A system and method for monitoring data input to a machine learning model.

Abstract

Systems and methods are provided to monitor data before it is input into a machine learning model. Generally, the systems and methods of this disclosure are designed to enable the secure use of machine learning modules within a virtual team environment. A chat module may be used to allow a user to control the use of one or more machine learning modules by entering commands. The chat module may be integrated into an existing user interface to add machine learning module functionality to the existing user interface. In some embodiments, a security module may monitor input data entered by a user into the chat module to prevent sensitive information from being disseminated to the machine learning module.

Description

Technical Field

[0001] Cross - reference to Related Applications This application claims priority to U.S. Patent Application No. 18 / 212,120, filed on June 20, 2023, and all of its applications are hereby incorporated by reference in their entirety into this specification.

[0002] Field of the Disclosure The subject matter of this disclosure generally relates to systems and methods for monitoring data before it is input into a machine learning model.

Background Art

[0003] Machine learning models have increasingly become important tools in enterprises due to their ability to enhance efficiency and improve user decision - making. Furthermore, the optimization of business plans and improved predictive maintenance for equipment are further enhancing business productivity. However, since machine learning models often collect data input by users to support the learning process, concerns have arisen regarding potential losses of the intellectual property of enterprises that use third - party machine learning models. Confidential information that once belonged to an enterprise may subsequently be generally redistributed, resulting in losses of the enterprise's intellectual property. The areas of intellectual property that are most concerning are the areas of trade secrets and / or patents. Trade secrets made public through machine learning models may be considered public property, and the public distribution of research / development of ideas not granted patent rights may result in a complete loss of the ability of the enterprise to obtain patents, or may even give competitors an advantage before a license agreement is arranged in the development around potentially patentable materials. Since they no longer have the competitiveness obtained by such intellectual property, the resulting revenue losses from such disclosures could lead to the economic collapse of the enterprise. Furthermore, such intellectual property losses can be purely accidental when an enterprise's employees attempt to use software for team development that incorporates a machine learning model to further enhance efficiency.

[0004] There are also legal issues that companies must consider when using machine learning models. Machine learning models often rely on large amounts of personal data, such as customer information or employee records. Compliance with data protection regulations can be difficult if companies are not careful, and it is crucial to obtain appropriate consent for the collection and use of this personal data to avoid large lawsuits and fines. Furthermore, machine learning models may integrate proprietary data or algorithms into any generated output, which can give rise to intellectual property issues different from those mentioned above. If companies are not careful to obtain appropriate licenses and permits for the use of any third-party data or algorithms, as well as sufficient protection to prevent infringement of their own intellectual property rights, costly lawsuits may follow. Moreover, machine learning models can be difficult to understand and interpret issues that arise in relation to transparency and accountability. Therefore, companies need to ensure that any machine learning model used to perform a particular task is transparent and understandable, and that they provide a clear explanation of how decisions are made. However, not all machine learning models mitigate these risks equally when given a particular task, making it difficult for companies to determine which machine learning model is best suited to their particular business.

[0005] Accordingly, there is a need in the art for systems and methods to monitor data fed into machine learning models, evaluate whether such data could result in data loss, and, if data loss is determined to have occurred, prevent such data from being transmitted to the machine learning model. Furthermore, there is a need in the art for methods to help companies monitor how their employees are using machine learning models and determine which machine learning models may be optimal for them. [Overview of the project] [Means for solving the problem]

[0006] A system and method are provided for monitoring data before it is input into a machine learning model. In one embodiment, the system allows the user to enter commands into a chat interface in such a way that the user can select which machine learning module will perform the desired task. In another embodiment, the system integrates machine learning capabilities into an existing user interface. In yet another embodiment, the system monitors data as it is entered into a chat interface to prevent sensitive information from being disseminated to a desired machine learning module. Generally, the systems and methods of this disclosure are designed to enable the secure use of machine learning modules in a virtual team environment. System 400 generally includes a computing entity 200 having a user interface 411, a security module 428, a machine learning module 425, a processor 220 operably connected to the computing entity 200, the security module 428, and the machine learning module 425, a display operably connected to the processor 220, and a persistent computer-readable medium connected to the processor 220 and having instructions stored thereon. Secondary security devices and authorization levels may be used to prevent unauthorized access to the system. [Effects of the Invention]

[0007] The above summary outlines some features of the systems and methods of this disclosure so that those skilled in the art may better understand the detailed description that follows. Additional features that form the subject matter of the claims are described below. Those skilled in the art should understand that these features can be readily utilized to design or modify other systems to perform the same purposes as those of the systems and methods disclosed herein. Those skilled in the art should also understand that such equivalent designs or modifications do not deviate from the scope of the systems and methods of this disclosure.

[0008] These and other features, aspects, and advantages of this disclosure will be better understood in relation to the following description, the attached claims, and the accompanying drawings. [Brief explanation of the drawing]

[0009] [Figure 1] This document presents a system that embodies features consistent with the principles of this disclosure. [Figure 2] This document presents a system that embodies features consistent with the principles of this disclosure. [Figure 3] This document presents a system that embodies features consistent with the principles of this disclosure. [Figure 4] This describes a system for managing data transfer between a computing device that has a chat module and a computing device that hosts a machine learning model. [Figure 5] This document describes a one-to-one chat user interface for managing data transfer between a computing device with a chat module and a computing device hosting a machine learning model. [Figure 6] This document describes a team user interface for managing data transfer between a computing device with a chat module and a computing device hosting a machine learning model. [Figure 7] This document describes a system used in an environment that embodies features consistent with the principles of this disclosure. [Figure 8] This diagram illustrates how individual access to data can be permitted or restricted based on the user's role and the administrator's role. [Figure 9] This flowchart illustrates a method step of a method that embodies features consistent with the principles of this disclosure. [Figure 10] This flowchart illustrates a method step of a method that embodies features consistent with the principles of this disclosure. [Figure 11]This flowchart illustrates a method step of a method that embodies features consistent with the principles of this disclosure. [Modes for carrying out the invention]

[0010] In the summary above, this detailed description, the claims below, and the accompanying drawings, certain features of the disclosure, including method steps, are referenced. It should be understood that the disclosure of the invention herein includes all possible combinations of such specific features. For example, where a particular feature is disclosed in the context of a particular aspect or embodiment of the invention or a particular claim, that feature may, as far as possible, be used in combination with and / or in the context of other particular aspects of the embodiments of the invention and in the invention in general.

[0011] The term “including” and its grammatical equivalent are used herein to mean the presence of other components, steps, etc., of optional choice. For example, a system “including” components A, B, and C may include only components A, B, and C, or it may include not only components A, B, and C, but also one or more other components. Where a method including two or more defined steps is referred to herein, the defined steps may be performed in any order or simultaneously (unless the context excludes that possibility), and the method may include one or more other steps performed before any of the defined steps, between two of the defined steps, or after all the defined steps (unless the context excludes that possibility). As will be apparent from the disclosures provided below, the present invention satisfies the need for a system and method that can reduce the amount of data transferred between computing devices. Hereinafter, the term “Security Information and Event Management (SIEM)” and its grammatical equivalent are used herein to mean a single security management system that includes both Security Information Management (SIM) and Security Event Management (SEM).

[0012] Figure 1 shows an exemplary environment 100 of a system 400 consisting of clients 105 connected to a server 110 and / or a database 115 via a network 150. A client 105 is a device of a user 405 that can be used to access the server 110 and / or database 115 through the network 150. The network 150 can consist of one or more networks of any kind, including but not limited to telephone networks, intranets, the Internet, memory devices, other types of networks, or combinations of networks, such as a local area network (LAN), wide area network (WAN), metropolitan area network (MAN), or public switched telephone network (PSTN). In a preferred embodiment, a computing entity 200 may function as a client 105 to a user 405. For example, a client 105 may include a personal computer, a wireless telephone, a streaming device, a "smart" television, a personal digital assistant (PDA), a laptop, a smartphone, a tablet computer, or another type of computing or communication interface 280. Server 110 may include devices for accessing, fetching, aggregating, processing, retrieving, serving, and / or maintaining documents. Figure 1 shows a preferred embodiment of Environment 100 for System 400, but in other embodiments, Environment 100 may include fewer components, different components, differently arranged components, and / or additional components than those shown in Figure 1. Alternatively, or additionally, one or more components of Environment 100 may perform one or more other tasks described as being performed by one or more other components of Environment 100.

[0013] As shown in Figure 1, one embodiment of the system 400 may include a server 110. Although shown as a single server 110 in Figure 1, in some embodiments, the server 110 may be implemented as multiple devices connected together via a network 150, distributed across a wide geographical area and performing different or similar functions. For example, two or more servers 110 may be implemented to operate as a single server 110 performing the same task. Alternatively, one server 110 may perform the functions of multiple servers 110. For example, a single server 110 may perform the tasks of a web server and an index server 110. Furthermore, it is understood that multiple servers 110 may be used to operably connect a processor 220 to a database 115 and / or other content repositories. The processor 220 may be operably connected to the server 110 via a wired or wireless connection. The types of servers 110 that may be used by the system 400 include, but are not limited to, search servers, document index servers, and web servers, or any combination thereof.

[0014] The search server may include one or more computing entities 200 designed to implement a search engine, such as a document / record search engine or a general web page search engine. The search server may include one or more web servers designed to receive, for example, search queries and / or inputs from a user 405, and to search one or more databases 115 in response to the search queries and / or inputs to provide the user 405 with documents or information related to the search queries and / or inputs. In some embodiments, the search server may include a web search server that can provide the user 405 with web pages, which may include references to web servers where the desired information and / or links reside. The references to web servers where the desired information reside may be included within frames and / or text boxes, or as links to the desired information and / or documents. The document index server may include one or more devices designed to index documents available through the network 150. The document index server may access other servers 110, such as web servers hosting content, to index the content. In some embodiments, a document index server may create an index of documents / records stored by other servers 110 connected to the network 150. The document index server may store and index, for example, content, information, and documents associated with user accounts, as well as user-generated content. A web server may include a server 110 that provides web pages to a client 105. For example, a web page may be an HTML-based web page. A web server may host one or more websites. In this specification, a website may refer to a collection of associated web pages. Often, a website may be associated with a single domain name, but some websites may potentially include two or more domain names. The concepts described herein may apply on a website basis. Alternatively, in some embodiments, the concepts described herein may apply on a web page basis.

