Method and system for detecting margin call factors using machine learning

Abstract

A method is provided to identify and capture all relevant features involved in the margin call process so that their root causes can be automatically identified using machine learning. [Solution] The method includes acquiring data related to a securities portfolio, automatically determining data patterns related to triggering a margin call and corresponding characteristics that contribute to triggering a margin call, automatically generating a margin call model based on the data patterns and characteristics, and evaluating a first securities portfolio using the margin call model to determine information related to whether to trigger a first margin call on the first securities portfolio.

Description

CROSS-REFERENCE TO RELATED APPLICATIONS

[0001] This application claims priority to U.S. Provisional Patent Application No. 62 / 946,096, filed December 10, 2019, the disclosure of which is incorporated herein in its entirety. [Technical Field]

[0002] The technology relates generally to methods and systems for detecting the causes of margin calls, and more particularly to methods and systems for identifying and capturing all relevant features involved in the margin call process so that their root causes can be automatically identified using machine learning. [Background technology]

[0003] Today, many financial institutions use margin calls to ensure that investors are within their maintenance margin requirements. Historically, the decision to call investors on margin was made ex post, based on varying degrees of success in mitigating investment risk. Summary of the Invention [Problem to be solved by the invention]

[0004] One drawback of using traditional analysis to determine whether a margin call is necessary for investors is that this analysis is often completed inefficiently using multiple systems and multiple data elements, after a portfolio has already exceeded an arbitrary threshold. Furthermore, this process is manual, and there is an element of subjectivity because data across different disciplines is interpreted by humans. As a result, financial institutions issuing margin calls spend a lot of resources on hindsight information, which inefficiently mitigates risk.

[0005] Therefore, there is a need for trading modeling that provides investors with information related to the various factors that led to a margin call so that they can efficiently correct any shortfalls. Furthermore, there is a need for trading modeling that provides financial institutions with information on the outlook for margin calls so that they can correct shortfalls without unnecessary risk exposure. Ultimately, there is a need to standardize the margin call process in how margin call features are interpreted across multiple domains and locations so that there is one unique interpretation that is free of human bias. [Means for solving the problem]

[0006] In particular, the present disclosure, in various aspects, embodiments, and / or specific features or subcomponents, provides various systems, servers, devices, methods, media, programs, and platforms for identifying and acquiring all relevant features involved in the margin call process such that the root causes can be automatically identified using machine learning.

[0007] According to one aspect of the present disclosure, there is provided a method for identifying and acquiring all relevant features involved in a margin call process such that root causes can be automatically identified. The method is implemented by at least one processor. The method may include obtaining a plurality of data related to a plurality of securities portfolios, using the plurality of data to automatically determine at least one data pattern related to triggering a margin call and at least one corresponding feature contributing to triggering the margin call, automatically generating at least one margin call model based on the at least one data pattern and the at least one feature, and evaluating a first securities portfolio using the at least one margin call model to determine first information related to whether to trigger a first margin call for the first securities portfolio.

[0008] According to an example embodiment, the evaluating may include selecting, by the at least one processor, a first one of the at least one data pattern and a first one of the at least one corresponding feature related to triggering the first margin call for the first portfolio of securities, and displaying, by the at least one processor, the first information in a graphical user interface together with the selected first one of the at least one data pattern and the selected first one of the at least one corresponding feature via a display.

[0009] According to an example embodiment, the at least one margin call model may be generated automatically using machine learning analysis of the plurality of data.

[0010] According to an example embodiment, the at least one margin call model may include at least one of a data model, a mathematical model, a process model, and a machine learning model.

[0011] According to an example embodiment, the method may further include evaluating a second securities portfolio with the at least one margin call model to determine second information related to whether to trigger a future second margin call on the second securities portfolio, and displaying the second information in a graphical user interface via a display.

[0012] According to an example embodiment, the graphical user interface may include at least one dashboard displaying at least one of monitoring information related to the first securities portfolio and classification information related to the at least one characteristic.

[0013] According to an example embodiment, the at least one feature may include at least one of a security appreciation feature, a security depreciation feature, an asset market value discrepancy feature, a security market value discrepancy feature, a security market value buy feature, an exposure discrepancy feature, a cash market value discrepancy feature, a collateral discrepancy feature, a cash market value sell feature, a haircut discrepancy feature, a security market value sell feature, a security liquidation price buy feature, a cash market value appreciation feature, a cash market value depreciation feature, a pledge market value out feature, a pledge market value discrepancy feature, a cash market value buy feature, a derivative exposure delta feature, a derivative initial margin exposure delta feature, a pledge liquidation price in feature, an accrued interest accumulation feature, a pledge market value in feature, a pledge market value appreciation feature, a pledge market value depreciation feature, a derivative negative mark-to-market delta feature, and an unlimited pledge discrepancy feature.

[0014] According to an example embodiment, the method may further include receiving feedback information from at least one user via a graphical user interface and updating the at least one margin call model based on the feedback information.

[0015] According to an exemplary embodiment, the feedback information includes at least one of a first information accuracy rank, a second information accuracy rank, and a feature accuracy rank.

[0016] According to an exemplary embodiment, the method may further include analyzing the plurality of data; identifying personally identifiable information in the analyzed plurality of data; and anonymizing the plurality of data by removing the identified personally identifiable information.

[0017] According to one aspect of the present disclosure, a computing device is disclosed that implements a method for identifying and acquiring all relevant features involved in a margin call process so that root causes thereof can be automatically identified. The computing device includes a processor, a memory, and a communication interface coupled to each of the processor and the memory. The processor may be configured to acquire a plurality of data related to a plurality of securities portfolios, use the plurality of data to automatically determine at least one data pattern related to triggering a margin call and at least one corresponding feature that contributes to triggering the margin call, automatically generate at least one margin call model based on the at least one data pattern and the at least one feature, and evaluate a first securities portfolio using the at least one margin call model to determine first information related to whether to trigger a first margin call for the first securities portfolio.

[0018] According to an example embodiment, for the evaluation, the processor may be further configured to select a first one of the at least one data pattern and a first one of the at least one corresponding feature related to triggering the first margin call for the first portfolio of securities, and display the first information in a graphical user interface via a display together with the selected first one of the at least one data pattern and the selected first one of the at least one corresponding feature.

[0019] According to an example embodiment, the processor may be further configured to automatically generate the at least one margin call model using machine learning analysis of the plurality of data.

[0020] According to an example embodiment, the at least one margin call model may include at least one of a data model, a mathematical model, a process model, and a machine learning model.

[0021] According to an exemplary embodiment, the processor may be further configured to evaluate a second securities portfolio using the at least one margin call model to determine second information related to whether to trigger a future second margin call on the second securities portfolio, and display the second information in a graphical user interface via a display.

[0022] According to an example embodiment, the graphical user interface may include at least one dashboard displaying at least one of monitoring information related to the first securities portfolio and classification information related to the at least one characteristic.

[0023] According to an exemplary embodiment, the at least one feature may be a security appreciation feature, a security depreciation feature, an asset market value discrepancy feature, a security market value discrepancy feature, a security market value buy feature, an exposure discrepancy feature, a cash market value discrepancy feature, a collateral discrepancy feature, a cash market value sell feature, a haircut discrepancy feature, a security market value sell feature, a security liquidation price buy feature, a cash market value appreciation feature, a cash market value depreciation feature, a pledge market value out feature, a pledge market value discrepancy feature, a cash market value buy feature, a derivative the delta feature of exposure to a pledge, the delta feature of exposure to a derivative initial margin, the in feature of pledge liquidation price, the accumulation feature of accrued interest, the in feature of market value of pledge, the accretion feature of market value of pledge, the depreciation feature of market value of pledge, the delta feature of negative mark-to-market of derivatives, the movement feature of loan amount fluctuations due to foreign exchange, the due date feature of accrued interest debited to private equity call or future value, and the difference feature of unlimited pledge.

[0024] According to an exemplary embodiment, the processor may be further configured to receive feedback information from at least one user via a graphical user interface and update the at least one margin call model based on the feedback information.

[0025] According to an example embodiment, the feedback information may include at least one of a first information accuracy rank, a second information accuracy rank, and a feature accuracy rank.

[0026] According to an exemplary embodiment, the processor may be further configured to analyze the plurality of data, identify personally identifiable information in the analyzed plurality of data, and anonymize the plurality of data by removing the identified personally identifiable information. [Brief explanation of the drawings]

