Computer implementation methods, information handling systems, and computer programs

Abstract

To provide a computer implementation method, an information handling system, and a computer program product for retaining data integrity in a cognitive multi agent system.SOLUTION: There is provided an approach to apply a first axiom to a set of data points for generating a first set of outputs by a first node. This approach applies a second axiom to a set of data points for generating a second set of outputs by a second node. The first node and the second node are parts of a computer network having a plurality of nodes. This approach calculates a first nuance on the basis of a set having disparity between the first set of outputs and the second set of outputs, and adjusts reliability of the first node in the computer network on the basis of the first nuance.SELECTED DRAWING: Figure 4

Description

[Technical Field]

[0001] This invention relates to maintaining data integrity in cognitive multi-agent systems. [Background technology]

[0002] Advances in computing hardware technology have led to a surge in the use of intelligent / cognitive computing devices for automation. The majority of these systems rely on sensors to collect the data necessary to automate decision-making processes. For example, in a nuclear power plant, manual temperature monitoring can be replaced by an array of thermal sensors and a programmable logic controller (PLC) for releasing coolant. Based on the temperature data collected from these sensors, the PLC can limit the manual interventions needed to maintain the reactor temperature. This can make a significant difference in the event of a catastrophe such as a reactor meltdown. Therefore, maintaining data integrity between these devices is crucial to ensuring the quality of automated decision-making. In a scenario where four out of ten thermal sensors malfunction, detecting the malfunctioning sensors becomes critical.

[0003] In many cases, these sensors / cognitive devices have their own cognitive processes that detect phenomena (e.g., temperature) and map them to numerical values. These devices are typically deployed as a network, allowing data points such as temperature values ​​to be shared among them. This data sharing enables large-scale automation. For example, a network of 50 devices can be used to estimate weather measurements for an area of ​​approximately 100 square kilometers. A set of N devices (e.g., N=5) can be responsible for every 10 square kilometers. By sharing weather data among N adjacent devices within an area, aggregated weather measurements for that area can be estimated. Each node within an area can determine its own accuracy based on the data transmitted by the N-1 adjacent nodes. [Overview of the project] [Problems that the invention aims to solve]

[0004] This invention provides a computer implementation method, an information handling system, and a computer program product for maintaining data integrity in cognitive multi-agent systems. [Means for solving the problem]

[0005] According to one embodiment of the present disclosure, an approach is provided which, by a first node, applies a first axiom to a set of data points in order to generate a first set of outputs. The approach, by a second node, applies a second axiom to the set of data points in order to generate a second set of outputs. The first node and the second node are part of a computer network comprising a plurality of nodes. The approach calculates a first nuance based on a set of discrepancies between the first set of outputs and the second set of outputs, and adjusts the reliability of the first node in the computer network based on the first nuance.

[0006] The above is a summary and therefore inevitably includes simplifications, generalizations, and omissions of details. As a result, those skilled in the art will understand that the summary is merely illustrative and not intended to be restrictive in any way. Other aspects, inventive features, and advantages of this disclosure, defined solely by the claims, will become apparent in the non-restrictive detailed description set forth below.

[0007] This disclosure will be better understood by referring to the accompanying drawings, and many of its purposes, features, and advantages will be apparent to those skilled in the art. [Brief explanation of the drawing]

[0008] [Figure 1] Figure 1 is a block diagram of a data processing system capable of implementing the method described herein. [Figure 2] Figure 2 provides an extension of the information handling system environment shown in Figure 1, illustrating that the methods described herein can be implemented in a wide variety of information handling systems operating in a network environment. [Figure 3] Figure 3 shows an example of a data integrity system that maintains the integrity of a multi-agent system. [Figure 4] Figure 4 is a flowchart illustrating an example of the procedure taken to evaluate the nuances of a node in a cognitive multi-agent system based on its axioms, calculate its reliability, and iteratively update the node's axioms accordingly. [Figure 5] Figure 5 shows an example of a data integrity system 350 that determines inconsistencies within nodes over time and calculates nuances of the nodes based on these inconsistencies. [Figure 6] Figure 6 shows an example of the relationship between node axioms and nuances. [Figure 7] Figure 7 shows an example of the relationship between nodes, axioms, nuances, and reliability. [Figure 8]Figure 8 is an illustrative diagram showing two embodiments of a cognitive multi-agent system in which the node axioms are iteratively updated based on nuance and reliability. [Modes for carrying out the invention]