[0015] In this specification, database 115 refers to a set of related data and the method of organizing it. Access to this data is typically provided by a database management system (DBMS), which consists of an integrated set of computer software that enables a user 405 to interact with one or more databases 115, providing access to all data contained within database 115. The DBMS provides a way of managing how that information is organized, by providing various functions that enable the input, storage, and retrieval of large amounts of information. Due to the close relationship between database 115 and the DBMS, as used herein, the term database 115 refers to both database 115 and the DBMS.

[0016] Figure 2 is an exemplary schematic diagram of a client 105, a server 110, and / or a database 115 (collectively referred to as the "Computing Entity 200"), which may correspond to one or more of the client 105, server 110, and database 115 according to one embodiment consistent with the principles of the present invention described herein. The Computing Entity 200 may include a bus 210, a processor 220, memory 304, storage device 250, peripheral devices 270, and a communication interface 280 (such as a wired or wireless communication device). The bus 210 may be defined as one or more conductors that allow communication between components of the Computing Entity 200. The processor 220 may be defined as logic circuits that process in response to basic instructions that operate the Computing Entity 200. The memory 304 may be defined as an integrated circuit that stores information for immediate use within the Computing Entity 200. The peripheral devices 270 may be defined as any hardware used by user 405 and / or the Computing Entity 200 to facilitate communication between the two. The storage device 250 may be defined as a device used to provide mass storage to the computing entity 200. The communication interface 280 may be defined as any transceiver-like device that enables the computing entity 200 to communicate with other devices and / or with the computing entity 200.

[0017] Bus 210 may include a high-speed interface 308 and / or a low-speed interface 312 that connect various components together in a manner that allows them to communicate with each other. The high-speed interface 308 manages operations that consume a large amount of bandwidth for the computing device 300, while the low-speed interface 312 manages operations that consume less bandwidth. In some preferred embodiments, the high-speed interface 308 of bus 210 may be coupled to a high-speed expansion port 310 that can receive various expansion cards such as memory 304, display 316, and a graphics processing unit (GPU). In other preferred embodiments, the low-speed interface 312 of bus 210 may be coupled to a storage device 250 and a low-speed expansion port 314. The low-speed expansion port 314 may include various communication ports such as USB, Bluetooth, Ethernet, wireless Ethernet, etc. Further, the low-speed expansion port 314 may be coupled to one or more peripheral devices 270 such as a keyboard, pointing device, scanner, and / or networking device, in which case the low-speed expansion port 314 facilitates the transfer of input data from the peripheral device 270 to the processor 220 via the low-speed interface 312.

[0018] The processor 220 may include any type of conventional processor or microprocessor that interprets and executes computer-readable instructions. The processor 220 is configured to perform the operations disclosed herein based on instructions stored within the system 400. The processor 220 may process instructions for execution within the computing entity 200, including instructions stored in memory 304 or on storage device 250, to display graphical information for a graphical user interface (GUI) on an external peripheral device 270, such as a display 316. The processor 220 may provide coordination of other components of the computing entity 200, such as control of the user interface 411, applications executed by the computing entity 200, and wireless communication via the communication interface 280 of the computing entity 200. The processor 220 may be any processor or microprocessor suitable for executing instructions. In some embodiments, the processor 220 may have a memory device in or coupled to it that is suitable for storing the data, content, or other information or material disclosed herein. In some cases, the processor 220 may be a component of a larger computing entity 200. The computing entity 200, which may house the processor 220, may include, but is not limited to, laptops, desktops, workstations, personal digital assistants, servers 110, mainframes, mobile phones, tablet computers, smart televisions, streaming devices, smartwatches, or any other similar devices. Accordingly, the subject matter of the invention disclosed herein may be implemented or utilized in whole or in part in devices that may include, but is not limited to, laptops, desktops, workstations, personal digital assistants, servers 110, mainframes, mobile phones, tablet computers, smart televisions, streaming devices, or any other similar devices.

[0019] Memory 304 stores information within computing device 300. In some preferred embodiments, memory 304 may include one or more volatile memory devices. In another preferred embodiment, memory 304 may include one or more non-volatile memory devices. Memory 304 may also include another form of computer-readable medium, such as magnetic, solid-state, or optical disks. For example, a portion of a magnetic hard drive may be partitioned as dynamic scratch space to enable temporary storage of information that may be used by processor 220 when a faster type of memory, such as random access memory (RAM), is required. The computer-readable medium may refer to a persistent computer-readable memory device. The memory device may refer to storage space within a single storage device 250 or scattered across multiple storage devices 250. Memory 304 may include main memory 230 and / or read-only memory (ROM) 240. In a preferred embodiment, main memory 230 may include RAM or another type of dynamic storage device 250 that stores information and instructions for execution by processor 220. ROM 240 may include a conventional ROM device or another type of static storage device 250 that stores static information and instructions for use by processor 220. Storage device 250 may include magnetic and / or optical recording media and their corresponding drives.

[0020] As described above, peripheral device 270 is a device that facilitates communication between user 405 and processor 220. Peripheral device 270 may include, but is not limited to, input devices and / or output devices. In this specification, an input device may be defined as a device that enables user 405 to input data and instructions, which are then converted into patterns of electrical signals in binary code understandable to computing entity 200. Input devices of peripheral device 270 may include one or more conventional devices that enable user 405 to input information to computing entity 200, such as a controller, scanner, telephone, camera, scanning device, keyboard, mouse, pen, voice recognition and / or biometric authentication mechanism. In this specification, an output device may be defined as a device that converts electrical signals received from computing entity 200 into a format understandable to user 405. Output devices of peripheral device 270 may include one or more conventional devices that output information to user 405, such as a display 316, printer, speaker, alarm, projector, etc. Additionally, storage devices 250, such as CD-ROM drives, and other computing entities 200 may function as peripheral devices 270 that can operate independently of the operationally connected computing entities 200. For example, a streaming device may transfer data to a smartphone, in which case the smartphone may use the data in a manner that separates it from the streaming device.

[0021] The storage device 250 can provide large-capacity storage to the computing entity 200. In some embodiments, the storage device 250 may include computer-readable media such as memory 304, storage device 250, or memory 304 on the processor 220. Computer-readable media may be defined as one or more physical or logical memory devices and / or carriers. Devices that can function as computer-readable media include, but are not limited to, arrays of devices, including hard disk drives, optical disk drives, tape drives, flash memory or other similar solid-state memory devices, or devices in a storage area network or other configuration. Examples of computer-readable media include, but are not limited to, magnetic media such as hard disks, floppy disks, and magnetic tapes; optical media such as CD-ROM disks and DVDs; magneto-optical media such as optical disks; and hardware devices specifically configured to store and execute programming instructions, such as ROM 240, RAM, flash memory, and similar.

[0022] In one embodiment, a computer program may be tangibly embodied in a storage device 250. The computer program may include instructions that, when executed by a processor 220, perform one or more steps, including methods such as those described herein. Instructions within the computer program may be transported to the processor 220 via a bus 210. Alternatively, the computer program may be transported to a computer-readable medium, in which case the information may then be accessed by the processor 220 via the bus 210 from the computer-readable medium as needed. In a preferred embodiment, software instructions may be read into memory 304 from another computer-readable medium, such as a data storage device 250, or from another device via a communication interface 280. Alternatively, hardware circuitry may be used instead of, or in combination with, software instructions to implement processes consistent with the principles described herein. Thus, embodiments consistent with the invention described herein are not limited to any particular combination of hardware circuitry and software.

[0023] Figure 3 shows exemplary computing entities 200 in the form of a computing device 300 and a mobile computing device 350, which can be used to perform various embodiments of the invention described herein. The computing device 300 is intended to represent various forms of digital computers, such as laptops, desktops, workstations, servers 110, databases 115, mainframes, and other suitable computers. The mobile computing device 350 is intended to represent various forms of mobile devices, such as scanners, scanning devices, personal digital assistants, mobile phones, smartphones, tablet computers, and other similar devices. The various components shown in Figure 3, and their connections, associations, and functions, are intended to be illustrative only and are not intended to limit the embodiments of the invention described herein. The computing device 300 can be implemented in several different forms, as shown in Figures 1 and 3. For example, the computing device 300 can be implemented as a server 110 or within a group of servers 110. The computing device 300 can also be implemented as part of a rack server system. In addition, the computing device 300 can be implemented as a personal computer, such as a desktop computer or a laptop computer. Alternatively, components from the computing device 300 may be combined with other components in a mobile device to create a mobile computing device 350. Each mobile computing device 350 may include one or more computing devices 300 and mobile devices, and the entire system may consist of multiple computing devices 300 and mobile devices communicating with each other, as shown by the mobile computing device 350 in Figure 3. A computing entity 200 consistent with the principles of the present invention described herein may perform certain receiving, communicating, generating, outputting, relating, and storing operations as needed to perform various methods described in more detail below.

[0024] In the embodiment shown in Figure 3, the computing device 300 may include a processor 220, memory 304, storage device 250, high-speed expansion port 310, low-speed expansion port 314, and a bus 210 that operably connects the processor 220, memory 304, storage device 250, high-speed expansion port 310, and low-speed expansion port 314. In one preferred embodiment, the bus 210 may include a high-speed interface 308 connecting the processor 220 to the memory 304 and high-speed expansion port 310, and a low-speed interface 312 connecting the low-speed expansion port 314 and storage device 250. Since each component is interconnected using the bus 210, they may be mounted on a common motherboard as shown in Figure 3, or in other ways as needed. The processor 220 may process instructions for execution within the computing device 300, including instructions stored in memory 304 or on storage device 250. Processing these instructions may cause the computing device 300 to display graphical information for a GUI on an output device, such as a display 316 coupled to the high-speed interface 308. In other embodiments, multiple processors and / or multiple buses may be used together with multiple memory devices and / or multiple types of memory, as needed. Furthermore, multiple computing devices may be connected, in which case each device provides some of the necessary operations.