[0027] The present disclosure is further explained in the detailed description by way of non-limiting examples of preferred embodiments of the present disclosure with reference to the noted several drawings in which like numerals represent like elements. [Figure 1] 1 illustrates an exemplary computer system. [Figure 2] 1 shows an exemplary diagram of a network environment. [Figure 3] We present an exemplary system for implementing a method for identifying and capturing all relevant features involved in a margin call process using machine learning so that their root causes can be automatically identified. [Figure 4] 1 is a flowchart of an exemplary process for implementing a method for identifying and capturing all relevant features involved in a margin call process so that their root causes can be automatically identified using machine learning. [Figure 5] FIG. 1 is a diagram of a machine learning process that can be used to implement a method for identifying and obtaining all relevant features involved in a margin call process so that their root causes can be automatically identified using machine learning, according to an example embodiment. [Figure 6]1 is a graphical user interface illustrating variable importance of features identified in a method for identifying and obtaining all relevant features involved in a margin call process using machine learning so that their root causes can be automatically identified, according to an example embodiment. [Figure 7] 1 is a screenshot illustrating a graphical user interface that can be used to implement a method for identifying and capturing all relevant features involved in a margin call process using machine learning so that their root causes can be automatically identified, according to an example embodiment. [Figure 8] 1 is a flowchart of an exemplary process for implementing a method for identifying and capturing all relevant features involved in a margin call process so that their root causes can be automatically identified using machine learning. [Figure 9] 1 is a flowchart of an exemplary process for implementing a method for identifying and capturing all relevant features involved in a margin call process so that their root causes can be automatically identified using machine learning. [Figure 10] 1 is a screenshot illustrating a graphical user interface of an uncategorized ticket that can be used to implement a method for identifying and capturing all relevant features involved in a margin call process so that their root causes can be automatically identified using machine learning, according to an example embodiment. [Figure 11] 1 is a screenshot illustrating a graphical user interface of an uncategorized tickets dashboard that can be used to implement a method for identifying and capturing all relevant features involved in a margin call process using machine learning so that their root causes can be automatically identified, according to an example embodiment. [Figure 12] 1 is a screenshot illustrating a classification check graphical user interface that can be used to implement a method for identifying and capturing all relevant features involved in a margin call process using machine learning so that their root causes can be automatically identified, according to an example embodiment. [Figure 13]1 is a screenshot illustrating a graphical user interface of a categorized ticket that can be used to implement a method for identifying and capturing all relevant features involved in a margin call process so that their root causes can be automatically identified using machine learning, according to an example embodiment. [Figure 14A] 1 is a screenshot illustrating a graphical user interface of an artificial intelligence function that can be used to implement a method for identifying and capturing all relevant features involved in a margin call process using machine learning so that their root causes can be automatically identified, according to an example embodiment. [Figure 14B] 1 is a screenshot illustrating a ticket details graphical user interface that can be used to implement a method for identifying and capturing all relevant features involved in a margin call process using machine learning so that their root causes can be automatically identified, according to an example embodiment. [Figure 15] 1 is a screenshot illustrating a graphical user interface of a categorized ticket dashboard that can be used to implement a method for identifying and capturing all relevant features involved in a margin call process using machine learning so that their root causes can be automatically identified, according to an example embodiment. [Figure 16] 1 is a screenshot illustrating a model accuracy graphical user interface that can be used to implement a method for identifying and obtaining all relevant features involved in a margin call process using machine learning so that their root causes can be automatically identified, according to an example embodiment. [Figure 17] 1 is a screenshot illustrating an accuracy monitoring graphical user interface that can be used to implement a method for identifying and capturing all relevant features involved in a margin call process using machine learning so that their root causes can be automatically identified, according to an example embodiment. [Figure 18]1 shows a show model illustrating feature interactions that can be used to implement a method for identifying and obtaining all relevant features involved in a margin call process using machine learning so that their root causes can be automatically identified, according to an example embodiment. [Figure 19A] 1 is a screenshot illustrating a graphical user interface of a user feedback form that can be used to implement a method for identifying and capturing all relevant features involved in a margin call process using machine learning so that their root causes can be automatically identified, according to an example embodiment. [Figure 19B] 1 is a screenshot illustrating an enhanced graphical user interface displayed based on user input in a user feedback form that can be used to implement a method for identifying and capturing all relevant features involved in a margin call process using machine learning so that their root causes can be automatically identified, according to an example embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0028] The embodiments and / or particular features or subcomponents of the present disclosure are intended to demonstrate one or more advantages, particularly as described above and mentioned below, through one or more of the various aspects.

[0029] The examples may also be embodied as one or more non-transitory or non-transitory computer-readable media on which are stored instructions for one or more aspects of the technology as described and illustrated in the examples herein. The instructions in some examples include executable code that, when executed by one or more processors, causes the processors to perform the steps necessary to implement the methodologies in the examples of the technology as described and illustrated in the examples.

[0030] 1 is an exemplary system used in accordance with embodiments described herein. System 100 is shown generally and may include a computer system 102, also shown generally.

[0031] Computer system 102 may include a set of executable instructions to cause computer system 102, alone or in combination with other devices described, to perform any one or more of the methods or computer-based functions disclosed herein. Computer system 102 may operate as a standalone device or may be connected to other systems or peripherals. For example, computer system 102 may include or be included in any one or more computers, servers, systems, communication networks, or cloud environments. Furthermore, the instructions may be executed in such cloud-based computing environments.

[0032] In a network deployment or implementation, computer system 102 may operate in the capacity of a server or as a client user computer in a server-client user network environment, a client user computer in a cloud computing environment, or a peer computer system in a peer-to-peer (or distributed) network environment. Computer system 102, or portions thereof, may be implemented as or incorporated into a variety of devices, such as a personal computer, a tablet computer, a set-top box, a personal digital assistant, a mobile device, a palmtop computer, a laptop computer, a desktop computer, a communication device, a wireless smartphone, a personal trusted device, a wearable device, a global positioning satellite (GPS) device, a web appliance, or any other machine capable of executing (sequentially or otherwise) a set of instructions that specify actions to be performed by the machine. Furthermore, while a single computer system 102 is shown, additional embodiments may include any collection of systems or subsystems that, individually or together, execute instructions or perform a function. The term "system" is taken throughout this disclosure to include any collection of systems or subsystems that individually or collectively execute one or more sets of instructions to perform one or more computer functions.

[0033] As shown in FIG. 1 , the computer system 102 may include at least one processor 104. The processor 104 is tangible and non-transitory. As used herein, the term “non-transitory” does not construe a permanent characteristic of a state, but rather a characteristic of a state that lasts for a period of time. The term “non-transitory” specifically negates fleeting characteristics, such as a particular carrier wave or signal, or other form of characteristic, that exists only temporarily at any time and in any place. The processor 104 is an article of manufacture and / or machine component. The processor 104 is configured to execute software instructions to perform functions as described in various embodiments herein. The processor 104 may be a general-purpose processor or part of an application-specific integrated circuit (ASIC). The processor 104 may also be a microprocessor, microcomputer, processor chip, controller, microcontroller, digital signal processor (DSP), state machine, or programmable logic circuit. The processor 104 may also be a logic circuit, including a programmable gate array (PGA), such as a field programmable gate array (FPGA), or other type of circuit including discrete gate and / or transistor logic. The processor 104 may be a central processing unit (CPU), a graphics processing unit (GPU), or both. Additionally, any processor described herein may include multiple processors, parallel processors, or both. Multiple processors may be contained in or connected to a single device or multiple devices.

[0034] The computer system 102 may also include computer memory 106. The computer memory 106 may include static memory, dynamic memory, or both in communication. Memory, as described herein, is a tangible storage medium capable of storing data as well as executable instructions and that is non-transitory while the instructions are stored therein. Again, as used herein, the term "non-transitory" is not to be construed as a permanent characteristic of a state, but rather as a characteristic of a state that lasts for a period of time. The term "non-transitory" specifically denies fleeting characteristics, such as a particular carrier wave or signal, or other form of characteristic, that exists only temporarily, anywhere at any time. Memory is an article of manufacture and / or machine component. Memory, as described herein, is a computer-readable medium from which data and executable instructions can be read by a computer. The memory described herein may be random access memory (RAM), read-only memory (ROM), flash memory, electrically programmable read-only memory (EPROM), electrically erasable EPROM (EEPROM), registers, hard disk, cache, removable disk, tape, compact disk read-only memory (CD-ROM), digital versatile disk (DVD), floppy disk, Blu-ray disk, or any other form of storage medium known in the art. Memory may be volatile or non-volatile, secure and / or encrypted, unsecure and / or non-encrypted. Of course, computer memory 106 may include any combination of memories or a single storage device.

[0035] The computer system 102 may further include a display 108, which may be, for example, a liquid crystal display (LCD), an organic light emitting diode (OLED), a flat panel display, a solid state display, a cathode ray tube (CRT), a plasma display, or other types of display, examples of which are well known to those skilled in the art.

[0036] The computer system 102 may also include at least one input device 110, such as a keyboard, a touch-sensitive input screen or pad, voice input, a mouse, a remote control device with a wireless keypad, a microphone connected to a voice recognition engine, a camera such as a video camera or a still camera, a cursor control device, a global positioning system (GPS) device, an altimeter, a gyroscope, an accelerometer, a proximity sensor, or any combination thereof. Those skilled in the art will appreciate that various embodiments of the computer system 102 may include multiple input devices 110. Moreover, those skilled in the art will further appreciate that the above list of exemplary input devices 110 is not intended to be exhaustive, and that the computer system 102 may include any additional or different input devices 110.

[0037] Computer system 102 may also include a media reading device 112 configured to read any one or more sets of instructions, e.g., software, from any of the memories described herein. When executed by the processor, the instructions can be used to perform one or more of the methods and processes described herein. In particular embodiments, the instructions reside, completely or at least partially, within memory 106, media reading device 112, and / or processor 110 during execution by computer system 102.

[0038] Additionally, computer system 102 may include any additional devices, components, parts, peripherals, hardware, software, or any combination thereof, which are well known and understood to be included with or within a computer system, such as, but not limited to, a network interface 114 and output devices 116. Output devices 116 may be, but are not limited to, speakers, audio output, video output, remote control output, printer, or any combination thereof.

[0039] The components of computer system 102 may be interconnected and communicate via a bus 118 or other communication link. As shown in FIG. 1, the components may each be interconnected and communicate via an internal bus. However, those skilled in the art will appreciate that any component may also be connected via an expansion bus. Moreover, bus 118 may be capable of communicating via any standard specification or other well-known and understood specification, including, but not limited to, Peripheral Component Interconnect (PCI), Peripheral Component Interconnect (PCI) Express, Parallel Advanced Technology Attachment, Serial Advanced Technology Attachment, and others.