[0009] The terms used herein are for the purpose of describing specific embodiments only and are not intended to limit them. Where used herein, the singular forms "a," "an," and "the" are intended to include the plural form unless the context explicitly indicates otherwise. Where used herein, the terms "comprises" or "comprising" or both specify the presence of a described feature, integer, step, operation, element, or component or combination thereof, but do not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components, or groups or combinations thereof.

[0010] All corresponding structures, materials, actions, and equivalent means or step-plus-function elements in the following claims are intended to include any structures, materials, or actions for performing a function in combination with other claimed elements, as specifically claimed. This disclosure is presented for illustrative and explanatory purposes, but is not intended to be exhaustive or to be limited to the disclosed forms. Many modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of this disclosure. The embodiments have been selected and described to best illustrate the principles and practical applications of this disclosure and to enable those skilled in the art to understand this disclosure in terms of various embodiments with various modifications suitable for the particular use to be intended.

[0011] The present invention may be a system, method, or computer program product or combination thereof, integrated at any possible level of technical detail. The computer program product may include a computer-readable storage medium storing computer-readable program instructions for causing a processor to perform aspects of the present invention.

[0012] A computer-readable storage medium can be a tangible device capable of holding and storing instructions used by an instruction execution device. Examples of computer-readable storage media may be electronic storage devices, magnetic storage devices, optical storage devices, electromagnetic storage devices, semiconductor storage devices, or appropriate combinations thereof. More specific examples of computer-readable storage media include portable computer diskettes, hard disks, RAM, ROM, EPROM (or flash memory), SRAM, CD-ROM, DVD, memory stick, floppy disk, punch cards, or grooved raised structures, and mechanically encoded devices on which instructions are recorded, and appropriate combinations thereof. Computer-readable storage devices as used herein should not be interpreted as transient signals themselves, such as radio waves or other freely propagating electromagnetic waves, electromagnetic waves propagating through waveguides or other transmission media (e.g., light pulses passing through optical fiber cables), or electrical signals transmitted through wires.

[0013] The computer-readable program instructions described herein can be downloaded from a computer-readable storage medium to each computer device / processor. Alternatively, they can be downloaded to an external computer or external storage device via a network (e.g., the Internet, LAN, WAN, or wireless network, or a combination thereof). The network may include copper transmission cables, optical transmission fibers, wireless transmission, routers, firewalls, switches, gateway computers or edge servers, or a combination thereof. A network adapter card or network interface within each computer device / processor receives computer-readable program instructions from the network and transfers them for storage in a computer-readable storage medium in the respective computer device / processor.

[0014] The computer-readable program instructions for carrying out the operations of the present invention may be source code or object code written in any combination of one or more programming languages, including assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, state-setting data, configuration data for integrated circuits, or object-oriented programming languages such as Smalltalk and C++, and procedural programming languages such as the "C" programming language and similar programming languages. The computer-readable program instructions may be executed entirely on the user's computer as a stand-alone software package, or partially on the user's computer. Alternatively, they may be executed partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the latter case, the remote computer may be connected to the user's computer via any type of network, including a LAN or WAN, or to an external computer (e.g., via the Internet using an Internet service provider). In some embodiments, an electronic circuit, including, for example, a programmable logic circuit, a field-programmable gate array (FPGA), or a programmable logic array (PLA), can execute the computer-readable program instructions by utilizing the state information of the computer-readable program instructions to customize the electronic circuit for the purpose of implementing aspects of the present invention.

[0015] Aspects of the present invention are described herein with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. Each block in the flowchart illustrations and / or block diagrams, and combinations of multiple blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer-readable program instructions.

