. The control parameters can impact an organization’s products or services. The compliance meter 310 can collect data from various sources, such as databases, APIs, legislation or knowledge banks, and file systems, including product information, configuration settings, usage patterns, or regulatory updates. The compliance meter 310 can evaluate the data to identify relationships between data points or correlate product data with relevant control parameters. The compliance meter 310 can use data mining, machine learning, or statistical analysis to extract insights from the data. Based on these insights, the compliance meter 310 can determine the compliance status of a product by evaluating it relative to the updates to the control parameters. These updates can include changes in regulatory requirements, new industry standards, or revised legislative directives. The compliance meter 310 can determine each product’s configuration, usage patterns, or operational parameters to identify any deviations from the current control parameters. The compliance meter 310 can implement annotations to the product 306 that fails to comply with updated control parameters. These annotations can indicate specific areas where the product 306 fails to satisfy regulatory requirements, industry standards, or legislative directives. The compliance meter 310 can include human-readable annotations detailing the control parameter violated, the level of non-compliance, or suggested actions to bring the product 306 into compliance. These annotations can be implemented as comments within the code, tags in configuration files, or entries in documentation.

[0085] The compliance monitoring system 300 can include recommendation board 312 that utilizes insights from the compliance meter 310 to provide actionable guidance on addressing compliance gaps in relation to control parameters and code segments. The recommendation board 312 can recommend modifications to code segments or updates to control parameters to satisfy regulatory requirements, industry standards, or legislative directives. The recommendation board 312 can recommend targeted actions, such as code revisions, security updates, or configuration adjustments, and provide technical guidance on interpreting control parameters or applying them directly to relevant code segments. The compliance monitoring system 300 can include a language model (LM) chat 314 to provide a communication interface. The LM chat 314 can allow users tointeract with the compliance monitoring system 300 in natural language. The LM chat 314 can be a voice-based LM chat engine. The LM chat 314 can facilitate real-time or near real-time conversations between users and the compliance monitoring system 300, where users can ask questions, request information on compliance status, or seek clarification on specific control parameters associated with one or more products 306. In some embodiments, the updates to the control parameters stored in the legislation bank or knowledge bank 308 can be retrieved by the LM chat 314 or the recommendation board 312 to provide users with timely insights or recommendations based on the latest compliance information.

[0086] FIG. 4 depicts an example legislation bank architecture 400 according to one or more aspects of the technical solutions described herein. As illustrated by way of example in FIG. 4, the legislation bank architecture 400 can expose a legislation bank 404 to a web or mobile application 402 via HTTP APIs. The HTTP APIs can be hosted on a serverless computing platform 408. In some embodiments, the serverless computing platform 408 can provide a serverless platform for hosting REST APIs. Within the legislation bank 404, an API gateway 406 and the serverless computing platform 408 can manage the hosting environment for the APIs, such that the legislation bank architecture 400 can be scaled automatically or can manage requests without a dedicated server infrastructure. The legislation bank 404 can store legislative or control parameters, including updates to the control parameters, in various structured formats (e.g., JSON, XML) to facilitate retrieval and organization of the control parameters. The legislation bank architecture 400 can store metadata associated with each legislation or control parameter within a serverless database 410. The serverless database 410 can provide a managed, multi-tenant database environment with features such as auto-scaling, multi-AZ (availability zone) replication, and high availability, such that the metadata storage remains robust and can scale with demand.

[0087] The legislation bank architecture 400 can include a knowledge bank 412 to store and retrieve detailed content related to legislation or control parameters. The knowledge bank 412 can implement natural language processing (NLP) techniques to provide intelligent search capabilities based on natural language queries. The knowledge bank 412 can support incremental learning, such that legislation bank architecture 400 can continuously update the knowledge base412 with updated information related to changes in regulations and standards. In some embodiments, the knowledge bank 412 can be accessed as a managed service.

[0088] The legislation bank architecture 400 can include object storage 414 to store legislative or control parameter data in a structured format, such as JSON, for flexible storage and retrieval of data structures. The object storage 414 can provide virtually unlimited storage capacity to store a vast amount of legislative or regulatory information. In some embodiments, the object storage 414 can implement encryption at rest and in transit to maintain the security and confidentiality of the stored data. The object storage 414 can facilitate easy access and integration with other components of the system.