[0025] The mobile computing device 350 may include a processor 220, memory 304, and peripherals 270 (among other components, a display 316, a communication interface 280, and a transceiver 368). The mobile computing device 350 may also include a storage device 250, such as a microdrive or other aforementioned storage device 250, to provide additional storage. Preferably, each component of the mobile computing device 350 is interconnected using a bus 210, which may allow some of the components of the mobile computing device 350 to be mounted on a common motherboard as shown in Figure 3, or in other ways as needed. In some embodiments, a computer program may be tangibly embodied within an information carrier. The computer program may include instructions that, when executed by the processor 220, perform one or more methods, such as those described herein. The information carrier is preferably a computer-readable medium, such as memory, additional memory 374, or memory 304 on the processor 220, such as ROM 240, which can be received via a transceiver or an external interface 362. The mobile computing device 350 can be implemented in several different forms, as shown in Figure 3. For example, the mobile computing device 350 can be implemented as part of a mobile phone, a smartphone, a personal digital assistant (PDCA), or other similar mobile device.

[0026] The processor 220 can execute instructions within the mobile computing device 350, including instructions stored in memory 304 and / or storage device 250. The processor 220 may be implemented as a chipset of chips that may include multiple separate analog and / or digital processors. The processor 220 may provide coordination of other components of the mobile computing device 350, such as control of the user interface 411, applications run by the mobile computing device 350, and wireless communication by the mobile computing device 350. The processor 220 of the mobile computing device 350 may communicate with the user 405 through a control interface 358 coupled to peripheral device 270 and a display interface 356 coupled to display 316. The display 316 of the mobile computing device 350 may include, but is not limited to, liquid crystal displays (LCDs), light-emitting diode (LED) displays, organic light-emitting diode (OLED) displays, and plasma displays (PDPs), holographic displays, augmented reality displays, virtual reality displays, or any combination thereof. The display interface 356 may include appropriate circuitry for displaying graphical and other information to the user 405 on the display 316. The control interface 358 may receive commands from the user 405 via peripheral devices 270 and convert those commands into computer-readable signals for the processor 220. In addition, an external interface 362 may be provided to communicate with the processor 220, which may enable near-field communication between the mobile computing device 350 and other devices. The external interface 362 may provide wired communication in some embodiments and wireless communication in other embodiments. In a preferred embodiment, multiple interfaces may be used within a single mobile computing device 350, as shown in Figure 3.

[0027] Memory 304 stores information within the mobile computing device 350. Devices that can function as memory 304 for the mobile computing device 350 include, but are not limited to, computer-readable media, volatile memory, and non-volatile memory. An expansion memory 374 is also provided and may be connected to the mobile computing device 350 through an expansion interface 372, which may include a single in-line memory module (SIM) card interface or a micro-secure digital (Micro-SD) card interface. The expansion memory 374 may include, but are not limited to, various types of flash memory and non-volatile random access memory (NVRAM). Such expansion memory 374 may provide extra storage space for the mobile computing device 350. In addition, the expansion memory 374 may store computer programs or other information that can be used by the mobile computing device 350. For example, the expansion memory 374 may have instructions stored thereon that, when executed by the processor 220, cause the mobile computing device 350 to perform the methods described herein. Furthermore, the additional memory 374 may have secure information stored thereon, and therefore, the additional memory 374 may be provided as a security module 428 for the mobile computing device 350, in which case the security module 428 may be programmed with instructions that permit the secure use of the mobile computing device 350. In addition, the additional memory 374, which has secure applications and secure information stored thereon, may allow a user 405 to place identification information on the additional memory 374 in a manner that cannot be hacked through the mobile computing device 350.

[0028] The mobile computing device 350 may communicate wirelessly through a communication interface 280, which may optionally include digital signal processing circuits. The communication interface 280 may provide communication under various modes or protocols, including, but not limited to, Global System Mobile Communications (GSM), Short Message Service (SMS), Enterprise Messaging System (EMS), Multimedia Messaging Service (MMS), Code Division Multiple Access (CDMA), Time Division Multiple Access (TDMA), Personal Digital Cellular (PDC), Wideband Code Division Multiple Access (WCDMA), IMT Multicarrier (CDMAX 0), and General-Purpose Packet Radio Service (GPRS), or any combination thereof. Such communication may occur, for example, through a transceiver 368. Short-range communication may occur using Bluetooth, Wi-Fi, or other such transceivers 368. In addition, a Global Positioning System (GPS) receiver module 370 may provide the mobile computing device 350 with additional navigation and location-related wireless data, which may be used as needed by the application running on the mobile computing device 350. Alternatively, the mobile computing device 350 may communicate audibly using an audio codec 360, which can receive audio information from the user 405 and convert the received audio information into a digital format that can be processed by the processor 220. The audio codec 360 may also generate audible sound for the user 405, for example, through a speaker in the headset of the mobile computing device 350. Such sound may include recorded audio such as voice calls and voice messages, and audio from music files. The sound may also include audio generated by applications running on the mobile computing device 350.

[0029] The power supply can be any power source that provides power to system 400. In a preferred embodiment, the primary power supply of the system is a fixed power source, such as a standard wall outlet. In one preferred embodiment, system 400 may include multiple power supplies that can supply power to system 400 under different circumstances. For example, system 400 may be coupled to a backup battery system that can supply power to system 400 only when its primary power supply is unable to supply power and the batteries in the backup battery system are charged. In this way, system 400 can receive power even when conventional power sources are not operating, enabling the user to use the system and thus the system can re-examine input data to prevent the leakage of sensitive data.

[0030] Figures 4–11 illustrate embodiments of a system 400 and methods for monitoring data before inputting it into a machine learning model to ensure that information contained within the input data 430B is not sensitive data to individuals and / or organizations. Figure 4 shows a preferred embodiment of the system 400 having a computing entity 200 with a user interface 411, a security module 428, and a machine learning module 425, all interconnected in an operable manner. Figure 5 shows an example of the user interface 411 of the computing entity 200. Figure 6 shows an example of the user interface 411 of the computing entity 200 and a report generated by the system 400. Figure 7 shows an example of the user interface 411 of the computing entity 200 and how a user 405 may manage the machine learning module 425 of the system 400. Figure 8 shows permission levels that may be used by the system 400 to control access to user content such as user data 430A, input data 430B, and usage data 430C. Figures 9–11 illustrate methods that may be performed by the system 400. It is understood that the various method steps associated with the method disclosed herein can be performed as operations by the system 400 shown in Figure 4.

[0031] In general, a user may control the system 400 via a chat module 505 having a chat interface that can be integrated as an add-on user interface to an existing user interface 411 in some embodiments, the chat module 505 may be used to integrate additional features into the existing user interface 411 or, if not integrated into the existing user interface 411, to provide team chat functionality. In a preferred embodiment, the chat module 505 enables the integration of a machine learning module 425, such as a natural language processing (NLP) engine, into a team environment. In embodiments where the chat module 505 is integrated into the existing user interface 411, the system 400 may be used to integrate the machine learning module 425 into the existing user interface 411. In embodiments of the system 400 that include an existing user interface 411, the existing user interface 411 is preferably also the chat module 505. In some preferred embodiments, the chat module 505 may replace an information input field 505A of the existing user interface 411 and redirect the input data 430B entered therein to a security module 428 of the system 400. The security module 428 may analyze the input data 430B based on security rules and take action based on whether or not it violates the security rules.

[0032] When user 405 inputs input data 430B into information input field 505A and / or feeds input data 430B into information stream 505B requesting a task to be performed by the machine learning module 425 of system 400, processor 220 may transmit input data 430B to machine learning module 425, depending on the actions performed by security module 428. In one preferred embodiment, if system 400 determines that a security rule has been violated, it may advise user 405 on how it may modify input data 430B to bring it into compliance with system 400's security rules. In another preferred embodiment, system 400 may warn other users 405 if a security rule has been violated. In yet another preferred embodiment, system 400 may use machine learning techniques to analyze user engagement and generate a report on said engagement. In yet another preferred embodiment, system 400 may be configured to learn from user engagement with system 400 in order to facilitate more sincere and professional communication among users 405 and to help schedule communication among users 405 to improve organizational efficiency. In another preferred embodiment, the system 400 may be configured to use robotic process animation (RPA) to automate certain tasks that are commonly performed by the user 405 of the system 400.

[0033] System 400 generally includes a computing entity 200 having a user interface 411, a security module 428, a machine learning module 425, a processor 220 operably connected to the computing entity 200, the security module 428, and the machine learning module 425, a display operably connected to the processor 220, and a persistent computer-readable medium coupled to the processor 220 and having instructions stored thereon. It will be understood by those skilled in the art that the term computing entity 200 may be used to refer to a single or more computing entities that may host various features of System 400. In one preferred embodiment, a database 115 may be operably connected to the processor 220 and may store various data of System 400, including, but not limited to, user data 430A, input data 430B, and usage data 430C. In a preferred embodiment, various data of System 400 transferred between computing entities are encrypted. Other embodiments may further include a server operably connected to the processor 220 and the database 115 to facilitate the transfer of data between them. In another preferred embodiment, a wireless communication interface may allow various elements of the system 400 to receive and transmit various data among themselves.

[0034] As described above, the processor 220 is configured to perform the operations disclosed herein based on instructions stored within the system 400. In one embodiment, the programming instructions responsible for the operations performed by the processor 220 are stored on a persistent computer-readable medium ("CRM"), which may be coupled to a server as shown in Figure 4. Alternatively, the programming instructions may be stored within or included within the processor 220. Examples of persistent computer-readable media include, but are not limited to, magnetic media such as hard disks, floppy disks, and magnetic tapes; optical media such as CD-ROM disks and DVDs; magneto-optical media such as optical disks; and hardware devices specifically configured to store and execute programming instructions, such as read-only memory (ROM), random access memory (RAM), flash memory, and the like. In some embodiments, the programming instructions may be stored as modules within the persistent computer-readable medium.

[0035] Data within System 400 may be stored within various profiles. In a preferred embodiment, System 400 includes user data 430A, input data 430B, and usage data 430C, which may be stored within a user profile 430. User profile 430 may be defined as a profile containing data about a particular user 405. Hereinafter, user data 430A may be defined as personal information of user 405 that helps System 400 identify user 405. The types of data that System 400 may use as user data 430A include, but are not limited to, the user's name, username, social security number, telephone number, gender, age, address, telephone number, email address, data protected by HIPPA and / or GDPR privacy regulations, or any combination thereof. In some preferred embodiments, user data may also include authentication and security data, such as passwords and security questions. In other preferred embodiments, user data may include light data loss prevention settings, such as keyword blocklists, PII constraints, and FINRA. In yet another preferred embodiment, user data may include AI preferences that indicate to the system 400 which AI services the user wants to use to perform a task. User data may also include user interface aesthetic preferences that allow the user to change the appearance of the user interface 411, and notification preferences that tell the system 400 when the user wants to receive alerts.