[0040] Computer system 102 may communicate with one or more additional computer devices 120 via network 122. Network 122 may be a local area network (LAN), a wide area network (WAN), the Internet, a telephone network, a short-range network, or any other network well known and understood in the art. A short-range network may include, for example, Bluetooth, ZigBee, infrared, near-field communication, ultra-wideband, or any combination thereof. Those skilled in the art will understand that additional known and understood networks 122 may be used in addition to or instead, and that the exemplary network 122 is not intended to be limiting or exhaustive. Also, while network 122 is shown in FIG. 1 as a wireless network, those skilled in the art will understand that network 122 may also be a wired network.

[0041] Additional computing device 120 is shown in FIG. 1 as a personal computer. However, those skilled in the art will understand that in alternative embodiments of the present application, computing device 120 may be a laptop computer, a tablet PC, a personal digital assistant (PDA), a mobile device, a palmtop computer, a desktop computer, a communications device, a wireless telephone, a personal trusted device, a web appliance, a server, or any other device capable of sequentially or otherwise executing a set of instructions that specify operations to be performed by the device. Of course, those skilled in the art will understand that the above-listed devices are merely exemplary devices, and that device 120 may be any additional device or apparatus well known and understood in the art without departing from the scope of the present application. For example, computing device 120 may be the same as or similar to computer system 102. Furthermore, those skilled in the art will understand that this device may be any combination of devices and apparatuses.

[0042] Of course, those skilled in the art will understand that the above listed components of computer system 102 are meant to be exemplary only and are not intended to be exhaustive and / or comprehensive. Furthermore, the above listed example components are exemplary and similarly are not meant to be exhaustive and / or comprehensive.

[0043] According to various embodiments of the present disclosure, the methods described herein may be implemented using a hardware computer system executing a software program. Furthermore, in exemplary, non-limiting embodiments, implementations may include distributed processing, component / object distributed processing, and parallel processing. Virtual computer system processes may be configured to implement one or more of the methods or functions described herein, and processors described herein may be used to support virtual processing environments.

[0044] As described herein, various embodiments provide an optimized method and system for identifying and capturing all relevant features involved in the margin call process so that their root causes can be automatically identified using machine learning.

[0045] 2 is a schematic diagram of an exemplary network environment 200 for implementing a method for identifying and acquiring all relevant features involved in a margin call process so that their root causes can be automatically identified using machine learning. In an exemplary embodiment, the method can be executed on any network computer platform, such as a personal computer (PC).

[0046] The method for identifying and acquiring all relevant features related to the margin call process so that their root causes can be automatically identified using machine learning may be implemented by a Margin Call Data Collection and Analysis (MCDCA) device 202. The MCDCA device 202 may be the same as or similar to the computer system 102 described with respect to FIG. 1. The MCDCA device 202 may store one or more applications, which may include executable instructions that, when executed by the MCDCA device 202, cause the MCDCA device 202 to, for example, send, receive, or otherwise process network messages and perform other operations described and illustrated below with reference to the figures. The one or more applications may be implemented as modules or components of other applications. Furthermore, the one or more applications may be implemented as operating system extensions, modules, plug-ins, etc.

[0047] Furthermore, the one or more applications may operate in a cloud-based computing environment. The one or more applications may run within or as virtual machines or virtual servers that may be managed in the cloud-based computing environment. The one or more applications, and even the MCDCA device 202 itself, may be located on one or more virtual servers operating in the cloud-based computing environment rather than being tied to one or more specific physical network computing devices. The one or more applications may also be running on one or more virtual machines (VMs) running on the MCDCA device 202. Additionally, in one or more embodiments of the present technology, the one or more virtual machines running on the MCDCA device 202 may be managed or monitored by a hypervisor.

[0048] In the network environment 200 of Figure 2, an MCDCA device 202 is connected to a plurality of server devices 204(1)-204(n), which host a plurality of databases 206(1)-206(n), and also to a plurality of client devices 208(1)-208(n) via one or more communications networks 210. A communications interface of the MCDCA device 202, such as the network interface 114 of the computer system 102 of Figure 1, operatively couples and communicates between the MCDCA device 202, the server devices 204(1)-204(n), and / or the client devices 208(1)-208(n), all connected by one or more communications networks 210. However, other types and / or numbers of communication networks or systems with other types and / or numbers of connections and / or configurations for other devices and / or elements may be used.

[0049] The MCDCA device 202, server devices 204(1)-204(n), and / or client devices 208(1)-208(n) may be connected via other topologies. However, the one or more communication networks 210 may be the same as or similar to the network 112 as described with respect to FIG. 1. In addition, the network environment 200 may include other network devices, such as one or more routers and / or switches, that are well known in the art and therefore not described herein. This technique provides many advantages, including a method, a non-transitory computer-readable medium, and an MCDCA device that effectively implements a method for identifying and acquiring all relevant features involved in a margin call process using machine learning so that the root causes can be automatically identified.

[0050] By way of example only, the one or more communications networks 210 may include one or more local area networks (LANs) or wide area networks (WANs) and may use TCP / IP over Ethernet and other industry-standard protocols. However, other types and / or numbers of protocols and / or communications networks may also be used. The one or more communications networks 210 in this example may employ any suitable interface mechanism and network communications technology, including, for example, any suitable form of communications traffic (e.g., voice, modem, etc.), a Public Switched Telephone Network (PSTN), an Ethernet-based Packet Data Network (PDN), combinations thereof, etc.

[0051] The MCDCA device 202 may be a standalone device or may be integrated with one or more other devices or apparatuses, such as one or more of the server devices 204(1)-204(n). In one specific example, the MCDCA device 202 may include or be hosted by one of the server devices 204(1)-204(n), although other configurations are also possible. Furthermore, one or more of the devices in the MCDCA device 202 may reside on the same or different communication networks, including, for example, one or more public, private, or cloud networks.

[0052] The server devices 204(1)-204(n) may be the same as or similar to the computer system 102 or computer device 120 described with respect to FIG. 1 and may include any feature or combination of features described with respect to FIG. 1. For example, any of the server devices 204(1)-204(n) may include, among other features, one or more processors, memory, and communication interfaces, which are connected together by a bus or other communication link. However, other numbers and / or types of network devices may be used. The server devices 204(1)-204(n) in this example may process requests received from the MCDCA devices 202 over one or more communication networks 210, for example, via HTTP-based and / or JavaScript Object Notation (JSON) protocols. However, other protocols may be used.

[0053] Server devices 204(1)-204(n) may be hardware or software, or may represent a system with multiple servers in a pool that may include an internal or external network. Server devices 204(1)-204(n) host databases 206(1)-206(n) configured to store data related to margin calls, data patterns, characteristics, and margin call models.

[0054] Although server devices 204(1)-204(n) are shown as single devices, one or more operations of each server device 204(1)-204(n) may be distributed across one or more separate networked computing devices that together comprise one or more of server devices 204(1)-204(n). Furthermore, server devices 204(1)-204(n) are not limited to any particular configuration. As such, server devices 204(1)-204(n) may include multiple networked computing devices that operate using a master / slave approach, whereby one of the networked computing devices of server devices 204(1)-204(n) operates to manage or coordinate the operation of the other networked computing devices.

[0055] The server devices 204(1)-204(n) may operate as multiple networked computing devices, for example, in a cluster architecture, a peer-to-peer architecture, a virtual machine, or a cloud architecture. As such, the techniques disclosed herein should not be construed as limited to a single environment, and other configurations and architectures are contemplated.

[0056] The multiple client devices 208(1)-208(n) may be the same as or similar to the computer system 102 or computing device 120 as described with respect to FIG. 1 and may include any feature or combination of features described with respect to FIG. 1. For example, the client devices 208(1)-208(n) in this example may include any type of computing device capable of interacting with the MCDCA device 202 via one or more communications networks 210. Thus, the client devices 208(1)-208(n) may be, for example, mobile computing devices, desktop computing devices, laptop computing devices, tablet computing devices, virtual machines (including cloud-based computers), etc. that host chat, email, or voice-to-text applications. In an exemplary embodiment, at least one client device 208 is a wireless mobile communicator, e.g., a smartphone.

[0057] Client devices 208(1)-208(n) may execute an interface application, such as a standard web browser or a standalone client application, that provides an interface for communicating with MCDCA device 202 over one or more communications networks 210 to convey user requests and information. Client devices 208(1)-208(n) may also include a display device, such as a display screen or touch screen, and / or an input device, such as a keyboard, among other features.

[0058] Although an exemplary network environment 200 is described and illustrated herein, including MCDCA devices 202, server devices 204(1)-204(n), client devices 208(1)-208(n), and one or more communication networks 210, other types and / or numbers of systems, devices, components, and / or elements in other topologies may be used. As will be appreciated by those skilled in the art, many variations in the specific hardware and software used to implement these examples are possible, and it is understood that the systems in the examples described herein are for illustrative purposes.

[0059] One or more of the devices depicted in network environment 200, such as MCDCA device 202, server device 204(1)-204(n), or client device 208(1)-208(n), may be configured to operate as virtual instances on the same physical machine. That is, one or more of MCDCA device 202, server device 204(1)-204(n), or client device 208(1)-208(n) may operate on the same physical device rather than as separate devices communicating over one or more communications networks 210. Additionally, there may be more or fewer MCDCA devices 202, server devices 204(1)-204(n), or client devices 208(1)-208(n) than shown in FIG. 2 .

[0060] Furthermore, two or more computing systems or devices may be substituted for any of the systems or devices in any of the examples. Accordingly, principles and advantages of distributed processes, such as redundancy and repetition, may also be implemented as desired to increase the robustness and performance of the devices and systems in these examples. These examples may also be implemented in one or more computer systems across any suitable network using any suitable interface mechanism and traffic technology, including, by way of example only, any suitable form of communication traffic (e.g., voice and modem), wireless traffic networks, cellular traffic networks, packet data networks (PDNs), the Internet, intranets, and combinations thereof.