[0089] The legislation bank architecture 400 can include a quality engineering as a service (QEAS) compliance meter 416. The QEAS compliance meter 416 can manage test suites for automation and store test annotations in a human-readable format. The QEAS compliance meter 416 can function as an established automation test framework or methodology within the compliance infrastructure to validate compliance with the updates to the control parameters. The QEAS compliance meter 416 can support compliance monitoring through integration with the AVA chatbot infrastructure. The QEAS compliance meter 416 can integrate with other components of the system, such as the One CI-CD pipeline, the OneUX framework, the metadata framework, and the Nexo framework.

[0090] FIG. 5 depicts an example authority connector 500 according to one or more aspects of the technical solutions described herein. As illustrated by way of example in FIG. 5, the authority connector 500 can facilitate the collection and transmission of reporting data between various products 502 and country-level authorities 512. The products 502, such as HCM platforms or global payroll platforms, can push their reporting data to a data stream processing platform 508. The data stream processing platform 508 can be managed by a third-party service provider. The data stream processing platform 508 can process and transform the reporting data into a standardized format. In some embodiments, the data stream processing platform 508 can retrieve the reporting data from the products 502 via REST API calls. These APIs, hosted on an APIgateway 504 or a serverless computing platform 506, can facilitate the efficient or scalable exchange of data between the products 502 and the data stream processing platform 508.

[0091] Once the reporting data lands in the data stream processing platform 508, the reporting can undergo collection, transformation, and storage within a serverless database 510. The serverless database 510, leveraging the serverless computing platform 506, can provide a scalable solution for data management. Following a schedule defined by an event bus 516, a serverless container 514 can be spun up on demand to prepare and submit reporting data to countrylevel authorities 512. The serverless container 514 can provide an efficient or scalable approach, as the serverless container is activated when desired, thereby minimizing resource consumption and cost. The event bus 516 can function as a central orchestrator, triggering various actions and workflows, including the activation of the serverless container 514.

[0092] FIG. 6 depicts an example compliance meter 600 according to one or more aspects of the technical solutions described herein. As illustrated by way of example in FIG. 6, the compliance meter 600 can utilize a QEAS framework 604 to automate testing and maintaining compliance across various control parameters. The QEAS framework 604 can receive data from, or otherwise interface with, various products 602, including, for example, a human capital management platform, global payroll platform, or a client service level. The QEAS framework 604 can manage and execute automation scripts in a human readable format. The compliance meter 600 can maintain test annotations for each control parameter. The test annotations can provide guidelines or best practices for implementing compliance checks within specific products. The QEAS framework 604 can execute the test annotations across different computing infrastructures, capturing compliance results specific to each environment. The QEAS framework 604 can store the test results in a serverless database 606, while any associated artifacts, such as logs and reports, can be stored in an object storage 608. The test results can be used as compliance indicators, allowing the compliance meter 600 to determine if a particular computing infrastructure within a specific country adheres to active legislative requirements or control parameters relevant to that country.]0093| FIG. 7 depicts an example LM chat 700 according to one or more aspects of the technical solutions described herein. As illustrated by way of example in FIG. 7, the LM chat 700 can facilitate user interaction with the system. Users 702 can access the LM chat 700 through various channels, such as browser-based applications 704, mobile applications 704, or voice assistants. When the user initiates a query or an audio input, the audio input can be processed by a conversational Al tool 706. The conversational Al tool 706 can utilize a text-to-speech tool 708 to synthesize text from the speech. The text-to-speech tool 708 can transcribe speech into structured text for further processing. The conversational Al tool 706 can evaluate the query to determine the underlying intent of the user 702. Once the intent is identified, the conversational Al tool 706 can trigger actions on a serverless computing platform 710 to satisfy the user’s request. The serverless computing platform 710 can access a knowledge bank 712 or a serverless database 714 to retrieve relevant information. For generic or ambiguous queries, the conversational Al tool 706 can redirect the search to the knowledge bank 712, which can leverage NLP-based search techniques to provide contextually accurate results. In some embodiments, the knowledge bank 712 can directly process text-based queries when specific information is desired. The conversational Al tool 706, text-to-speech tool 708, and knowledge bank 712 can be managed by third-party AI / ML services. Additionally, various components of the LM chat 700 shown in FIG. 7 may be similar to, and include any of the structure and functionality of, the systems of FIGS. 3- 6. For example, the QEAS compliance meter 716 shown in FIG. 7 can be similar to the compliance meter 310 in FIG. 3 or the compliance meter 600 in FIG. 6. Similarly, the object storage 718 shown in FIG. 7 can be similar to the object storage 608 in FIG. 6.