[0036] In this specification, input data 430B is data entered into the user interface 411 by a user 405 of system 400. The types of data that may be used by system 400 as input data 430B include, but are not limited to, text data, image data, audio data, or any combination thereof. Image data may include a series of images ordered in a manner that creates a single image or video, which may or may not further include audio data. Usage data 430C may be defined as data related to each user 405's individual use of the machine learning module 425. User data 430A, input data 430B, and security rules combined with permission levels are preferably used by system 400 to help prevent the undesirable dissemination of sensitive information to the public and / or machine learning module 425. A user 405 is preferably associated with a specific user profile 430 based on their username. However, it is understood that a user 405 may be associated with a user profile 430 in a variety of ways without departing from the subject matter of the invention as herein.

[0037] In some preferred embodiments, the system 400 may categorize user profiles 430 into groups and subgroups (or user roles). In preferred embodiments, various groups and subgroups of the system 400 may grant users 405 permission levels that give them access to data within the system 400. For example, a user profile 430 for a company's regional manager may be given permission level 800, which allows the regional manager to manage security rules for all branches under their control, and this may allow the regional manager to set minimum security levels for users 405 within those groups, which may include multiple subgroups. A user profile 430 for a sub-user operating a company branch within a particular region of the regional manager may be given permission level 800, which grants the sub-user the ability to manage security rules within those particular subgroups, provided that it does not conflict with the security rules set by the regional manager. Thus, users 405 of the system 400 may modify the security rules that can be applied to other users 405 of the system 400 in accordance with the permission levels 800 of various users and sub-users.

[0038] As shown in Figure 4, the system 400 may include a database 115 operably connected to the processor 220. The database 115 may be operably connected to the processor 220 via a wired or wireless connection. In a preferred embodiment, the database 115 is configured to store user data 430A, input data 430B, and usage data 430C therein. Alternatively, the user data 430A, input data 430B, and usage data 430C may be stored on a persistent computer-readable medium. The database 115 may be a relational database, and therefore the user data 430A, input data 430B, and usage data 430C associated with each user profile 430 in a plurality of user profiles 430 may be stored at least partially in one or more tables. Alternatively, the database 115 may be an object database, and therefore the user data 430A, input data 430B, and usage data 430C associated with each user profile 430 in a plurality of user profiles 430 may be stored at least partially as objects. In some cases, the database 115 may include a relational and / or object database, as well as a server dedicated to managing user data 430A, input data 430B, and usage data 430C in the manner disclosed herein.

[0039] Information presented via a display may be called a soft copy of information, as it exists electronically and is presented temporarily. Information stored on a persistent computer-readable medium may be called a hard copy of information. For example, a display may present a soft copy of the visual representation of image data via a liquid crystal display (LCD), in which case the hard copy of the image data may be stored on a local hard drive. For example, a display may present a soft copy of audio information via a speaker, in which case the hard copy of the audio information is stored in the memory of the mobile computing device. For example, a display may present a soft copy of user data 430A via a hologram, in which case the hard copy of user data 430A is stored in the database 115. Displays include, but are not limited to, cathode ray tube monitors, LCD monitors, light-emitting diode (LED) monitors, gas plasma monitors, screen readers, speech synthesizers, holographic displays, speakers, and fragrance generators, or any combination thereof.

[0040] The user interface 411 can be defined as a space where interaction between user 405 and system 400 can occur. In one embodiment, the interaction may occur in a manner that allows user 405 to control the operation of system 400. The user interface 411 may include, but is not limited to, an operating system, a command-line user interface, an interactive interface, a web-based user interface, a zooming user interface, a touchscreen, a task-based user interface, a touch user interface, a text-based user interface, an intelligent user interface, a brain-computer interface (BCI), and a graphical user interface, or any combination thereof. System 400 may present data from the user interface 411 to user 405 via a display operably connected to processor 220. The display may be defined as an output device that transmits data, which may include, but is not limited to, visual, auditory, tactile, kinesthetic, olfactory, and gustatory, or any combination thereof.

[0041] Devices that can function as a communication interface include, but are not limited to, near-field communication (NFC), Bluetooth, infrared (IR), radio frequency communication (RFC), radio-frequency identification (RFID), and ANT+, or any combination thereof. In one embodiment, a communication interface may broadcast two or more types of signals. For example, a communication interface including an IR transmitter and an RFID transmitter may broadcast IR signals and RFID signals. Alternatively, a communication interface may broadcast signals of only one type. For example, an ID badge may be fitted with a communication interface that broadcasts only an NFC signal containing a unique ID that needs to be received by a computing entity equipped with an NFC receiver before it is permitted to disseminate the information to the machine learning module 425.

[0042] In one preferred embodiment, the system 400 may further include secondary security devices such as a biometric scanner, a camera configured to collect image data for facial recognition, or an ID badge having a unique identifier. In one preferred embodiment, the secondary security device may be operably connected to the computing entity 200 in such a way that it communicates directly with the computing entity 200 and not with other computing entities 200. For example, the secondary security device may be connected to the company's computing entity 200, so that user 405 must scan their thumbprint and / or face with biometric features before starting the computing entity 200. This may serve as an additional precaution used to prevent the unintentional sharing of protected information, such as intellectual property and sensitive user data. The computing entity 200, servers, databases, and secondary security devices may be connected via wired or wireless connections.

[0043] In another preferred embodiment, the secondary security device may include a transmitter having a unique ID, which may be transmitted to the computing entity 200 in the form of a computer-readable signal before the processor 220 decides whether access to the system 400 is permitted. The unique ID contained in the signal broadcast by the transmitter may include, but is not limited to, a unique identifier code, a social security number, a PIN, etc. For example, a computer-readable signal broadcast by the secondary security device in the form of an ID badge may contain information that warns the system 400 that a particular user 405 is within range of a particular computing device, which may cause the system 400 to automatically start the particular computing device. Alternatively, the system 400 may be configured to prevent the computing entity 200 from starting if a particular user 405 is within range of the system 400. If a user 405 without the appropriate authorization level is within range of the system 400, the system 400 will not start. For example, a confidential research area having a system 400 locally installed on a computer may disable certain functions of the chat module 505 if an outsider is within a predefined range within a computing device hosting the chat module 505, due to the outsider's ID badge sending a computer-readable signal to the computing device that disables the functions of the chat module 505.

[0044] Devices capable of functioning as transmitters include, but are not limited to, near-field communication (NFC), Bluetooth, infrared (IR), radio frequency communication (RFC), radio-frequency identification (RFID), and ANT+, or any combination thereof. In one embodiment, a transmitter may broadcast two or more types of signals. For example, a transmitter including an IR transmitter and an RFID transmitter may broadcast IR signals and RFID signals. Alternatively, a transmitter may broadcast only one type of signal. For example, an ID badge may be fitted with a transmitter that broadcasts only an NFC signal containing a unique ID that needs to be received by a computing device equipped with an NFC receiver before being activated by user 405.

[0045] User 405 preferably inputs and accesses data in System 400 by entering commands / tasks within the user interface 411 of the computing entity 200. In a preferred embodiment, as shown in Figures 7 and 8, User 405 may access data in System 400 by using the user interface 411 of the computing entity 200 to log in to User Profile 430, which has permission level 800 that allows User 405 to input and / or access User Data 430A, Input Data 430B, and Usage Data 430C of User Profile 430. After logging in to User Profile 430 via User Interface 411, User 405 may input Data 430B into the chat module 505 of User Interface 411, thereby allowing the security module 428 to determine the security status of Input Data 430B before sending it to the machine learning module 425. Some preferred embodiments may require further secondary security measures before allowing the transfer of Input Data 430B to the machine learning module 425. For example, system 400 may require biometric authentication before allowing user 405 to disseminate information to machine learning module 425 via user interface 411, but will not require biometric authentication if that information is already shared with the designated user 405 of system 400.

[0046] In a preferred embodiment, user 405 may select a signature within the user interface 411 to access the chat module 505, such as an image or a “new chat button”. At least one NLP engine is used by the system 400 to interpret commands entered into the chat module 505. User 405 may change which NLP engine is used within the chat module 505 by modifying user data 430A in their user profile 430, as shown in Figure 7. In a preferred embodiment, user 405 may modify user data 430A contained within their user profile 430 via the “User Preferences” window in the user interface 411. Once user 405 begins sending commands, content, and messages to the NLP engine via the chat module 505, the system 400 will create a chat catalog from any input data 430B provided by user 405 and any task data provided by various machine learning modules 425 used by the chat module 505. In a preferred embodiment, user 405 may navigate the chat catalog to view past engagements and resulting task data. In some preferred embodiments, the system 400 may parse the chat catalog into various data components, such as text strings, image files, document files, video files, audio files, or any combination thereof. In another preferred embodiment of the system 400, a user 405 may delete or archive chat history data if they have the appropriate permission level 800.

[0047] As described above, system 400 may react to input data entered into the chat module 505 before providing the input data to the AI ​​engine. In a preferred embodiment, user data 430A may cause system 400 to use different machine learning modules 425 in response to commands entered by user 405. For example, if system 400 determines that sending the input data could result in a point-of-sale (PoS) system exceeding a maximum cost threshold set by user 405, system 400 may prevent the input data from being sent to machine learning module 425. Alternatively, user 405 may enter a command causing system 400 to override the maximum cost threshold so that user 405 can access a preferred machine learning module 425 if needed. In another preferred embodiment, system 400 may manage how input data and task data are exchanged between different machine learning modules 425 based on user input. For example, user 405 may enter a command causing system 400 to send first task data, created by the first AI engine, to a second AI engine to generate second task data. For example, user 405 may use a first machine learning module 425 to create a specific image file from text and then input a command to enlarge the specific image file using a second machine learning module 425. For example, user 405 may instruct system 400 to ask an NLP engine to create a 60-second speech and then feed the resulting task data into a second AI engine that converts text into an audio file. In some embodiments, system 400 may ask user 405 to approve the first task data before sending it to the second machine learning module 425. The task data created by system 400's machine learning module 425 is preferably stored in a CRM and / or database.