[0061] 3 as including a margin call data collection and analysis (MCDCA) module 302. However, the MCDCA module 302 may include, for example, other rules, policies, modules, databases, or applications. As described below, the MCDCA module 302 is configured to implement a method for identifying and acquiring all relevant features involved in the margin call process such that their root causes can be automatically identified using machine learning.

[0062] Utilizing the network environment of FIG. 2, an exemplary process 300 for implementing a mechanism for identifying and acquiring all relevant characteristics involved in a margin call process so that the root causes thereof can be automatically identified is shown as being performed in FIG. 3. Specifically, first client device 208(1) and second client device 208(2) are shown in communication with MCDCA device 202. In this regard, first client device 208(1) and second client device 208(2) may be “clients” of MCDCA device 202 and are described as such herein. Nevertheless, it is known and understood that first client device 208(1) and / or second client device 208(2) are not necessarily “clients” of MCDCA device 202 or any of the entities described in connection with this specification. Any additional or different relationships may exist between either or both of the first client device 208(1) and the second client device 208(2) and the MCDCA device 202, or no relationships may exist.

[0063] Additionally, the MCDCA device 202 is shown to have access to a margin call data repository 206(1) and a data pattern, feature, and margin call model database 206(2). The MCDCA module 302 may be configured to access these databases to implement a method for identifying and obtaining all relevant features involved in the margin call process using machine learning so that their root causes can be automatically identified.

[0064] The first client device 208(1) may be, for example, a smartphone. It will be appreciated that the first client device 208(1) may be any of the additional devices described herein. The second client device 208(2) may be, for example, a personal computer (PC). It will be appreciated that the second client device 208(2) may also be any of the additional devices described herein.

[0065] This process may be performed over one or more communication networks 210, which may comprise multiple networks such as those described above. For example, in an exemplary embodiment, either or both of first client device 208(1) and second client device 208(2) may communicate with MCDCA device 202 over broadband or cellular communications. Of course, these embodiments are merely exemplary and are not limiting or exhaustive.

[0066] Once activated, the MCDCA module 302 executes a process for identifying and obtaining all relevant features involved in the margin call process so that their root causes can be automatically identified using machine learning. An exemplary overall process for identifying and obtaining all relevant features involved in the margin call process so that their root causes can be automatically identified using machine learning is shown in flowchart 400 of FIG.

[0067] In step S402 of process 400 of Figure 4, data related to multiple securities portfolios may be obtained. In an exemplary embodiment, this data may include information related to securities portfolios of multiple investors. In yet another exemplary embodiment, the multiple investors may include unrelated institutions and / or related entities (i.e., subsidiaries), and the multiple investors may also include individual investors and / or institutional investors (i.e., hedge funds).

[0068] In yet another exemplary embodiment, this data may be analyzed and anonymized to remove identifying information of individuals corresponding to multiple investors. In yet another exemplary embodiment, a securities portfolio may include any group of financial assets, such as, for example, stocks, bonds, commodities, currencies, cash equivalents, mutual funds, exchange-traded funds, and closed funds. Other types of non-traded products, such as time deposits, pledges, and guarantees, may also be part of the portfolio. Additionally, a securities portfolio may also include privately traded securities, such as, for example, real estate, artwork, and private investments.

[0069] In yet another exemplary embodiment, the data may be obtained from any publicly accessible source (e.g., stock market data from an exchange) and privately held sources (e.g., account data from a bank). In yet another exemplary embodiment, the data may include any information such as, for example, trade classification data, stock price data, asset value data, and bond yield data related to each financial asset in a securities portfolio.

[0070] In yet another exemplary embodiment, the trade segment data may be any buy or sell orders, such as, for example, market orders, limit orders, stop orders (i.e., stop-loss orders), and limit buy orders. In yet another exemplary embodiment, the trade segment data may be any requests by a financial institution to an investor, such as, for example, federal margin calls, maintenance margin calls, and exchange margin calls, related to the investor's securities portfolio. As will be appreciated by those skilled in the art, the trade segment data may be any trades and corresponding trade information related to each security in the securities portfolio.

[0071] In step S404, data patterns related to margin call triggering and corresponding features contributing to margin call triggering may be automatically determined using this data. In an exemplary embodiment, the margin call may include generalized margin call scenarios derived from analysis of the acquired data. In yet another exemplary embodiment, the margin call may include predetermined scenarios established by the organization as standard triggers for margin calls.

[0072] In yet another exemplary embodiment, analyzing the data to automatically determine data patterns and features may include comparing the data to a pre-developed analytical model. In yet another exemplary embodiment, analyzing the data to automatically determine data patterns and features may include utilizing machine learning analytical techniques on the data. In yet another exemplary embodiment, the machine learning analytical techniques may include supervised learning algorithms, such as regression analysis, decision tree analysis, random forest analysis, k-nearest neighbor analysis, logistic regression analysis, etc. In yet another exemplary embodiment, the machine learning analytical techniques may include unsupervised learning algorithms, such as deductive analysis, k-means analysis, etc. In yet another exemplary embodiment, the machine learning analytical techniques may include reinforcement learning algorithms, such as Markov decision processes, etc.

[0073] In yet another exemplary embodiment, data patterns may be determined using historical data pattern recognition techniques (i.e., comparisons with known data patterns in historical data) and statistical recognition pattern techniques (i.e., machine learning and artificial intelligence related to recognizing patterns and regularities in data). For example, in historical data pattern recognition techniques, first data may be received and compared to pre-developed data patterns for portfolios of securities known to have triggered margin calls in the past. Alternatively, for example, in statistical recognition pattern techniques, data of past margin calls may be analyzed by a computing system using algorithms that enable the computing system to recognize data patterns. The identified data patterns may then be applied to a first portfolio of securities to predict the probability of a margin call for the first portfolio of securities.

[0074] In an exemplary embodiment, the features may be automatically determined using the machine learning analysis techniques described above based on data patterns. In yet another exemplary embodiment, the features may include underlying causes of margin calls based on their contribution and / or degree of impact on a margin call. For example, a decline in the value of Bond A may contribute to an already valid margin call due to a decline in the price of Stock B. In yet another exemplary embodiment, the degree of impact of each feature may be expressed as a percentage of its actual contribution to a margin call and a percentage probability of its contribution to a margin call.

[0075] In yet another exemplary embodiment, the features include a security appreciation feature, a security depreciation feature, an asset market value discrepancy feature, a security market value discrepancy feature, a security market value buy feature, an exposure discrepancy feature, a cash market value discrepancy feature, a collateral discrepancy feature, a cash market value sell feature, a haircut discrepancy feature, a security market value sell feature, a security liquidation price buy feature, a cash market value appreciation feature, a cash market value depreciation feature, a pledge market value out feature, a pledge market value discrepancy feature, a cash market value buy feature, a derivative exposure feature, a The pledge may include at least one of a delta feature of pledge surplus, a delta feature of derivative initial margin exposure, an in feature of pledge liquidation price, an accumulation feature of accrued interest, an in feature of market value of pledge, an accretion feature of market value of pledge, a depreciation feature of market value of pledge, a delta feature of negative mark-to-market of derivatives, a movement feature of loan amount fluctuations due to foreign exchange, a due date feature of accrued interest debited to private equity call or future value, and a difference feature of unlimited pledge.

[0076] In step S406, a margin call model may be automatically generated based on the data patterns and features. In an exemplary embodiment, the margin call model may include a mathematical model, such as a financial model designed to represent real-world financial conditions. In yet another exemplary embodiment, the margin call model may include a mathematical model that represents the likelihood of a margin call for any given portfolio of securities. In yet another exemplary embodiment, the margin call model may be generated for any of the trade segments described above. In yet another exemplary embodiment, the margin call model may be automatically generated to represent a particular analysis, such as an expected risk analysis and an expected outcome analysis. In yet another exemplary embodiment, the margin call model may be generated using any combination of data patterns and features.

[0077] In yet another exemplary embodiment, the margin call model may include a training model, such as a machine learning model, that is generated for further training with additional data. Once the training model is sufficiently trained, it may be deployed and used in various connected systems. In yet another exemplary embodiment, the training of the training model may be sufficient when a model evaluation method, such as a holdout method, a K-fold cross-validation method, or a bootstrap method, determines that the model's true positive rate, true negative rate, false positive rate, and false negative rate correspond to predetermined ranges. In yet another exemplary embodiment, the training model may be operational (i.e., available for active use by the organization) while continuing to train using new data. In yet another exemplary embodiment, the margin call model may be generated using at least one of an artificial neural network technique, a decision tree technique, a support vector machine technique, a Bayesian network technique, and a genetic algorithm technique.

[0078] In yet another exemplary embodiment, the margin call model may be trained based on at least one attribute, which may correspond to internally available data as well as externally retrieved data. The attributes are: ticket amount, platform amount, maturity amount, available collateral today, available collateral yesterday, total loan amount today, total loan amount yesterday, total haircuts today, total haircuts yesterday, accrued interest today, accrued interest yesterday, derivative initial margin (IM) exposure today, derivative market-to-market (MTM) exposure today, derivative IM exposure yesterday, derivative MTM exposure yesterday, total exposure today, total exposure yesterday, security liquidation value (LV) today, security LV yesterday, unlimited cross pledge today, unlimited cross pledge yesterday, restricted cross pledge LV today, restricted cross pledge LV yesterday, mortgaged property value LV, mortgage exposure, market value value (MV), amount of purchased securities for loan amount, amount of purchased securities for loan amount, amount of sold securities for loan amount, amount of purchased assets for loan amount, margin call threshold (amount), future value of debit cash, future value of debit cash for loan, collateral delta amount due to LV, collateral delta amount due to price, collateral delta amount due to foreign exchange (FX), cash delta amount due to FX, change in cash loan amount, artificial intelligence margin call factor amount, and margin call factor amount.