[0094] FIG. 8 depicts an example cloud-based language platform 800 according to one or more aspects of the technical solutions described herein. As illustrated by way of example in FIG. 8, the cloud-based language platform 800 can include a cloud-based personalized engine 808. A serverless computing platform 804 can retrieve the updates to the control parameters and user profde data from a serverless database 802 and other sources. The retrieved data can then be stored in an object storage 806. The cloud-based personalized engine 808 can access the data from the object storage 806, evaluate the user profile and preferences, and generate personalizedrecommendations based on the control parameters and profile attributes. In some embodiments, recommendations for the user can be generated by storing rule-based preferences in the serverless database 802. The cloud-based personalized engine 808 can retrieve the stored preferences and apply rule-based logic to generate recommendations directed to the user’s specified criteria or other requirements.]0095| The cloud-based personalized engine 808 can process and evaluate legislative documents to identify updates and changes. The cloud-based personalized engine 808 can generate summaries of the changes and evaluate their potential impact on various stakeholders, including payroll practitioners, client employees, and service teams. For example, the cloud-based personalized engine 808 can summarize the impact of new tax laws on pay structures or provide insights into changes in benefit eligibility criteria. The cloud-based personalized engine 808 can generate executive summaries for HR leadership, providing a high-level overview of legislative changes and their implications for the organization. By leveraging text summarization techniques, the cloud-based personalized engine 808 can extract information from legal documents and present it in a concise or easy-to-understand format. Additionally, various components of the cloud-based language platform 800 shown in FIG. 8 may be similar to, and include any of the structure and functionality of, the systems of FIGS. 3-7. For example, the API gateway 810 shown in FIG. 8 can be similar to the API gateway 406 in FIG. 4 or the API gateway 504 in FIG. 5. Similarly, the serverless computing platform 812 shown in FIG. 8 can be similar to the serverless computing platform 408 in FIG. 4 or the serverless computing platform 506 in FIG. 5.

[0096] FIG. 9 depicts an example system 900 according to one or more aspects of the technical solutions described herein. As illustrated by way of example in FIG. 9, a system 900 can include an application tier 902 and compliance engine 904. The application tier 902 can include one or more components, such as an application layer, products, compliance meter, regulatory authorities, and database. The application tier 902 can interact with the compliance engine 904 via API calls. The API calls can allow the application tier 902 to request compliance checks, where the application layer can send specific data or code segments to the compliance engine for evaluation and validation. The compliance engine 904 can return detailed reports on compliancestatus, potential vulnerabilities, or recommended actions. The compliance engine 904 can trigger automated remediation by initiating actions such as updating configurations or code segments based on predefined rules, policies, or updates to control parameters. The compliance engine 904 can include one or more components, such as a data processor 906, a privacy layer 908, and an availability zone 910. The data processor 906 can monitor the system for updates, cache frequently accessed data to improve performance, or identify potential issues. The privacy layer 908 can identify and mask personally identifiable information (PII) to protect user privacy, implement data anonymization and pseudonymization techniques, and manage the lifecycle of PII data. The availability zone 910 can provide a scalable infrastructure for the compliance engine 904. The availability zone 910 can include API gateways to expose services and control access, utilize serverless computing to efficiently process data and execute tasks, implement a serverless secrets manager to securely store and manage sensitive information, and integrate Al tools to automate tasks, analyze data, and generate insights. In some embodiments, the compliance engine 904 can include cloud watch and auto-scaling functionalities to monitor system performance or automatically adjust resources to satisfy changing demands.