[0048] In yet another preferred embodiment, system 400 may be used to create new multimedia by inputting commands and descriptions of what is desired using multiple AI modules. For example, user 405 may input commands that instruct system 400 to create a meme and a description of what user 405 wants the meme to represent. System 400 would then use an NLP module, such as ChatGPT, to generate language for that meme, and subsequently instruct an AI image module, such as Midjourney, to create the meme using the description provided by user 405 and the language generated by the NLP module. For example, user 405 may instruct system 400 to generate a deepfake video and provide a description of what user 405 wants the deepfake video to convey. System 400 would then use an NLP module to create a script for the deepfake video based on the provided description, and subsequently use murf Al to convert the script into an audio file. Using the script and audio files, a video could be created using reface, which would then be combined with the audio files to create the final deepfake video.

[0049] In another preferred embodiment, input data may be input to system 400 by user 405 to create weighted task data, in which case the weighted task data is weighted responses from at least two machine learning modules 425. For example, user 405 may input a command within chat module 505 to cause system 400 to ask several machine learning modules 425 about the historical timeline of World War I. System 400 may then use all of the task data provided by the several machine learning modules 425 to provide user 405 with a weighted response. In some preferred embodiments, system 400 may use machine learning modules 425 to create weighted task data. For example, user 405 may input a command to cause system 400 to ask five different NLP engines to write a paper on ransomware attacks. User 405 may then ask system 400 to create a final paper using the task data provided by the NLP engines, which may cause system 400 to input the task data from the five different NLP engines into another machine learning module 425 that compiles that task data into a final paper. As described above, user 405 may automate this process so that system 400 automatically sends first task data from multiple machine learning modules 425 to another machine learning module 425 in order to create weighted task data. In one preferred embodiment, one of the machine learning modules 425 used to create the first task data may also be used to create weighted task data. For example, a first NLP engine, a second NLP engine, and a third NLP engine may be used to create the first task data, and the second NLP engine may be used again to combine the first task data generated by the three NLP engines into weighted task data.

[0050] In some embodiments, user 405 may create custom commands that cause system 400 to function in a certain way when the custom commands are entered into chat module 505. In one preferred embodiment, this can be achieved by including a hash symbol / hashtag string within the input data entered into chat module 505. For example, if user 405 wants both the default NLP engine and a non-default NLP engine to create task data for a particular command, user 405 may add #[non-default NLP engine name] to their input data to cause system 400 to request task data from both the default and non-default NLP engines. For example, user 405 may enter input data to request a 1,000-word story about a family of bears in the woods and add #[non-default NLP engine name (storyboard)] to cause the image AI engine to create an image storyboard to support that story.

[0051] As described above, the chat module 505 allows user 405 to input input data 430B into the information input field 505A, which is then monitored by the security module 428 of system 400 to ensure that the input data 430B does not violate selected security rules. Additional actions that system 400 may perform include, but are not limited to, automating common tasks, including scheduling meetings between users 405, tracking user 405's calendar, analyzing communication patterns, analyzing the tone and sentiment of communications, usage analysis, billing analysis, and data entry and file management. If the input data 430B is to be analyzed for security violations, system 400 sends the input data 430B to the security module 428, where it is analyzed. In one preferred embodiment, the computing entity 200 hosts an existing user interface 411 and a chat module 505 that integrates the functionality of system 400 described herein into the existing user interface 411. A separate computing entity 200 preferably hosts a security module 428 to which the chat user interface 411 redirects input data 430B. If the security module 428 determines that the input data 430B does not pose a security threat, the processor 220 may send the input data 430B to the machine learning module 425. The machine learning module 425 is preferably hosted on a separate computing entity from the chat module 505 and the security module 428, but it will be understood by those skilled in the art that the chat module 505, the machine learning module 425, and the security module 428 may exist in any combination of computing entities without departing from the subject matter of the invention described herein.

[0052] The security module 428 preferably includes a set of security rules that the security module 428 uses to inspect input data 430B, which is at least one of the input data entered into the information input field 505A or fed into the information stream 505B of the chat module 505. These security rules may include, but are not limited to, keywords, keyword and phrase strings, regular expression patterns, voice pattern files, document templates that function as "digital fingerprints," pixel evaluation for image filtering, state laws, and federal laws. Special security rules may be available for enterprise users to enable and configure. Some examples of these special security rules are, but are not limited to, inappropriate words and phrases (e.g., adult, explicit, hate), FINRA, PII, HIPAA, file types, and metadata tags, or any combination thereof. In one preferred embodiment, a user 405 with an appropriate authorization level 800 may create special security rules that apply only to users 405 within a certain group of the system 400. These special security rules may include certain terms and / or strings of terms that users 405 are prohibited from entering as input data 430B to the system 400 for transmission to the machine learning module 425 of the system 400. For example, an administrator may create several security rules regarding terms related to the intellectual property of the company in which they are employed to prevent users 405 from disclosing the intellectual property to the public or third-party machine learning module 425. In another preferred embodiment, the system 400 may allow users 405 with an appropriate authorization level 800 to select security rules that apply to a particular group of users. For example, a first manager of a company may modify security rules to apply to users 405 in the first manager's group who have strict tone requirements, while a second manager within the same company may modify security rules to apply to users 405 in the second manager's group who have zero or low tone requirements.Accordingly, the System 400 of this Disclosure may include a set of pre-configured security rules that a user 405 with an appropriate authorization level may select in addition to creating customized security rules.

[0053] In some preferred embodiments, the system 400 may use one or more machine learning techniques to determine whether it violates security rules of the security module 428. For example, the system 400 may use a combination of natural language processing and reinforcement learning to determine what is represented in the input data 430B that is entered into the information input field 505A. The system 400 may then use this insight into the meaning of the input data 430B to compare the input data 430B to rules or regulations and then determine whether the input data 430B potentially violates security rules. For example, the system 400 may use machine learning techniques to analyze patterns in the input data 430B and provide suggestions to improve communication between users 405 or with other machine learning modules 425. For example, the system 400 may be configured to analyze input data 430B directed to ChatGPT instructing ChatGPT to perform a specific task and then recommend an alternative way to communicate the specific task to ChatGPT.

[0054] In another preferred embodiment, system 400 may use two or more machine learning techniques to facilitate a more efficient work environment based on security rules. For example, system 400 may use Delv Al to summarize the frequency and content of communication between team members and use a machine learning module 425 such as Grammarly to suggest that some members communicate more frequently, rarely, more carefully, and more task-related. In yet another preferred embodiment, system 400 may analyze data over a period of time using one or more machine learning modules 425 to create a ready-made report that provides a summary of a point in time. In a preferred embodiment, at least two machine learning modules 425 are used to create the ready-made report. For example, system 400 may use Google Bard and LLaMA to analyze data to generate a list of the most frequently asked questions by customers and a FAQ that can be answered. Furthermore, system 400 may use an image AI generation module such as MidJoumey to create images that can be used to help answer questions within the FAQ.

[0055] In another preferred embodiment, System 400 may automate certain tasks using RPA. Tasks that can be automated in this manner include, but are not limited to, scheduling group meetings, recording minutes of group meetings, revising previous minutes, updating agendas, updating calendars, file management, and user analysis. For example, System 400 may be configured to automatically generate a monthly report on user usage of System 400, in which case the report would include data quantifying information consumed, collected, and generated by System 400's machine learning module 425. For example, System 400 may be configured to automatically schedule group meetings for user groups using Sidekick Al and to record minutes of those group meetings using a transcriber program configured to convert speech to text, such as Otter, which System 400 may later use as input data 430B. For example, based on the input data 430B of the minutes, the system 400 may automatically perform tasks such as scheduling future group meetings, revising previous minutes, or updating user calendars by analyzing the input data 430B of the minutes with respect to terms that may indicate such tasks that need to be performed.

[0056] In some preferred embodiments, system 400 may employ machine learning techniques to assist in automating certain tasks. For example, system 400 may use deep learning to recognize patterns that could enable system 400 to automatically schedule group meetings by determining the booking status of each user 405 in the user group. In yet another preferred embodiment, system 400 may use machine learning techniques to automatically assess the compatibility between users 405 in a group and warn the group administrator if system 400 determines that there is a potential conflict between group members that is or could be a violation of security rules. For example, if two users 405 in a group use forbidden language toward each other more times than system 400's minimum threshold, processor 220 may send a message to the group manager to warn them about the potential conflict. RPA can also be used to automatically translate content or transcripts into other languages ​​preferred by users 405. For example, system 400 may configure a virtual classroom setting such that the instructor's remarks are automatically translated into the preferred language of each student within that virtual classroom setting. In some preferred embodiments, the system 400 may be configured to automatically detect misleading or false content in order to prevent the spread of misleading / false information within the group. For example, a political science research group may use automated deepfake detection to prevent members from sharing content that has been determined to be a deepfake.

[0057] In another preferred embodiment, system 400 may be used to assist user 405 in analyzing SIEM data to further enhance security for the organization. For example, user 405 may configure system 400 in such a way that it supplies SIEM data to an NLP module, which in turn can provide advice or suggestions to the organization that may enhance data security. In another preferred embodiment, system 400 may be configured to analyze SIEM data and alert administrators if unusual events occur. For example, user 405 may configure system 400 to provide SIEM data in real time to a machine learning module 425 configured to detect anomalous user behavior. If user 405 downloads several files in addition to those normally downloaded within a given period, system 400 may send a computer-readable signal to security personnel that user 405 is behaving unusually by downloading a larger amount of data than usual. In another preferred embodiment, system 400 may be configured to generate security reports using SIEM data, which may be used to enhance data security. For example, system 400 could use Power BI to generate a report using Azure Sentinel data dumps, which could be used to inform the company of its most troublesome data security issues and why they should be resolved. For example, system 400 could allow user 405 to input commands to turn Splunk SIEM data into a security report, using a second machine learning module 425 to notify the company's security department of any unusual behavior by various users 405 of system 400. In some preferred embodiments, system 400 could be used to generate a report rating the data breach threats that various users 405 of system 400 might cause.In a preferred embodiment, the report generated by the system 400 enables the company to better understand what user 405 is seeking help with and to determine what types of data are being shared with third parties.