[0079] In step S408, the targeted individual's securities portfolio may be evaluated using the margin call model to determine a first set of information related to whether to trigger a margin call on the targeted securities portfolio. In an exemplary embodiment, a user may identify a first securities portfolio to the MCDCA module 302 for this evaluation via input to a graphical user interface. In yet another exemplary embodiment, the first set of information related to whether to trigger a margin call may include margin call prediction information such as a probability metric and information of an actual requirement for triggering a first margin call on the first securities portfolio. In yet another exemplary embodiment, the evaluation may also include determining a first data pattern, a classification of associated parameters, and a first characteristic of the first securities portfolio. The first data pattern and the first characteristic may correspond to the first securities portfolio and may be related to triggering a first margin call.

[0080] Next, in step S410, the MCDCA module 302 may display the first set of information in a graphical user interface via the display. In yet another exemplary embodiment, the first information may include the first data pattern, the first characteristic, and any other data related to the first securities portfolio. In yet another exemplary embodiment, the first information may be displayed on any connected computing device, such as, for example, a computer, laptop, smartphone, or smartwatch. The display of the information may include, for example, information that the analyst can use to determine whether to execute a margin call and / or information that provides specific advice regarding whether to execute a margin call. In yet another exemplary embodiment, the display of the information may include an alert, such as, for example, an alert notification of an impending margin call.

[0081] In yet another exemplary embodiment, the first securities portfolio may be received in a combined securities portfolio (i.e., one file containing multiple securities portfolios) or each securities portfolio may be received in a single, separate entry. In yet another exemplary embodiment, the first securities portfolio may be stored in a connected system and retrieved upon user request (i.e., a user may enter a first client number into a graphical user interface to analyze the first securities portfolio).

[0082] In yet another exemplary embodiment, the information related to triggering a margin call may include the probability of a margin call for a given portfolio of securities. In yet another exemplary embodiment, the information may include at least one characteristic related to and contributing to an expected margin call for a given portfolio of securities.

[0083] In yet another exemplary embodiment, a user may select a particular data pattern and corresponding feature from the graphical user interface. The selected data pattern and corresponding feature may be related to triggering a margin call for the currently analyzed securities portfolio. After the user selects the data pattern and corresponding feature, the graphical user interface may display the margin call information of interest based on the selection.

[0084] In yet another exemplary embodiment, the graphical user interface may display the margin call information in various formats, such as, for example, a dashboard. The dashboard may display additional information related to the securities portfolio, such as, for example, monitoring information and classification information related to established characteristics and margin call information related to the securities portfolio.

[0085] In yet another exemplary embodiment, the graphical user interface may include graphical elements, such as text boxes, selection surfaces, and flagging features, to receive feedback information from a user. This feedback information may include data such as a first information accuracy rank, a second information accuracy rank, and a feature accuracy rank. The feedback information may also include predicted information accuracy corresponding to a particular data pattern. In yet another exemplary embodiment, the feedback information may be used to update the margin call model to increase the prediction accuracy and analytical accuracy of the margin call model. As will be appreciated by those skilled in the art, the margin call model may be updated based on machine learning techniques, such as training the margin call model based on the feedback information.

[0086] FIG. 5 is a diagram 500 of a machine learning process that can be used to implement a method for identifying and capturing all relevant features involved in a margin call process so that the underlying causes can be automatically identified using machine learning, according to an example embodiment. FIG. 5 illustrates the interrelationship of various steps in the machine learning process. In an example embodiment, the steps may include "Data Collection" 502, "Clean / Transform" 504, "Exploration / Visualization" 506, "Model" 508, "Evaluation" 510, and "Deployment" 512, where different steps can be performed periodically to improve accuracy as more data is available from previous margin calls. New data is retrofitted into the system to learn from previous margin calls.

[0087] As shown in Figure 5, the machine learning process may begin with a data collection step 502. The collected data may then be cleaned to remove unnecessary information and converted into a processable data format in a clean / transform step 504. The clean / transformed data is then moved to an exploration / visualization step 506, where usable information, such as data patterns and features, is determined. At this step in the process, if the collected data may be insufficient to determine usable information, the process may return to the data collection step 502 to acquire additional data. However, once usable information has been determined, a model may be generated in a model step 508.

[0088] Once a model is generated, the process may return to the data collection step 502 to gather more data, to the clean / transform step 504 to further refine the data already collected, or to the evaluation step 510. In the evaluation step 510, the developed model may be tested to determine its accuracy. If the model is sufficiently accurate based on predetermined criteria, the process may proceed directly to the deployment step 512. However, if additional refinement is required, the process may return to the data collection step 502 or the clean / transform step 504. In the deployment step 512, the model may be used to analyze and predict outcomes for user inputs. After being deployed, the machine learning process is continuously refined using additional collected data, additional available information, and additional models.

[0089] FIG. 6 illustrates a graphical user interface 600 showing the variable importance of features identified in a method for identifying and acquiring all relevant features involved in a margin call process so that their underlying causes can be automatically identified using machine learning, according to an exemplary embodiment. The graphical user interface shown in FIG. 6 has the header "Variable Importance." The graphical user interface may also include a bar graph with an x-axis showing the percentage of importance numerically and a y-axis showing the contributing features textually. In an exemplary embodiment, the contributing features are arranged in ascending and descending order of relative importance. For example, the most important features may be placed at the top of the bar graph, or vice versa, the least important features may be placed at the bottom of the bar graph. In yet another exemplary embodiment, the relative importance may include the percentage of each feature's contribution to a trade segment, such as a margin call.

[0090] In yet another exemplary embodiment, the graphs shown may include any visual representation of the data, such as, for example, line graphs, pie charts, mosaic charts, spider charts, stacked bar graphs, etc. In yet another exemplary embodiment, the contribution features may be a security appreciation feature, a security depreciation feature, an asset market value discrepancy feature, a security market value discrepancy feature, a security market value buy feature, an exposure discrepancy feature, a cash market value discrepancy feature, a collateral discrepancy feature, a cash market value sell feature, a haircut discrepancy feature, a security market value sell feature, a security liquidation price buy feature, a cash market value appreciation feature, a cash market value depreciation feature, a pledge market value out feature, a pledge market value discrepancy feature, a cash market value buy feature, a derivative exposure feature, a The features may include at least one of a delta feature of a pledge exposure, a delta feature of a derivative initial margin exposure, a pledge liquidation price in feature, an accrued interest accumulation feature, a pledge market value in feature, a pledge market value accretion feature, a pledge market value depreciation feature, a delta feature of a derivative negative mark-to-market, a movement feature of loan amount changes due to foreign exchange, a due date feature of accrued interest debited to a private equity call or future value, and an open-ended pledge difference feature.

[0091] FIG. 7 is a screenshot illustrating a graphical user interface 700 that can be used to implement a method for identifying and capturing all relevant features involved in a margin call process so that their underlying causes can be automatically identified using machine learning, according to an exemplary embodiment. Graphical user interface 700 may include an exemplary graphical user interface that facilitates interaction between a user and various automatically generated margin call models. As shown, graphical user interface 700 may include a sortable table of multiple investors and their corresponding securities portfolios. In an exemplary embodiment, this table provides an aggregate view of the processed securities portfolio. In graphical user interface 700, this table may be sorted by categories titled "Ticket ID," "Client Name," "Age," "Aging Bucket," "Client ID," "Capital Advisor," "Credit Executive," "Margin Call Amount USD," "Margin Call Factor," and "Predicted Margin Call Factor." Graphical user interface 700 may also include user-selectable buttons, such as a filter button and a refresh button, to further refine the displayed information.

[0092] In an exemplary embodiment, information in the graphical user interface 700 may be displayed as any combination of letters, words, and numbers. In yet another exemplary embodiment, the aging bucket category may display information as a range of dates. In yet another exemplary embodiment, the margin call amount category may display any legally recognized fiat currency amount selectable by the user. In yet another exemplary embodiment, the margin call drivers category may display features known to have contributed to margin calls in the collected data. In yet another exemplary embodiment, the predicted margin call drivers category may display features predicted to contribute to future margin calls. In yet another exemplary embodiment, a comparison of the predicted margin call drivers category and the margin call driver categories for a particular securities portfolio may be used to validate a margin call model. In yet another exemplary embodiment, the predicted drivers of data may be sent to the graphical user interface 700 by data management software, such as SPLUNK, via an application programming interface (API), such as a Representational State Transfer (REST) ​​API.

[0093] 8 is a flowchart 800 of an exemplary process for implementing a method for identifying and obtaining all relevant features involved in a margin call process so that their root causes can be automatically identified using machine learning. As shown, flowchart 800 may include user interfaces operable on various user terminals labeled "System 1," "System 2," "System 3," and "System n." The user terminals may be connected to a system configured to implement a method for identifying and obtaining all relevant features involved in a margin call process so that their root causes can be automatically identified using machine learning.

[0094] The exemplary process in flowchart 800 begins in step S802 with the collection of features related to shared characteristics of margin calls. The collected features may include, for example, accrued interest FX rate fluctuations and market price fluctuations. Then, in step S804, a margin call model may be trained using artificial intelligence (AI) or machine learning (ML) techniques using the collected features. In step S806, margin call factors may be predicted using the trained margin call model. The predicted margin call factors may then be validated in step S808 using known outcomes to further improve the accuracy of the developed margin call model. Finally, in step S810, model retrofitting may be achieved by repeating the steps using additional collected features to further train the developed margin call model.