[0097] The compliance engine 904 can include an explanation generator 1002, as shown by way of example in FIG. 10. The explanation generator 1002 can include one or more components, such as a generative Al model and a response model. The explanation generator 1002 can receive input prompts corresponding to compliance reports, audit findings, or user queries. The input prompts can be used to guide the generative Al model in generative relevant or informative explanations. The input prompts can be directed to specific use cases, such as summarizing updates to control parameters, generating compliance assessment reports, or providing guidance on remediation actions. In this regard, the generative Al model can include large language models, which can be trained on a vast dataset of compliance parameters or corresponding code segments of the respective products. The generative Al model can generate human-quality text, such as explanations, summaries, or recommendations. The response model can be used for formatting and delivering the generated explanations to the user. For example, the response model can present information in various formats, including plain text, HTML, or otherstructured formats. Additionally, various components of the system 1000 shown in FIG. 10 may be similar to, and include any of the structure and functionality of, the system of FIG. 9.

[0098] FIG. 11 depicts an example operational system 1100 according to one or more aspects of the technical solutions described herein. As illustrated by way of example in FIG. 11, an operational system 1100 can execute a series of operations to process user input indicative of updates to control parameters and generate compliance assessment reports or modify code segments associated with non-compliant products to comply with the updates.[0099| A user can initiate the operation by logging into the system via the application interface 1102. The user can input, via the application interface 1102, a query or request related to updates to control parameters, such as legislative or regulatory changes. A compliance meter 1104 can receive the query. The compliance meter 1104 can interact with the data repository 1108 via API calls 1106 to retrieve relevant data, including rules, applicable regulations, historical compliance information, or product-specific configurations. The API call 1106 can facilitate communication between the compliance meter 1104 and the data repository 1108.

[0100] The data repository 1108 can query transaction data 1110 using a prompt message to retrieve additional relevant data, such as transaction records, user profiles, or historical data related to the user’s query or contextual information. A prompt engine 1114 can integrate the user’s query with relevant data extracted from the data repository 1108 and the transaction data 1110 to generate a refined prompt formatted in a structured schema suitable for processing via an API call to the LLM API 1112. The LLM API 1112 can receive the API call and process the refined prompt. The LLM API 1112 can implement robust auditing and monitoring mechanisms to track the performance of the model, identify potential biases, or detect any deviations from established guidelines. The LLM API 1112 can generate a response, which can be a text-based explanation, a code snippet, or a combination of both. The generated response is transmitted to the application interface 1102, where the response is presented to the user.

[0101] FIG. 12 depicts a block diagram of a computing system 1200 for implementing the embodiments of the technical solutions discussed herein, in accordance with various aspects. FIG.12 illustrates a block diagram of an example computing system 1200, which can also be referred to as the computer system 1200. Computing system 1200 can be used to implement elements of the systems and methods described and illustrated herein. Computing system 1200 can be included in and run any device (e.g., a server, a computer, a cloud computing environment or a data processing system).101021 Computing system 1200 can include at least one bus data bus 1205 or other communication device, structure or component for communicating information or data. Computing system 1200 can include at least one processor 1210 or processing circuit coupled to the data bus 1205 for executing instructions or processing data or information. Computing system 1200 can include one or more processors 1210 or processing circuits coupled to the data bus 1205 for exchanging or processing data or information along with other computing systems 1200. Computing system 1200 can include one or more main memories 1215, such as a random access memory (RAM), dynamic RAM (DRAM), cache memory or other dynamic storage device, which can be coupled to the data bus 1205 for storing information, data and instructions to be executed by the processor(s) 1210. Main memory 1215 can be used for storing information (e g., data, computer code, commands or instructions) during execution of instructions by the processor(s) 1210.

[0103] Computing system 1200 can include one or more read only memories (ROMs) 1220 or other static storage device 1225 coupled to the bus 1205 for storing static information and instructions for the processor(s) 1210. Storage devices 1225 can include any storage device, such as a solid-state device, magnetic disk or optical disk, which can be coupled to the data bus 1205 to persistently store information and instructions.

[0104] Computing system 1200 can be coupled via the data bus 1205 to one or more output devices 1235, such as speakers or displays (e.g., liquid crystal display or active matrix display) for displaying or providing information to a user. Input devices 1230, such as keyboards, touch screens or voice interfaces, can be coupled to the data bus 1205 for communicating information and commands to the processor(s) 1210. Input device 1230 can include, for example, a touchscreen display (e.g., output device 1235). Input device 1230 can include a cursor control, such as a mouse, a trackball, or cursor direction keys, for communicating direction information and command selections to the processor(s) 1210 for controlling cursor movement on a display.