[0058] In another preferred embodiment of System 400, the chat module of System 400 may be configured to provide real-time assistance to a user 405 of System 400. The chatbot is preferably configured to answer common questions and provide advice on how to use the functions of System 400 more efficiently. For example, if user 405 is having trouble sharing files within Microsoft Teams, the chatbot may provide step-by-step instructions on how to share files. The chatbot module may also be used to integrate a machine learning module 425 into an existing chat application. For example, user 405 may integrate the chatbot module into an existing chat application and use an indicator 515 in an input field 505A of the existing chat application to indicate that user 405 wants a particular machine learning module 425 to perform a task. The chatbot module may then ask the machine learning module 425 to perform the indicated task and return the resulting task data to user 405 via an information stream 505B of the existing chat application. In addition, the chatbot module can be used to enable system 400 to perform other functions of system 400, including but not limited to data loss prevention, RPA, and analytics.

[0059] In a preferred embodiment, input data 430B is entered by user 405 into information input field 505A of user interface 411 via an input device of the user's computing entity 200. User 405 may then provide user interface 411 with a command requesting that the input data 430B entered into information input field 505A be placed in information stream 505B of user interface 411, in which case user interface 411 will then cause the user's computing entity 200 to send the input data 430B to the computing entity 200 hosting the security module 428, so that the input data 430B can be examined by the security module 428 to determine whether user 405 is attempting to spread protected information in violation of any security rules of the security module 428. If the security module determines that the input data 430B does not violate any security rules, it may send the input data 430B to a computing entity hosting a machine learning module 425, where the machine learning module 425 can perform a task based on the input data 430B and then send the task results to the user's computing entity 200. The input data 430B submitted by user 405 to the information stream 505B via the user interface 411 is preferably stored in a data record, which can be accessed by system 400 in such a way that certain input data 430B may then be deleted from the data record. For example, the security rules of system 400 may be updated to include new constraints. The security module 428 may scan the data record to perform data loss prevention with respect to the data record and then use it to delete any problematic input data 430B in the data record, if any.

[0060] To prevent unauthorized user 405 from accessing information of other users 405, system 400 may employ security methods. As shown in Figure 8, the security methods of system 400 may include multiple permission levels 800, which may grant user 405 access to user content 815, 835, and 855 within system 400, while simultaneously denying user 405 without the appropriate permission level 800 the ability to view user content 815, 835, and 855. To access user content 815, 835, and 855 stored within system 400, user 405 may be required to make a request through the user interface. Access to the data within system 400 may be permitted or denied by processor 220 based on verification of the permission level 800 of the requesting user 805, 825, and 845. If the authorization level 800 for the requesting users 805, 825, and 845 is sufficient, the processor 220 may grant the requesting users 805, 825, and 845 access to the user content 815, 835, and 855 stored in the system 400. Conversely, if the authorization level 800 for the requesting users 805, 825, and 845 is insufficient, the processor 220 may deny the requesting users 805, 825, and 845 access to the user content 815, 835, and 855 stored in the system 400. In one embodiment, the authorization level 800 may be based on user roles 810, 830, and 850 and administrator role 870, as shown in Figure 8. User roles 810, 830, and 850 allow requesting users 805, 825, and 845 to access user content 815, 835, and 855 that user 405 has uploaded and / or otherwise obtained through the use of system 400. Administrator role 870 allows administrator 856 to access system-wide data.

[0061] In one embodiment, user roles 810, 830, and 850 can be assigned to users in such a way that requesting users 805, 825, and 845 can view a user profile 430 containing user data, input data, and usage data 430C via the user interface 411. To access data within the system 400, user 405 can make a user request to the processor 220 via the user interface 411. In one embodiment, the processor 220 may grant or deny the request based on the permission level 800 associated with the requesting users 805, 825, and 845. Only users 405 with the appropriate user role 810, 830, 850, or administrator role 870 can access the data within the user profile 430. For example, as shown in Figure 8, requesting user 1 805 has permission to view user 1's content 815 and user 2's content 835, while requesting user 2 825 only has permission to view user 2's content 835. Alternatively, user content 815, 835, and 855 may be restricted so that users can only view a limited amount of user content 815, 835, and 855. For example, a requesting user 3 845 may be granted permission level 800, which allows them to view only user 3's content 855 related to their use of a machine learning model, but does not allow other data that is determined to be user 3's content 855. In the example shown in Figure 8, administrator 865 grants user 405 a new permission level 800, enabling administrator 865 to grant user 405 a higher permission level 800 or a lower permission level 800. For example, administrator 865, who has administrator role 870, may grant other users a higher permission level 800 so that they can view user 3's content 855 and / or any other user 405's content 815, 835, and 855. Therefore, the authorization level 800 of system 400 can be assigned to user 405 in various ways without departing from the subject matter of the invention described herein.

[0062] Some preferred embodiments of System 400 may further include a point-of-sale (PoS) system which may be used to purchase access to various machine learning modules 425 of System 400. The user interface 411 of the computing entity 200 may be operably connected to a PoS that enables the purchase of access to the machine learning modules 425, which may then be integrated into a chat module 505. In a preferred embodiment, the machine learning modules 425 that may be integrated into the chat module 505 may be presented to the user 405 as stamps via the user interface 411 in list form, allowing the user 405 to select a stamp representing the desired machine learning module 425, as shown in Figure 7, although other methods may be used to represent the machine learning modules 425 without departing from the subject matter of the invention described herein. When a machine learning module 425 requiring payment is selected by the user 405 via the user interface 411, the PoS may automatically communicate with the computing device in a manner that enables the user 405 to take the necessary steps to access the desired machine learning module 425. Once the desired machine learning module 425 is purchased, the system 400 can then integrate that machine learning module 425 into the chat module 505.

[0063] In another preferred embodiment, the PoS may be used to allow user 405 to purchase additional features of system 400. For example, the RPA feature of system 400 may be locked to a free version of chat module 505. By using the PoS, user 405 may unlock the RPA feature by making a monthly payment. Once paid via the PoS, processor 220 may update user 405's permission level 800 to allow user 405 to access features of chat module 505 that they may not have been able to access previously. User 405 may access the PoS history invoice via user interface 411, which may be stored by system 400 as user data 430A. In a preferred embodiment, system 400 deletes invoices older than 36 months. User 405 may update payment information stored within their user profile 430 and used by the PoS via user interface 411. In some embodiments, a group of users may have a single payment method stored within the user profile 430 of the group of users, in which case the user 405 has the appropriate authorization level 800. For example, a global administrator of a business account containing multiple users 405 may pay for access to the entire group. If two or more global administrators are assigned to one group, when the first global administrator attempts to cancel a payment via PoS, the second global administrator must verify that the cancellation is necessary.

[0064] In a preferred embodiment, System 400 may present multiple billing models to User 405 before one or more of System 400's features are unlocked. Billing models that may be used by System 400 include, but are not limited to, per-user billing with pre-set limits, per-user billing with pre-set limits and allowable overruns backed by a budget, per-user billing where the company provides its own consumption API keys to each platform and is responsible for their use according to the DLP budget, per-user billing where the company uses API keys owned by System 400 and is therefore billed per call consumption model, per-user billing for archived users, private hosting consulting fees, private hosting customization work, per-user billing for privately hosted solutions, and support hours, or any combination thereof. Computing devices may be operablely connected to PoS via Bluetooth, Wi-Fi, or other such transceivers, but are not limited to these methods of communication.

[0065] Figure 9 provides a flowchart 900 illustrating a preferred method step that may be used to perform a method of checking input data against security rules. Step 905 indicates the start of the method. During step 910, the processor 220 may accept input data from an input device of the system 400, in which case the input data is entered into the system 400 by the user 405. During step 915, the processor 220 may query for security rules based on the user data of the user in the user profile and / or group data of a particular group of the user. In a preferred embodiment, the processor 220 may query persistent computer-readable media and / or databases for security rules related to the user data 430A and / or group data. Based on the results of the query, the processor 220 may perform an action during step 920. If the processor 220 determines that no security rules applicable to a particular user 405 are found, the system 400 may proceed to the method termination step 945. If the processor 220 determines that it has found a security rule applicable to a specific user 405, the system 400 may proceed to step 925.

[0066] During step 925, the processor 220 may retrieve security rules related to user 405 and / or the group. Once retrieved, the processor 220 may perform a query during step 930 to determine whether the security rules have been violated by user 405. In a preferred embodiment, the system 400 looks for prohibited terms and term strings within the security rules to determine whether the security rules have been violated. Based on the results of the query, the processor 220 may perform an action during step 935. If it is determined that the input data does not violate the security rules, the system 400 may proceed to the method termination step 945. If the processor 220 determines that the input data violates the security rules of the system 400, the processor 220 may, during step 940, prevent the input data from being taken in a way that could result in a security violation. In some preferred embodiments, the system 400 may also be configured to warn user 405 of the system 400 of the violation of the security rules. Once system 400 has prevented the problematic input data from being taken in a way that could result in a security breach, system 400 may proceed to method termination step 945.

[0067] Figure 10 provides a flowchart 1000 illustrating a preferred method step that may be used to implement a method for a chat module 505 to receive input data from an existing chat application and perform a desired task. Step 1005 marks the start of the method. During step 1010, the processor 220 may receive input data, including a chat module indicator 515, from a user 405 via an input device, in which case the input data, including the chat module indicator 515, is entered into an information input field 505A of an existing chat application. The processor 220 then transmits the input data to the chat module 505 during step 1015, and then, during step 1020, may determine which machine learning module 425 should be used based on the chat module indicator 515. Once the chat module 505 has received the input data, it may query to determine which task should be performed during step 1025. Based on the results of the query, the processor 220 may perform an action during step 1030. If a task cannot be determined based on the input data, the system 400 may, during step 1035, send a computer-readable signal to the information stream 505B of the chat module 505 to indicate that the task cannot be performed based on the input data, as illustrated in Figure 5, and then proceed to step 1055. If it is determined that the task can be performed, the processor 220 may send a computer-readable signal to the determined machine learning module 425 containing instructions for the task to be performed during step 1040. In some preferred embodiments, the chat module 505 may be configured to send the input data to the security module 428 to determine whether the input data violates security rules before sending the input data to the machine learning module 425.During step 1045, the processor 220 may send completed task data from the machine learning module 425 to the chat module 505, and then during step 1050, the completed task data may be sent from the chat module 505 to the information stream 505B of the existing chat application. Once the completed task data has been sent to the existing chat application, the system 400 may proceed to the method termination step 1055.