[0095] FIG. 9 illustrates a flowchart 900 of an exemplary process for implementing a method for identifying and capturing all relevant features involved in a margin call process so that their root causes can be automatically identified using machine learning. As shown, flowchart 900 begins with a business problem that can be solved using machine learning techniques. A machine learning problem formulation method may be used to formulate the business problem and identify possible technical solutions involving machine learning techniques that can be used to solve the business problem. Once the business problem is formulated, implementation of the machine learning techniques may begin with collecting data related to the business problem. The collected data may be integrated with other relevant data, cleaned, and prepared for data visualization and analysis. Feature engineering of the analyzed and visualized data uses domain knowledge of the data to create features related to attributes or properties of the business problem.

[0096] The engineered features may then be used to train a developed model and tune its parameters. The trained model may then be evaluated using known outcomes to determine whether the business objectives associated with the solution to the business problem are met. A determination that the business objectives are not met may result in a reconsideration of the engineered features and a re-implementation of the machine learning techniques, starting with data collection. However, if it is determined that the business objectives are not being met, the developed model may be deployed to predict and solve the business problem. After deployment, the model may be maintained using monitoring and debugging techniques that allow the model to be continually retrained with newly collected data.

[0097] 10 is a screenshot 1000 illustrating a graphical user interface for an unclassified ticket that can be used to implement a method for identifying and capturing all relevant features involved in a margin call process so that their root causes can be automatically identified using machine learning, according to an example embodiment. As shown in FIG. 10, the unclassified ticket is displayed in a margin call tool consistent with an embodiment of the present application. The margin call tool may include a table containing multiple unclassified tickets and a counter titled "Not Reviewed" that counts the number of unclassified tickets.

[0098] 11 is a screenshot 1100 illustrating a graphical user interface of an unclassified tickets dashboard that can be used to implement a method for identifying and capturing all relevant features involved in a margin call process so that their root causes can be automatically identified using machine learning, according to an example embodiment. As shown in FIG. 11, the unclassified tickets dashboard provides information for multiple tickets. In an example embodiment, the unclassified tickets dashboard shows tickets that have not yet been classified by artificial intelligence and may also include historical margin calls that have been previously classified.

[0099] FIG. 12 is a screenshot 1200 illustrating a classification confirmation graphical user interface that can be used to implement a method for identifying and capturing all relevant features involved in a margin call process so that root causes can be automatically identified using machine learning, according to an exemplary embodiment. As shown in FIG. 12, a confirmation email may be received within a certain period of time to confirm successful classification of the selected data. The confirmation email may include information corresponding to the selected classification data. As will be appreciated by those skilled in the art, the confirmation email may be received via any email client, such as the owner's email client or a third-party developed email client.

[0100] 13 is a screenshot 1300 illustrating a graphical user interface of classified tickets that can be used to implement a method for identifying and capturing all relevant features involved in a margin call process so that their root causes can be automatically identified using machine learning, according to an example embodiment. As shown in FIG. 13, classified tickets are displayed in a margin call tool consistent with embodiments of the present application. Tickets that have been successfully classified by artificial intelligence are displayed in the margin call tool.

[0101] 14A is a screenshot 1400A illustrating a graphical user interface of an artificial intelligence function that can be used to implement a method for identifying and capturing all relevant features involved in a margin call process using machine learning so that the underlying causes can be automatically identified, according to an example embodiment. As shown in FIG. 14A, classified tickets are displayed in a margin call tool consistent with embodiments of the present application. The margin call tool may also provide details of the available tickets, including the artificial intelligence classifiers and attributes.

[0102] 14B is a screenshot 1400B of a graphical user interface showing ticket details that can be used to implement a method for identifying and capturing all relevant features involved in a margin call process so that their root causes can be automatically identified using machine learning, according to an example embodiment. As shown in FIG. 14B, the classified tickets are displayed in a margin call tool consistent with embodiments of the present application. The margin call tool may also provide available ticket details for a particular ticket, such as the ticket identifier, the factors that led to the determination of the margin call on the ticket, and the associated attribute values ​​used to determine the factors for the margin call.

[0103] FIG. 15 is a screenshot 1500 illustrating a graphical user interface of a dashboard of categorized tickets that can be used to implement a method for identifying and capturing all relevant features involved in a margin call process so that their root causes can be automatically identified using machine learning, according to an example embodiment. As shown in FIG. 15 , categorized tickets are displayed in a dashboard consistent with an embodiment of the present application. The dashboard may display tickets categorized by artificial intelligence along with corresponding data for analysis. The data may include the number of valid margin calls, the number of invalid margin calls, and the total number of margin calls. The dashboard may also present the data in graphical elements, such as bar graphs, for example, by margin call cause by area.

[0104] FIG. 16 is a screenshot 1600 illustrating a model accuracy graphical user interface that can be used to implement a method for identifying and capturing all relevant features involved in the margin call process so that their root causes can be automatically identified using machine learning, according to an example embodiment. As shown in FIG. 16 , model analysis data is displayed in a dashboard consistent with embodiments of the present application. The dashboard may, for example, show the artificial intelligence analysis accuracy for the current model for a given factor of margin call, such as the factor of accrued interest. The model analysis data may include the number of valid margin calls, the number of invalid margin calls, and the total number of margin calls. The dashboard may also present data in graphical elements, such as bar graphs, for example, for margin call factors by region.

[0105] 17 is a screenshot 1700 illustrating an accuracy monitoring graphical user interface that can be used to implement a method for identifying and capturing all relevant features involved in the margin call process so that the underlying causes can be automatically identified using machine learning, according to an example embodiment. As shown in FIG. 17, historical model data is displayed in a dashboard consistent with embodiments of the present application. The dashboard may show a graphical illustration representing monitoring accuracy improvement as a function of time as more data is used for model training. The graphical illustration may include graphs, such as line graphs and bar charts, that track the increase in accuracy for the model as a function of time.

[0106] 18 is a show model 1800 illustrating feature interactions that can be used to implement a method for identifying and obtaining all relevant features involved in a margin call process so that their root causes can be automatically identified using machine learning, according to an exemplary embodiment. As shown in FIG. 18, interactions between variables and features may be highlighted in chart format. As will be appreciated by those skilled in the art, variables and features may be presented on either the X-axis or the Y-axis of the chart, with intersecting squares indicating interactions between a particular variable and a particular feature.

[0107] 19A is a screenshot 1900A illustrating a graphical user interface of a user feedback form that can be used to implement a method for identifying and capturing all relevant features involved in a margin call process so that the underlying causes can be automatically identified using machine learning, according to an example embodiment. Screenshot 1900A provides graphical user elements, such as menus or other utility windows, that are presented to the user.

[0108] As shown in Figure 19A, the graphical user elements may include a user feedback form related to the performance of the artificial intelligence classification. In an exemplary embodiment, data collected from the user feedback form may be used to further train a margin call model using techniques consistent with embodiments herein to improve classification accuracy. The user feedback form may include information corresponding to a particular ticket as well as user-selectable graphical elements for receiving user input.

[0109] 19B is a screenshot 1900B illustrating an enhanced graphical user interface displayed based on user input in a user feedback form that can be used to implement a method for identifying and capturing all relevant features involved in a margin call process so that their root causes can be automatically identified using machine learning, according to an exemplary embodiment. Screenshot 1900B provides enhanced graphical user elements, such as menus or other utility windows, that are presented to the user. The enhanced graphical user elements may be presented to the user based on user input received from the graphical user elements of FIG. 19A.

[0110] As shown in FIG. 19B, the graphical user element may include an enhanced user feedback form related to the outcome of the artificial intelligence classification. In an exemplary embodiment, data collected from the enhanced user feedback form may be used to further train a margin call model using techniques consistent with embodiments herein to improve classification accuracy. The enhanced user feedback form may display information based on user selections in previous graphical user elements. The enhanced user feedback form may include additional information corresponding to a particular ticket as well as additional user-selectable graphical elements for receiving user input.

[0111] Thus, this technique provides an optimized process for identifying and capturing all relevant features involved in the margin call process so that their root causes can be automatically identified using machine learning.

[0112] While the present invention has been described with reference to certain exemplary embodiments, the terms used are terms of description and illustration, rather than of limitation. Changes may be made within the purview of the appended claims without departing from the scope and spirit of the present disclosure in this aspect, as described and amended herein. While the present invention has been described with reference to particular methods, materials, and embodiments, the invention is not intended to be limited to the particular disclosures, and the invention extends, for example, to all functionally equivalent structures, methods, and uses within the scope of the appended claims.

[0113] For example, while a computer-readable medium may be described as a single medium, the term "computer-readable medium" includes a single medium or multiple media, such as a centralized or distributed database, and / or associated caches and servers that store one or more sets of instructions. The term "computer-readable medium" is also intended to include any medium capable of storing, encoding, or executing a set of instructions for execution by a processor, or any medium that causes a computer system to perform any one or more of the embodiments disclosed herein.

[0114] The computer-readable medium may comprise one or more non-transitory computer-readable media and / or one or more transitory computer-readable media. In certain non-limiting exemplary embodiments, the computer-readable medium may include a solid-state memory, such as a memory card, or other package containing one or more non-volatile read-only memories. The computer-readable medium may also be a random access memory or other volatile re-writable memory. In addition, the computer-readable medium may include a magneto-optical or optical medium, such as a disk or tape, or other storage device that captures a carrier wave signal, such as a signal communicated over a transmission medium. Accordingly, the present disclosure is considered to include any computer-readable medium, or other equivalent and successor medium, on which data or instructions may be recorded.