[0105] The processes, systems and methods described herein can be implemented by the computing system 1200 in response to the processor 1210 executing an arrangement of instructions contained in main memory 1215. Such instructions can be read into main memory 1215 from another computer-readable medium, such as the storage device 1225. Execution of the arrangement of instructions contained in main memory 1215 causes the computing system 1200 to perform the illustrative processes described herein. One or more processors 1210 in a multiprocessing arrangement can also be employed to execute the instructions contained in main memory 1215. Hard-wired circuitry can be used in place of or in combination with software instructions together with the systems and methods described herein. Systems and methods described herein are not limited to any specific combination of hardware circuitry and software.

[0106] Although an example computing system has been described in FIG. 12, the subject matter, including the operations described in this specification, can be implemented in other types of digital electronic circuitry, or in computer software, firmware, or hardware, including the structures disclosed in this specification and their structural equivalents, or in combinations of one or more of them.

[0017] The foregoing examples have been provided merely for the purpose of explanation and are in no way to be construed as limiting of the present disclosure. While aspects of the present disclosure have been described with reference to an exemplary embodiment, it is understood that the words which have been used herein are words of description and illustration, rather than words of limitation. Changes can be made, within the purview of the appended claims, as presently stated and as amended, without departing from the scope and spirit of the present disclosure in its aspects. Although aspects of the present disclosure have been described herein with reference to particular means, materials and embodiments, the present disclosure is not intended to be limited to theparticulars disclosed herein; rather, the present disclosure extends to all functionally equivalent structures, methods and uses, such as are within the scope of the appended claims.

[0108] The subject matter and the operations described in this specification can be implemented in digital electronic circuitry, or in computer software, firmware, or hardware, including the structures disclosed in this specification and their structural equivalents, or in combinations of one or more of them. The subject matter described in this specification can be implemented as one or more computer programs, e.g., one or more circuits of computer program instructions, encoded on one or more computer storage media for execution by, or to control the operation of, data processing apparatuses. Alternatively or in addition, the program instructions can be encoded on an artificially generated propagated signal, e.g., a machine-generated electrical, optical, or electromagnetic signal that is generated to encode information for transmission to suitable receiver apparatus for execution by a data processing apparatus. A computer storage medium can be, or be included in, a computer-readable storage device, a computer-readable storage substrate, a random or serial access memory array or device, or a combination of one or more of them. While a computer storage medium is not a propagated signal, a computer storage medium can be a source or destination of computer program instructions encoded in an artificially generated propagated signal. The computer storage medium can also be, or be included in, one or more separate components or media (e.g., multiple CDs, disks, or other storage devices include cloud storage). The operations described in this specification can be implemented as operations performed by a data processing apparatus on data stored on one or more computer-readable storage devices or received from other sources.