[0068] Figure 11 provides a flowchart 1100 illustrating a preferred method step that may be used to carry out a method of collecting information on users 405 and generating reports on user consumption and use of machine learning modules 425. Step 1105 marks the start of the method. During step 1110, the processor 220 collects usage data from multiple users 405, and then during step 1115, it may store the usage data in the users' user profiles, in which case the usage data relates to the machine learning modules 425 used by each user 405, information provided to the machine learning modules 425, information collected by the machine learning modules 425 in relation to the provided information, and task data generated by the machine learning modules 425. The system 400 may then, during step 1120, perform a query to determine whether a user 405 has requested a usage analysis. Based on the results of the query, the system 400 may perform an action during step 1123. If system 400 determines that no user 405 has requested a usage analysis, system 400 may proceed to method termination step 1150. If system 400 determines that user 405 has requested a usage analysis, system 400 may then query during step 1125 to determine whether user 405 has an appropriate authorization level to receive the usage analysis. Based on the result of the query, processor 220 may perform an action during step 1130. If system 400 determines that user 405 does not have an appropriate authorization level 800 to receive the usage analysis, system 400 may proceed to method termination step 1150. If system 400 determines that user 405 has an appropriate authorization level 800 to receive the usage analysis, system 400 may determine during steps 1135 and 1140, respectively, which user 405 and / or group of users are specified in the input data, and then generate a usage report.System 400 may send a usage report to user 405 during step 1145 before proceeding to method termination step 1150.

[0069] The subject matter described herein may be embodied in systems, apparatus, methods, and / or items, depending on the desired configuration. Specifically, various embodiments of the subject matter described herein may be realized in digital electronic circuits, integrated circuits, specially designed application-specific integrated circuits (ASICs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include embodiments in one or more computer programs that may be executable and / or interpretable on a programmable system including a storage system and at least one programmable processor, which may be dedicated or general-purpose, connected to receive data and instructions from there and transmit data and instructions thereto.

[0070] These computer programs, also called programs, software, applications, software applications, components, or code, may contain machine instructions to a programmable processor and may be implemented in high-level procedural and / or object-oriented programming languages, as well as / or assembly machine languages. In this specification, the term “persistent computer-readable medium” refers to any computer program, product, device, and / or apparatus, such as magnetic disks, optical disks, memory, and programmable logic devices (PLDs), used to provide machine instructions and / or data to a programmable processor, including persistent computer-readable medium that receives machine instructions, such as computer-readable signals. The term “computer-readable signal” refers to any signal used to provide machine instructions and / or data to a programmable processor. To provide interaction with a user, the subjects described herein may be implemented on a computer having displays, such as cathode ray tubes (CRDs), liquid crystal displays (LCDs), and light-emitting displays (LEDs), for providing information to the user, as well as pointing devices, such as keyboards and mice or trackballs, by which the user can provide input to the computer. The displays may include, but are not limited to, visual, auditory, tactile, kinesthetic, olfactory, and gustatory displays, or any combination thereof.

[0071] Other types of devices may also be used to facilitate interaction with the user. For example, feedback provided to the user may be any form of sensory feedback, such as visual, auditory, or tactile feedback, and input from the user may be received in any form, including but not limited to acoustic, voice, or tactile input. The subject matter described herein may be implemented as a computing system including backend components such as a data server, or middleware components such as an application server, or frontend components such as a client computer having a graphical user interface or a web browser through which the user can interact with the system described herein, or as any combination of such backend, middleware, or frontend components. The components of the system may be interconnected by digital data communications of any form or medium, such as a communication network. Examples of communication networks may include, but are not limited to, a local area network ("LAN"), a wide area network ("WAN"), a metropolitan area network ("MAN"), and the Internet.

[0072] The embodiments described herein do not represent all embodiments that are consistent with the subject matter described herein. Rather, they are merely examples that are consistent with aspects related to the subject matter described herein. While two or three variations are described in detail above, other modifications or additions are possible. Specifically, further features and / or variations can be provided in addition to those described herein. For example, the embodiments described herein may cover various combinations and partial combinations of the disclosed features and / or combinations and partial combinations of some of the further features disclosed above. In addition, the logical flows shown in the accompanying drawings and / or described herein do not necessarily require a specific order or sequence shown in order to achieve the desired result. It will be readily apparent to those skilled in the art that various other changes in the details, apparatus, and arrangement of the parts and method steps described and illustrated to illustrate the subject matter of this invention can be made without departing from the principles and scope of the subject matter of the invention.

Claims

1. A computing device operably connected to a display and an input device, The computing device includes an existing user interface, an add-on user interface, and security rules. The information input field of the existing user interface of the computing device is configured to receive input data from the input device. The user interface of the add-on is incorporated into the existing user interface in such a manner that the input data entered into the information input field is received by the user interface of the add-on. The user interface of the add-on is operablely connected to multiple machine learning methods that are separate from the user interface of the add-on and the existing user interface. The user interface of the add-on, if the input data includes an intelligence indicator, causes the input data to be sent to one of the multiple machine learning methods. The intelligence indicator specifically instructs the user interface of the add-on which of the multiple machine learning methods should be used to process the input data. When the intelligence indicator is entered into the information input field along with the input data, the input data is checked for violations of the security rules. Computing devices and Multiple computing entities including the aforementioned multiple machine learning methods, A processor operably connected to the computing device and the plurality of computing entities, A persistent computer-readable medium coupled to the processor, The said persistent computer-readable medium includes instructions stored thereon, which, when executed by the processor, are transmitted to the processor. Receiving the input data intercepted by the user interface of the add-on, The user interface of the add-on intercepts the input data directed to the information input field of the existing user interface. To determine whether the input data includes the intelligence indicator, If the input data includes the intelligence indicator, the input data is examined for any violation of the security rules. If the input data includes the intelligence indicator and there is no violation of the security rules, the input data intercepted by the add-on's user interface is transmitted to the plurality of computing entities. The machine learning methods of the plurality of computing entities are determined based on the intelligence indicators. Processing the input data using the machine learning techniques of the plurality of computing entities in order to create task data, The task data is generated by the machine learning method according to the task instructions contained in the input data. Transmitting the task data generated by the machine learning method to the existing user interface of the computing device, and The task data generated by the aforementioned machine learning method is presented within the information stream of the existing user interface. Perform an action that includes Persistent computer-readable media and including, A system for enhancing existing user interfaces and monitoring data transfer between computing devices.

2. The system according to claim 1, wherein the security rules are used to check for violations of the security rules before the input data entered into the information input field of the existing user interface is transmitted to the machine learning method among the plurality of machine learning methods.

3. When executed by the aforementioned processor, the aforementioned processor: The input data entered into the information input field is checked for violations of the security rules before being transmitted to one of the machine learning methods among the multiple machine learning methods. The system according to claim 1, further comprising additional instructions stored on the persistent computer-readable medium for causing additional operations, including,

4. The system according to claim 3, wherein the comparison of the input data with the security rules helps prevent the input data entered into the information input field from being transmitted to the machine learning method among the plurality of machine learning methods if a violation of the security rules is determined.

5. When executed by the aforementioned processor, the aforementioned processor: The transmission of the input data from the information input field to the machine learning method among the multiple machine learning methods is terminated if there is a violation of the security rules. The system according to claim 4, further comprising the additional instructions stored on the persistent computer-readable medium for causing additional operations, including the execution of additional operations.

6. The system according to claim 1, wherein a first machine learning method is used to create first task data, a second machine learning method is used to create second task data, the first task data and the second task data are provided to a third machine learning method, the third machine learning method uses the first task data and the second task data to create weighted task data, the weighted task data is a combination of the first task data and the second task data.

7. The system according to claim 1, wherein the input data entered into the information input field is stored in a data record as past input data if the input data includes the intelligence indicator.

8. The system according to claim 7, wherein the security rules are used to examine the past input data stored in the data record for any violation of the security rules.

9. The system according to claim 1, wherein the input data entered into the information input field is stored in an input catalog as past input data, the task data created by the machine learning method is stored in the input catalog as past task data, the input catalog allows a user to view the past input data entered into the information input field and the past task data associated with the past input data, and the input catalog is stored in at least one of the persistent computer-readable media or database.

10. When executed by the aforementioned processor, the aforementioned processor: The input data entered into the information input field and the task data created by the machine learning method are stored as an input catalog in at least one of the persistent computer-readable media or the database, wherein the input catalog allows the user to view the input data entered into the information input field and the task data associated with the input data. The system according to claim 9, further comprising the additional instructions stored on the persistent computer-readable medium for causing additional operations, including,

11. A first computing device operably connected to a display and an input device, The first computing device includes an existing user interface and an add-on user interface. The information input field of the existing user interface of the first computing device is configured to receive input data from the input device, The user interface of the add-on is incorporated into the existing user interface in such a manner that the input data entered into the information input field is received by the user interface of the add-on. The user interface of the add-on is operablely connected to multiple machine learning methods that are separate from the user interface of the add-on and the existing user interface. The user interface of the add-on, if the input data includes an intelligence indicator, causes the input data to be sent to one of the multiple machine learning methods. The intelligence indicator specifically instructs the user interface of the add-on which of the machine learning methods should be used to process the input data. The first computing device and A second computing device having multiple security rules, The second computing device is configured to receive the input data from the first computing device if the input data includes the intelligence indicator. The second computing device is configured to, if the input data includes the intelligence indicator, to inspect the input data for violations of security rules using the plurality of security rules. A second computing device, A third computing device having the aforementioned machine learning method, The third computing device is configured to receive the input data from the first computing device if the input data includes the intelligence indicator and the input data does not violate the plurality of security rules. A third computing device, A processor operably connected to the first computing device, the second computing device, and the third computing device, A persistent computer-readable medium coupled to the processor, The persistent computer-readable medium, when executed by the processor, allows the processor to: Receiving the input data intercepted by the user interface of the add-on through the first computing device, The user interface of the add-on intercepts the input data directed to the information input field of the existing user interface. To determine whether the input data includes the intelligence indicator, If the input data includes the intelligence indicator, transmit the input data from the first computing device to the second computing device. Analyze the input data using the aforementioned security rules to determine whether there is a security breach. The input data intercepted by the user interface of the add-on is transmitted from the first computing device to the third computing device if the input data includes the intelligence indicator and does not violate the security rules. Processing the input data using the machine learning method to create task data, The task data is generated by the machine learning method according to the task instructions contained in the input data. Transmitting the task data generated by the machine learning method to the existing user interface of the first computing device, The task data generated by the aforementioned machine learning method is presented within the information stream of the existing user interface. Includes instructions stored above which to perform an action, Persistent computer-readable media and A system that enhances existing chat applications, including monitoring data transferred between computing devices.