[0115] Although this application describes certain embodiments that may be implemented as a computer program or code segments on a computer-readable medium, it is understood that dedicated hardware implementations, such as application specific integrated circuits, programmable logic arrays, and other hardware devices, can be configured to implement one or more embodiments described herein. Applications that include the various embodiments described herein may broadly encompass a variety of electronic and computer systems. Thus, this application may encompass software, firmware, and hardware implementations, or combinations thereof. Nothing in this application should be interpreted as being implemented or capable of being implemented solely by software and not implemented in hardware.

[0116] Although this specification describes components and functions that may be implemented in particular embodiments with reference to particular standards and protocols, the disclosure is not limited to such standards and protocols. Such standards are periodically replaced by faster or more efficient equivalents having essentially the same functionality. Accordingly, replacement standards and protocols having the same or similar functionality are considered equivalents thereof.

[0117] The illustrations of the embodiments described herein are intended to provide a general understanding of various embodiments. The illustrations are not intended to be a complete description of all elements and features of apparatus and systems that utilize the structures or methods described herein. Many other embodiments will be apparent to those skilled in the art upon consideration of the present disclosure. Other embodiments may be utilized and derived from the present disclosure. Consequently, structural and logical substitutions and changes may be made without departing from the scope of the present disclosure. Additionally, the illustrations are merely representational and may not be drawn to scale. Certain proportions in the illustrations may be exaggerated, while other proportions may be minimized. Therefore, the present disclosure and the figures should be considered illustrative, not restrictive.

[0118] One or more embodiments of the present disclosure may be referred to herein individually and / or collectively by the term "invention" for convenience only, and without any intention to limit the scope of the present application to any particular invention or inventive concept. Furthermore, while specific embodiments have been shown and described herein, it should be understood that any subsequent constructions designed to achieve the same or similar purpose may be substituted for the specific embodiment shown. The present disclosure is intended to cover any and all subsequent modifications or variations of the various embodiments. Combinations of the above embodiments, as well as other embodiments not specifically described herein, will be apparent to those skilled in the art upon consideration of this description.

[0119] The Abstract is submitted with the understanding that it will not be used to interpret or limit the scope or meaning of the claims. Moreover, in the foregoing Detailed Description, various features may be grouped together or described in a single embodiment for the purpose of streamlining the disclosure. This disclosure is not to be interpreted as implying an intention that the claimed embodiments require more features than are expressly recited in each claim. Rather, as the following claims reflect, inventive subject matter may be directed to less than all of the features of any of the disclosed embodiments. Accordingly, the following claims are incorporated into the Detailed Description, with each claim standing on its own merits in defining each claimed subject matter.

[0120] The subject matter disclosed above is considered to be illustrative and not restrictive. The appended claims are intended to cover all such modifications, enhancements, and other embodiments that fall within the true spirit and scope of the present disclosure. Accordingly, to the maximum extent permitted by law, the scope of the present disclosure shall be determined by the broadest permissible interpretation of the following claims and their equivalents, and is not limited or constrained by the above detailed description. The present invention includes the following embodiments. [Aspect 1] 1. A method of providing information related to margin call triggers for a first portfolio of securities, comprising: The method is implemented by at least one processor; The method comprises: obtaining, by the at least one processor, a plurality of data relating to a plurality of securities portfolios; automatically determining, by the at least one processor, using the plurality of data to determine at least one data pattern related to triggering a margin call and at least one corresponding feature contributing to triggering the margin call; automatically generating, by the at least one processor, at least one margin call model based on the at least one data pattern and the at least one feature; evaluating, by the at least one processor, the first securities portfolio with the at least one margin call model to determine first information related to whether to trigger a first margin call on the first securities portfolio; A method comprising: [Aspect 2] In the method of embodiment 1, The evaluating step includes: selecting, by the at least one processor, a first one of the at least one data pattern and a first one of the at least one corresponding feature related to triggering the first margin call for the first portfolio of securities; displaying, by the at least one processor, the first information in a graphical user interface together with the selected first one of the at least one data pattern and the selected first one of the at least one corresponding feature via a display; The method further comprises: [Aspect 3] In the method of embodiment 1, The method, wherein the at least one margin call model is automatically generated using machine learning analysis of the plurality of data. [Aspect 4] In the method of embodiment 1, 10. The method of claim 1, wherein the at least one margin call model comprises at least one of a data model, a mathematical model, a process model, and a machine learning model. [Aspect 5] In the method of embodiment 1, evaluating, by the at least one processor, a second portfolio of securities using the at least one margin call model to determine second information related to whether to trigger a future second margin call on the second portfolio of securities; displaying, by the at least one processor, the second information in a graphical user interface via a display; and The method further comprises: [Aspect 6] In the method of embodiment 5, The graphical user interface displays a list of securities related to the first portfolio of securities. and at least one dashboard displaying at least one of monitoring information and classification information related to the at least one characteristic. [Aspect 7] In the method of embodiment 1, The at least one feature may be a security appreciation feature, a security depreciation feature, an asset market value discrepancy feature, a security market value discrepancy feature, a security market value buy feature, an exposure discrepancy feature, a cash market value discrepancy feature, a collateral discrepancy feature, a cash market value sell feature, a haircut discrepancy feature, a security market value sell feature, a security liquidation price buy feature, a cash market value appreciation feature, a cash market value depreciation feature, a pledge market value out feature, a pledge market value discrepancy feature, a cash market value buy feature, a derivative exposure the method comprising at least one of a delta feature of a pledge, a delta feature of a derivative initial margin exposure, an in feature of a pledge liquidation price, an accumulation feature of accrued interest, an in feature of a pledge market value, an accretion feature of a pledge market value, a depreciation feature of a pledge market value, a delta feature of a negative mark-to-market of a derivative, a movement feature of loan amount changes due to foreign exchange, a due date feature of accrued interest debited to a private equity call or future value, and a difference feature of an unlimited pledge. [Aspect 8] In the method of embodiment 1, receiving, by the at least one processor, feedback information from at least one user via a graphical user interface; updating, by the at least one processor, the at least one margin call model based on the feedback information; The method further comprises: [Aspect 9] In a method according to embodiment 8, The method, wherein the feedback information includes at least one of a first information accuracy rank, a second information accuracy rank, and a feature accuracy rank. [Aspect 10] In the method of embodiment 1, analyzing, by the at least one processor, the plurality of data; identifying, by the at least one processor, personally identifiable information in the analyzed plurality of data; anonymizing, by the at least one processor, the plurality of data by removing the identified personally identifiable information; The method further comprises: [Aspect 11] 1. A computing device implemented to perform a method for providing information related to the triggering of a margin call for a first portfolio of securities, the method comprising: the computing device, a processor; Memory and a communication interface connected to each of the processor and the memory; Equipped with The processor: Acquire multiple data related to multiple securities portfolios; using the plurality of data to automatically determine at least one data pattern related to triggering a margin call and at least one corresponding feature contributing to triggering the margin call; automatically generating at least one margin call model based on the at least one data pattern and the at least one feature; and determining first information relating to whether to trigger a first margin call on the first securities portfolio. configured to evaluate using at least one margin call model; Computing devices. [Aspect 12] 12. The computing device of claim 11, For the evaluation, the processor further comprises: selecting a first one of the at least one data pattern and a first one of the at least one corresponding feature related to triggering the first margin call for the first portfolio of securities; configured to display the first information in a graphical user interface together with the selected first one of the at least one data pattern and the selected first one of the at least one corresponding feature via a display. Computing devices. [Aspect 13] In the computing device according to aspect 11, the processor is further configured to automatically generate the at least one margin call model using machine learning analysis of the plurality of data, a computing device. [Aspect 14] In the computing device according to aspect 11, the at least one margin call model includes at least one of a data model, a mathematical model, a process model, and a machine learning model, a computing device. [Aspect 15] In the computing device according to aspect 11, the processor is further evaluating a second securities portfolio using the at least one margin call model to determine second information related to whether a future second margin call will be triggered for the second securities portfolio, configured to display the second information on a graphical user interface via a display, a computing device. [Aspect 16] In the computing device according to aspect 15, the graphical user interface includes at least one dashboard that displays at least one of monitoring information related to the first securities portfolio and classification information related to the at least one feature, a computing device. [Aspect 17] In the computing device according to aspect 11, The at least one feature may be a security appreciation feature, a security depreciation feature, an asset market value discrepancy feature, a security market value discrepancy feature, a security market value buy feature, an exposure discrepancy feature, a cash market value discrepancy feature, a collateral discrepancy feature, a cash market value sell feature, a haircut discrepancy feature, a security market value sell feature, a security liquidation price buy feature, a cash market value appreciation feature, a cash market value depreciation feature, a pledge market value out feature, a pledge market value discrepancy feature, a cash market value buy feature, a derivative exposure delta ... a computing device comprising at least one of a derivative initial margin exposure delta feature, a pledge liquidation price in feature, an accrued interest accumulation feature, a pledge market value in feature, a pledge market value accretion feature, a pledge market value depreciation feature, a derivative negative mark-to-market delta feature, a foreign exchange driven loan amount movement feature, a private equity call or future value debited accrued interest due date feature, and an unlimited pledge difference feature. [Aspect 18] 12. The computing device of claim 11, The processor further comprises: receiving feedback information from at least one user via the graphical user interface; configured to update the at least one margin call model based on the feedback information. Computing devices. [Aspect 19] 19. The computing device of claim 18, The computing device, wherein the feedback information includes at least one of a first information accuracy rank, a second information accuracy rank, and a feature accuracy rank. [Aspect 20] 12. The computing device of claim 11, The processor further comprises: Analyzing the plurality of data; Identifying information that can identify an individual from the analyzed plurality of data; and anonymizing the plurality of data by removing the identified personally identifiable information. Computing devices.