[0109] The terms “computing device,” “component” or “data processing apparatus” or the like encompass various apparatuses, devices, and machines for processing data, including, by way of example, a programmable processor, a computer, a system on a chip, or multiple ones, or combinations of the foregoing. The apparatus can include special purpose logic circuitry, e.g., an FPGA (field programmable gate array) or an ASIC (application specific integrated circuit). The apparatus can also include, in addition to hardware, code that creates an execution environment for the computer program in question, e.g., code that constitutes processor firmware, a protocolstack, a database management system, an operating system, a cross-platform runtime environment, a virtual machine, or a combination of one or more of them. The apparatus and execution environment can realize various different computing model infrastructures, such as web services, distributed computing and grid computing infrastructures.A computer program (also known as a program, software, software application, app, script, or code) can be written in any form of programming language, including compiled or interpreted languages, declarative or procedural languages, and can be deployed in any form, including as a stand-alone program or as a module, component, subroutine, object, or other unit suitable for use in a computing environment. A computer program can correspond to a file in a file system. A computer program can be stored in a portion 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 multiple coordinated files (e.g., files that store one or more modules, sub programs, or portions of code). A computer program can be deployed to be executed on one computer or on multiple computers that are located at one site or distributed across multiple sites and interconnected by a communication network.10111] The processes and logic flows described in this specification can be performed by one or more programmable processors executing one or more computer programs to perform actions by operating on input data and generating output. The processes and logic flows can also be performed by, and apparatuses can also be implemented as, special purpose logic circuitry, e.g., an FPGA (field programmable gate array) or an ASIC (application specific integrated circuit). Devices suitable for storing computer program instructions and data can include non-volatile memory, media and memory devices, including, by way of 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. The processor and the memory can be supplemented by, or incorporated in, special purpose logic circuitry.[Oil 2] The subj ect matter described herein can be implemented in a computing system that includes a back end component, e.g., as a data server, or that includes a middleware component, e.g., an application server, or that includes a front end component, e.g., a client computer having a graphical user interface or a web browser through which a user can interact with an implementation of the subject matter described in this specification, or a combination of one or more such back end, middleware, or front end components. The components of the system can be interconnected by any form or medium of digital data communication, e.g., a communication network. Examples of communication networks include a local area network (“LAN”) and a wide area network (“WAN”), an inter-network (e.g., the Internet), and peer-to-peer networks (e.g., ad hoc peer-to-peer networks).

[0113] While operations are depicted in the drawings in a particular order, such operations are not required to be performed in the particular order shown or in sequential order, and all illustrated operations are not required to be performed. Actions described herein can be performed in a different order.[01.1.4] Having now described some illustrative implementations, it is apparent that the foregoing is illustrative and not limiting, having been presented by way of example. In particular, although many of the examples presented herein involve specific combinations of method acts or system elements, those acts and those elements can be combined in other ways to accomplish the same objectives. Acts, elements and features discussed in connection with one implementation are not intended to be excluded from a similar role in other implementations.[01151 The phraseology and terminology used herein is for the purpose of description and should not be regarded as limiting. The use of “including,” “comprising,” “having,” “containing,” “involving,” “characterized by,” “characterized in that” and variations thereof herein, is meant to encompass the items listed thereafter, equivalents thereof, and additional items, as well as alternate implementations consisting of the items listed thereafter exclusively. In one implementation, the systems and methods described herein consist of one, each combination of more than one, or all of the described elements, acts, or components.[0116| Any references to implementations or elements or acts of the systems and methods herein referred to in the singular can also embrace implementations including a plurality of these elements, and any references in plural to any implementation or element or act herein can also embrace implementations including only a single element. References in the singular or plural form are not intended to limit the presently disclosed systems or methods, their components, acts, or elements to single or plural configurations. References to any act or element being based on any information, act or element can include implementations where the act or element is based at least in part on any information, act, or element.

[0117] Any implementation disclosed herein can be combined with any other implementation or embodiment, and references to “an implementation,” “some implementations,” “one implementation” or the like are not necessarily mutually exclusive and are intended to indicate that a particular feature, structure, or characteristic described in connection with the implementation can be included in at least one implementation or embodiment. Such terms as used herein are not necessarily all referring to the same implementation. Any implementation can be combined with any other implementation, inclusively or exclusively, in any manner consistent with the aspects and implementations disclosed herein.10118] References to “or” can be construed as inclusive so that any terms described using “or” can indicate any of a single, more than one, and all of the described terms. References to at least one of a conjunctive list of terms can be construed as an inclusive OR to indicate any of a single, more than one, and all of the described terms. For example, a reference to “at least one of ‘A’ and ‘B’” can include only ‘A’, only ‘B’, as well as both ‘A’ and ‘B’. Such references used in conjunction with “comprising” or other open terminology can include additional items.

[0119] Where technical features in the drawings, detailed description or any claim are followed by reference signs, the reference signs have been included to increase the intelligibility of the drawings, detailed description, and claims. Accordingly, neither the reference signs nor their absence have any limiting effect on the scope of any claim elements.[0I20| Modifications of described elements and acts such as substitutions, changes and omissions can be made in the design, operating conditions and arrangement of the disclosed elements and operations without departing from the scope of the present disclosure.