12. The system according to claim 11, wherein the input data is transmitted to the second computing device when it is entered into the information input field.

13. The system according to claim 12, wherein the input data is checked against the security rules when it is received by the second computing device.

14. The system according to claim 11, wherein a first machine learning method is used to create first task data, a second machine learning method is used to create second task data, the first task data and the second task data are provided to a third machine learning method, the third machine learning method uses the first task data and the second task data to create weighted task data, the weighted task data is a combination of the first task data and the second task data.

15. The system according to claim 11, wherein the input data entered into the information input field is stored in a data record as past input data if the input data includes the intelligence indicator.

16. The system according to claim 15, wherein the security rules are used to examine the past input data stored in the data record for any violation of the security rules.

17. The system according to claim 11, wherein the input data entered into the information input field is stored in the input catalog as past input data, the task data created by the machine learning method is stored in the input catalog as past task data, the input catalog allows the user to view the past input data entered into the information input field and the past task data associated with the past input data, and the input catalog is stored in at least one of the persistent computer-readable medium or database.

18. When executed by the aforementioned processor, the aforementioned processor: The input data entered into the information input field and the task data created by the machine learning method are stored as an input catalog in at least one of the persistent computer-readable media or the database, wherein the input catalog allows the user to view the input data entered into the information input field and the task data associated with the input data. The system according to claim 17, further comprising the additional instructions stored on the persistent computer-readable medium for causing additional operations, including the execution of additional operations.

19. When executed by the aforementioned processor, the aforementioned processor: The input catalog shall be inspected for any violation of the security rules. The system according to claim 18, further comprising the additional instructions stored on the persistent computer-readable medium for causing additional operations, including the execution of additional operations.

20. When executed by the aforementioned processor, the aforementioned processor: When it is found that at least one of the input data or task data violates the multiple security rules, correct the input catalog. The system according to claim 19, further comprising the additional instructions stored on the persistent computer-readable medium for causing additional operations, including the execution of additional operations.

21. One or more persistent computer-readable media coupled to a processor, having instructions stored thereon, wherein when the instructions are executed by the processor, the processor This involves receiving input data from the user interface of an add-on that is integrated into an existing user interface. The user interface of the add-on intercepts the input data entered into the existing user interface, The user interface of the aforementioned add-on is operablely connected to multiple machine learning methods. The input data is to determine whether it includes an intelligence indicator, The intelligence indicator specifies which of the multiple machine learning methods should be used to process the input data. If the input data includes the intelligence indicator, the input data intercepted by the add-on's user interface is sent to the machine learning method as specified by the intelligence indicator. The task data is created based on the input data via the machine learning method as specified by the intelligence indicator, The task data is generated by the machine learning method according to the task instructions contained in the input data. To transmit the task data generated by the machine learning method to the existing user interface, and The task data generated by the machine learning method is presented via the existing user interface. One or more persistent computer-readable media that perform the operation of [the specified action].

22. The input data is checked for violations of security rules before being transmitted to the machine learning method, in one or more persistent computer-readable media according to claim 21.

23. When executed by the aforementioned processor, the aforementioned processor: The input data is inspected for violations of security rules before being sent to the machine learning method. One or more persistent computer-readable media according to claim 21, further comprising additional instructions for causing additional operations, including,

24. The intelligence indicator specifies the creation of first task data and second task data via a first machine learning method, as described in claim 21, for one or more persistent computer-readable media.

25. The intelligence indicator specifies the creation of a third task data via a third machine learning method using the first task data and the second task data, one or more persistent computer-readable media according to claim 24.

26. When executed by the aforementioned processor, the aforementioned processor: The first task data is created via a first machine learning method as specified by the intelligence indicator, The second task data is created via a second machine learning method as specified by the intelligence indicator, Creating third task data via a third machine learning method as specified by the intelligence indicator, The first task data and the second task data are provided to the third machine learning method to assist in the creation of the third task data. Transmit the third task data generated by the third machine learning method to the existing user interface, The third task data is presented via the existing user interface. One or more persistent computer-readable media according to claim 25, further comprising additional instructions for causing additional operations, including,

27. When executed by the aforementioned processor, the aforementioned processor: The input data and task data are stored in a data record. One or more persistent computer-readable media according to claim 21, further comprising additional instructions for causing additional operations, including,

28. When executed by the aforementioned processor, the aforementioned processor: To inspect the input data and task data stored in the data record for violations of security rules. One or more persistent computer-readable media according to claim 37, further comprising additional stored instructions that cause additional operations to be performed.

29. When executed by the aforementioned processor, the aforementioned processor: If a violation of the aforementioned security rule is determined, the input data and task data shall be deleted from the data record. One or more persistent computer-readable media according to claim 38, further comprising additional instructions for causing additional operations, including,

30. One or more persistent computer-readable media coupled to a processor, having instructions stored thereon, wherein when the instructions are executed by the processor, the processor This involves receiving input data from the user interface of an add-on that is integrated into an existing user interface. The user interface of the add-on intercepts the input data entered into the existing user interface, The user interface of the aforementioned add-on is operable with multiple machine learning methods, The input data is to determine whether it includes an intelligence indicator, The intelligence indicator specifies which of the plurality of machine learning methods is used as the first machine learning method for processing the input data. The intelligence indicator specifies which of the plurality of machine learning methods is used as the second machine learning method for processing the input data, If the input data includes the intelligence indicator, the input data intercepted by the user interface of the add-on is transmitted to the first machine learning method as specified by the intelligence indicator. If the input data includes the intelligence indicator, the input data intercepted by the user interface of the add-on is sent to the second machine learning method as specified by the intelligence indicator. Creating first task data based on the input data via the first machine learning method as specified by the intelligence indicator, The first task data is generated by the first machine learning method in accordance with the task instructions contained in the input data. Creating second task data based on the input data via the second machine learning method as specified by the intelligence indicator, The second task data is generated by the second machine learning method in accordance with the task instructions contained in the input data. The first task data generated by the first machine learning method and the second task data generated by the second machine learning method are transmitted to the existing user interface. The first task data and the second task data are presented via the existing user interface. One or more persistent computer-readable media that perform the operation of [the specified action].

31. The input data is checked for violations of security rules before being transmitted to the first machine learning method and the second machine learning method, in one or more persistent computer-readable media according to claim 30.

32. When executed by the aforementioned processor, the aforementioned processor: The input data is inspected for violations of security rules before being sent to the machine learning method. One or more persistent computer-readable media according to claim 30, further comprising additional instructions for causing additional operations, including,

33. The intelligence indicator specifies the creation of a third task data via a third machine learning method using the first task data and the second task data, one or more persistent computer-readable media according to claim 30.

34. When executed by the aforementioned processor, the aforementioned processor: Creating third task data via a third machine learning method as specified by the intelligence indicator, The first task data and the second task data are provided to the third machine learning method to assist in the creation of the third task data. Transmit the third task data generated by the third machine learning method to the existing user interface, The third task data is presented via the existing user interface. One or more persistent computer-readable media according to claim 33, further comprising additional instructions for causing additional operations, including,

35. When executed by the aforementioned processor, the aforementioned processor: The input data, the first task data, and the second task data are stored in a data record. One or more persistent computer-readable media according to claim 30, further comprising additional instructions for causing additional operations, including,

36. When executed by the aforementioned processor, the aforementioned processor: To inspect the input data, first task data, and second task data stored in the data record for violations of security rules. One or more persistent computer-readable media according to claim 35, further comprising stored additional instructions that cause additional operations to be performed.

37. When executed by the aforementioned processor, the aforementioned processor: If a violation of the security rules is determined, the input data, the first task data, and the second task data shall be deleted from the data record. One or more persistent computer-readable media according to claim 36, further comprising additional instructions for causing additional operations, including,

38. One or more persistent computer-readable media coupled to a processor, having instructions stored thereon, wherein when the instructions are executed by the processor, the processor This involves receiving input data from the user interface of an add-on that is integrated into an existing user interface. The user interface of the add-on intercepts the input data entered into the existing user interface, The user interface of the aforementioned add-on is operable with multiple machine learning methods, The input data is to determine whether it includes an intelligence indicator, The intelligence indicator specifies which of the plurality of machine learning methods is used as the first machine learning method for processing the input data. The intelligence indicator specifies which of the plurality of machine learning methods is used as the second machine learning method for processing the input data, If the input data includes the intelligence indicator, the input data intercepted by the user interface of the add-on is transmitted to the first machine learning method as specified by the intelligence indicator. If the input data includes the intelligence indicator, the input data intercepted by the user interface of the add-on is sent to the second machine learning method as specified by the intelligence indicator. Creating first task data based on the input data via the first machine learning method as specified by the intelligence indicator, The first task data is generated by the first machine learning method in accordance with the task instructions contained in the input data. Creating second task data based on the input data via the second machine learning method as specified by the intelligence indicator, The second task data is generated by the second machine learning method in accordance with the task instructions contained in the input data. The first task data generated by the first machine learning method and the second task data generated by the second machine learning method are transmitted to the third machine learning method as indicated by the intelligence indicator. The third task data is created using the third machine learning method, as specified by the intelligence indicator, with support from the first task data and the second task data. Transmit the third task data generated by the third machine learning method to the existing user interface, The third task data is presented within the existing user interface. One or more persistent computer-readable media that perform the operation of [the specified action].

39. When executed by the aforementioned processor, the aforementioned processor: The input data and the third task data are stored in the data record. One or more persistent computer-readable media according to claim 38, further comprising additional instructions for causing additional operations, including,

40. When executed by the aforementioned processor, the aforementioned processor: The input data and the third task data stored in the data record are to be examined for violations of security rules, If a violation of the security rules is determined, the input data and the third task data shall be deleted from the data record. One or more persistent computer-readable media according to claim 39, further comprising additional instructions for causing additional operations, including,

Images