Claims

1. 1. A method of providing information related to margin call triggers for a first portfolio of securities, comprising: The method is implemented by at least one processor; The method comprises: obtaining, by the at least one processor, a plurality of data related to a plurality of securities portfolios from publicly accessible sources and privately held sources; analyzing, by the at least one processor, the plurality of data; identifying, by the at least one processor, personally identifiable information corresponding to at least one investor in the analyzed plurality of data; removing, by the at least one processor, the identified personally identifiable information from the plurality of data and anonymizing the remaining data of the plurality of data; automatically determining, by the at least one processor, using the anonymized data, at least one data pattern related to triggering a margin call and at least one corresponding feature contributing to triggering the margin call; identifying, by the at least one processor, a contributing factor to at least one margin call from the anonymized data; determining, by the at least one processor, a degree of influence for each of at least one of the contribution factors, the degree of influence including a percentage of contribution to a margin call and a probability of contribution to a margin call; determining, by the at least one processor, at least one corresponding characteristic from at least one of the contributing factors based on the degree of influence; and automatically generating, by the at least one processor, at least one margin call model in memory based on the at least one data pattern and the at least one corresponding feature of a plurality of securities portfolios to output first information regarding whether a first margin call should be triggered for the securities portfolio when the securities portfolio is input into the at least one margin call model, wherein the at least one data pattern and the at least one corresponding feature include margin call root causes based on the contributing factors and degree of influence to the margin call, and the first information includes a probability metric and corresponding trigger requirement for the input securities portfolio; training, by said at least one processor, at least one margin call model using the additional data; determining, by the at least one processor, an accuracy of the trained at least one margin calling model; determining whether the accuracy of the at least one trained margin call model is sufficient based on a predetermined criterion, where the predetermined criterion is that the true positive rate, true negative rate, false positive rate, and false negative rate output by the at least one trained margin call model fall within predetermined ranges; if the accuracy of the at least one trained margin call model is determined to be insufficient, iteratively collecting additional data and modifying the at least one trained margin call model incorporating the additional data until the accuracy of the at least one trained margin call model is determined to be sufficient based on predetermined criteria; deploying the trained at least one margin call model if the accuracy of the trained at least one margin call model is determined to be sufficient; and evaluating, by the at least one processor, using the deployed trained at least one margin call model to determine first information related to whether to trigger a first margin call on the first securities portfolio, wherein the first securities portfolio is evaluated using the deployed trained at least one margin call model when the first securities portfolio is input to the trained at least one margin call model; and the first information includes a probability metric and a corresponding trigger requirement; determining, by the at least one processor, at least one advisory action based on the first information using the deployed trained at least one margin call model, including whether to execute a margin call; receiving, by the at least one processor, feedback information from at least one user via a graphical user interface; improving accuracy of the deployed trained at least one margin call model based on the automatically determined at least one advisory action and the feedback information; and A method comprising:

2. 10. The method of claim 1, The method, wherein the at least one margin call model is automatically generated using machine learning analysis of the anonymized data.

3. 10. The method of claim 1, 10. The method of claim 1, wherein the at least one margin call model comprises at least one of a data model, a mathematical model, a process model, and a machine learning model.

4. 10. The method of claim 1, evaluating, by the at least one processor, a second portfolio of securities using the at least one margin call model to determine second information related to whether to trigger a future second margin call on the second portfolio of securities; displaying, by the at least one processor, the second information in a graphical user interface via a display; and The method further comprises:

5. 5. The method of claim 4, the graphical user interface includes at least one dashboard that displays at least one of monitoring information related to the first securities portfolio and classification information related to the at least one corresponding characteristic.

6. 10. The method of claim 1, The at least one corresponding feature may be a security appreciation feature, a security depreciation feature, an asset market value discrepancy feature, a security market value discrepancy feature, a security market value buy feature, an exposure discrepancy feature, a cash market value discrepancy feature, a collateral discrepancy feature, a cash market value sell feature, a haircut discrepancy feature, a security market value sell feature, a security liquidation price buy feature, a cash market value appreciation feature, a cash market value depreciation feature, a pledge market value out feature, a pledge market value discrepancy feature, a cash market value buy feature, a derivative exposure feature, a the method comprising at least one of a delta feature of pledge, a delta feature of derivative initial margin exposure, an in feature of pledge liquidation price, an accumulation feature of accrued interest, an in feature of market value of pledge, an accretion feature of market value of pledge, a depreciation feature of market value of pledge, a delta feature of negative mark-to-market of derivatives, a movement feature of loan amount changes due to foreign exchange, a due date feature of accrued interest debited on private equity call or future value, and a difference feature of unlimited pledge.

7. 10. The method of claim 1, The method, wherein the feedback information includes at least one of a first information accuracy rank, a second information accuracy rank, and a feature accuracy rank.

8. 1. A computing device implemented to perform a method for providing information related to the triggering of a margin call for a first portfolio of securities, the computing device comprising: the computing device, a processor; Memory and a communication interface connected to each of the processor and the memory; Equipped with The processor: Obtaining a plurality of data relating to a plurality of securities portfolios from publicly accessible and privately held sources; Analyzing the plurality of data; Identifying personally identifiable information corresponding to at least one investor in the analyzed plurality of data; removing the identified personally identifiable information from the plurality of data sets and anonymizing the remaining data of the plurality of data sets; using the anonymized data to automatically determine at least one data pattern related to triggering a margin call and at least one corresponding feature contributing to triggering the margin call; Identifying contributing factors to at least one margin call from the anonymized data; and determining a degree of influence for each of at least one of the contribution factors, the degree of influence including a percentage of contribution to a margin call and a probability of contribution to a margin call; determining at least one corresponding characteristic from the at least one contributing factor based on the degree of influence; It will be automatically confirmed by automatically generating in the memory at least one margin call model based on the at least one data pattern and the at least one corresponding feature of a plurality of securities portfolios to output first information regarding whether a first margin call should be triggered for the securities portfolio when the securities portfolio is input into the at least one margin call model, wherein the at least one data pattern and the at least one corresponding feature include margin call root causes based on the contributing factors and degrees of influence to the margin call, and the first information includes a probability metric and corresponding trigger requirements for the input securities portfolio; Train at least one margin call model using the additional data; determining the accuracy of the at least one trained margin calling model; determining whether the accuracy of the at least one trained margin call model is sufficient based on a predetermined criterion, the predetermined criterion being that the true positive rate, true negative rate, false positive rate, and false negative rate output by the at least one trained margin call model fall within predetermined ranges; if the accuracy of the at least one trained margin call model is determined to be insufficient, iteratively collecting additional data and modifying the at least one trained margin call model incorporating the additional data until the accuracy of the at least one trained margin call model is determined to be sufficient based on predetermined criteria; deploying the trained at least one margin call model if the accuracy of the trained at least one margin call model is determined to be sufficient; wherein the evaluating step includes determining first information related to whether to trigger a first margin call on the first securities portfolio using the deployed trained at least one margin call model, and wherein the first securities portfolio is input to the trained at least one margin call model and the first securities portfolio is evaluated using the deployed trained at least one margin call model; moreover, the first information includes a probability metric and a corresponding trigger requirement; The processor: using the deployed trained at least one margin call model to determine at least one advisory action based on the first information, including whether or not to execute a margin call; receiving feedback information from at least one user via the graphical user interface; improving accuracy of the deployed trained at least one margin call model based on the automatically determined at least one advisory action and the feedback information; Computing devices.

9. 9. The computing device of claim 8, The computing device, wherein the processor is further configured to automatically generate the at least one margin call model using machine learning analysis of the anonymized data.

10. 9. The computing device of claim 8, The at least one margin call model comprises at least one of a data model, a mathematical model, a process model, and a machine learning model.

11. 9. The computing device of claim 8, The processor further comprises: evaluating a second portfolio of securities using the at least one margin call model to determine second information related to whether to trigger a second future margin call on the second portfolio of securities; configured to display the second information in a graphical user interface via a display; Computing devices.

12. 12. The computing device of claim 11, a computing device, wherein the graphical user interface includes at least one dashboard that displays at least one of monitoring information related to the first securities portfolio and classification information related to the at least one corresponding characteristic.

13. 9. The computing device of claim 8, The at least one corresponding feature may be a security appreciation feature, a security depreciation feature, an asset market value discrepancy feature, a security market value discrepancy feature, a security market value buy feature, an exposure discrepancy feature, a cash market value discrepancy feature, a collateral discrepancy feature, a cash market value sell feature, a haircut discrepancy feature, a security market value sell feature, a security liquidation price buy feature, a cash market value appreciation feature, a cash market value depreciation feature, a pledge market value out feature, a pledge market value discrepancy feature, a cash market value buy feature, a derivative exposure A computing device comprising at least one of a delta feature, a derivative initial margin exposure delta feature, a pledge liquidation price in feature, an accrued interest accumulation feature, a pledge market value in feature, a pledge market value accretion feature, a pledge market value depreciation feature, a derivative negative mark-to-market delta feature, a foreign exchange driven loan amount movement feature, a private equity call or future value debit accrued interest due date feature, and an unlimited pledge difference feature.

14. 9. The computing device of claim 8, The computing device, wherein the feedback information includes at least one of a first information accuracy rank, a second information accuracy rank, and a feature accuracy rank.

Images