Claims

WHAT IS CLAIMED IS:

1. A system, comprising: one or more processors, coupled with memory, to: receive, from a client device, a textual input via a chatbot interface, the textual input indicative of an update to a control parameter; generate, using a large language model, a search query based on the textual input; filter, using the search query, data related to control parameters stored in a database; detect, using the large language model, updates to the control parameters based on the filtered data; identify, based on the updates to the control parameters and product metadata, one or more products that fail to comply with the updates; determine code segments within a source code corresponding to the one or more products, the code segments related to the control parameters; modify the code segments to comply with the updates to the control parameters; and execute the modified code segments to perform an operation.

2. The system of claim 1, wherein the control parameters correspond to a set of rules defining operational standards for the one or more products.

3. The system of claim 1, wherein the one or more processors are further configured to: input the textual input into the large language model to cause the large language model to classify the textual input into one or more search terms; and generate, using the large language model, the search query based on the one or more search terms.

4. The system of claim 1, wherein the large language model is trained on a dataset comprising at least one of regulatory requirements, industry standards, or legislative directives.

5. The system of claim 1, wherein the one or more processors are further configured to: identify, responsive to comparing the product metadata of the one or more products to compliance thresholds defined by the updates to the control parameters, that the one or more products fail to comply with the updates to the control parameters.

6. The system of claim 1, wherein the product metadata includes information related to product configurations, operational settings, usage patterns, or deployment environments.

7. The system of claim 1, wherein the one or more processors are further configured to: determine the code segments within the source code responsive to mapping the control parameters to the code segments implementing functionalities related to the control parameters.

8. The system of claim 1, wherein the one or more processors are further configured to: execute the large language model, trained on a dataset of control parameters, to determine the code segments within the source code.

9. The system of claim 1, wherein the one or more processors are further configured to: generate code modifications to the code segments based on the updates to the control parameters; and execute the generated code modifications to comply with the updates to the control parameters.

10. The system of claim 1, wherein the one or more processors are further configured to: input the modified code segments and the updates to the control parameters into the large language model to cause the large language model to generate a data structure; andparse the data structure to display, via the client device, an electronic report indicating compliance of the one or more products with the updates to the control parameters.

11. A method, comprising: generating, using a large language model, a search query based on an input received via a chatbot interface indicative of an update to a control parameter; filtering, using the search query, data related to control parameters stored in a database; detecting, using the large language model, updates to the control parameters based on the filtered data; identifying, based on the updates to the control parameters and product metadata, one or more products that fail to comply with the updates; determining code segments within a source code corresponding to the one or more products, the code segments related to the control parameters; modifying the code segments to comply with the updates to the control parameters; and executing the modified code segments to perform an operation.

12. The method of claim 11, wherein the control parameters correspond to a set of rules defining operational standards for the one or more products.

13. The method of claim 11, further comprising: inputting the input into the large language model to cause the large language model to classify the input into one or more search terms; and generating, using the large language model, the search query based on the one or more search terms.

14. The method of claim 11, wherein the large language model is trained on a dataset comprising at least one of regulatory requirements, industry standards, or legislative directives.

15. The method of claim 11, further comprising:identifying, responsive to comparing the product metadata of the one or more products to compliance thresholds defined by the updates to the control parameters, that the one or more products fail to comply with the updates to the control parameters.

16. The method of claim 11, wherein the product metadata includes information related to product configurations, operational settings, usage patterns, or deployment environments.

17. The method of claim 11, further comprising: determining the code segments within the source code responsive to mapping the control parameters to the code segments implementing functionalities related to the control parameters.

18. The method of claim 11, further comprising: executing the large language model, trained on a dataset of control parameters, to determine the code segments within the source code.

19. The method of claim 11 , further comprising: generating code modifications to the code segments based on the updates to the control parameters; and executing the generated code modifications to comply with the updates to the control parameters.

20. A non-transitory computer readable medium including one or more instructions stored thereon and executable by a processor to: filter using a large language model, data related to control parameters used by one or more products, the control parameters indicated via a chatbot interface; detect, using the large language model, updates to the control parameters based on the filtered data; identify, based on the updates to the control parameters and product metadata, that the one or more products fail to comply with the updates to the control parameters;determine code segments within a source code corresponding to the one or more products, the code segments related to the control parameters; and modify the code segments to comply with the updates to the control parameters.

Images