[0016] The above computer-readable program instructions may be provided to a processor of a computer or other programmable data processing apparatus for producing a machine. Thereby, these instructions executed via the processor of such computer or other programmable data processing apparatus create means for performing the functions / operations specified in one or more blocks in a flowchart or a block diagram or both. The above computer-readable program instructions may further be stored in a computer-readable storage medium that can be instructed to function in a specific manner for a computer, a programmable data processing apparatus or other apparatus or a combination thereof. Thereby, the computer-readable storage medium in which the instructions are stored constitutes a product containing instructions for performing the modes of functions / operations specified in one or more blocks in a flowchart or a block diagram or both.

[0017] Also, a computer-executable process may be generated by loading the computer-readable program instructions into a computer, other programmable apparatus, or other apparatus and causing a series of operational steps to be performed on the computer, other programmable apparatus, or other apparatus. Thereby, the instructions executed on the computer, other programmable apparatus, or other apparatus perform the functions / operations specified in one or more blocks in a flowchart or a block diagram or both.

[0018] The flowcharts and block diagrams in the figures illustrate the configuration, functions, and operations of executable modes of systems, methods, and computer program products according to various embodiments of the present invention. In this regard, each block in a flowchart or block diagram may represent a module, segment, or part of an instruction, which constitutes one or more executable instructions for implementing a specified logical function. In some alternative embodiments, the functions shown in the blocks may differ from the order shown in the figures. For example, two consecutively shown blocks may actually be achieved as a single process, executed simultaneously or substantially simultaneously, executed in a partially or entirely overlapping manner in time, or, in some cases, executed in reverse order, depending on the functions involved. It should also be noted that each block in a block diagram or flowchart diagram or both, and any combination of blocks in a block diagram or flowchart diagram or both, can be implemented by a special-purpose hardware-based system that performs a specified function or operation, or a combination of special-purpose hardware and computer instructions. The following detailed description generally follows the above-mentioned overview of the present disclosure and further explains and expands the definitions of various modes and embodiments of the present disclosure as needed.

[0019] Figure 1 shows an information handling system 100, a simplified example of a computer system capable of performing the computing operations described herein. The information handling system 100 includes one or more processors 110 coupled to a processor interface bus 112. The processor interface bus 112 connects the processors 110 to a northbridge 115, also known as a memory controller hub (MCH). The northbridge 115 connects to system memory 120 and provides means for the (one or more) processors 110 to access the system memory. A graphics controller 125 also connects to the northbridge 115. In one embodiment, a peripheral component interconnect (PCI) express bus 118 connects the northbridge 115 to the graphics controller 125. The graphics controller 125 connects to a display device 130, such as a computer monitor.

[0020] The northbridge 115 and southbridge 135 are connected to each other using bus 119. In some embodiments, the bus is a Direct Media Interface (DMI) bus that transfers data at high speed in each direction between the northbridge 115 and the southbridge 135. In some embodiments, a PCI bus connects the northbridge and the southbridge. The southbridge 135, also known as the Input / Output (I / O) Controller Hub (ICH), is generally a chip that implements the ability to operate at a slower speed than the capabilities provided by the northbridge. The southbridge 135 generally provides a variety of buses used to connect various components. These buses include, for example, PCI and PCI Express buses, ISA buses, System Management Buses (SMBus or SMB), or Low Pin Count (LPC) buses or combinations thereof. The LPC bus often connects low-bandwidth devices such as the boot ROM 196 and “legacy” I / O devices (using “super I / O” chips). "Legacy" I / O devices (198) may include, for example, serial and parallel ports, keyboards, mice, or floppy disk controllers or a combination thereof. Other components often included in the Southbridge 135 are a Direct Memory Access (DMA) controller, a Programmable Interrupt Controller (PIC), and a Storage Device Controller that uses bus 184 to connect the Southbridge 135 to a non-volatile storage device 185 such as a hard disk drive.

[0021] The ExpressCard 155 is a slot for connecting hot-pluggable devices to the information handling system. The ExpressCard 155 connects to the Southbridge 135 using both USB and PCI Express buses, thus supporting both PCI Express and USB (Universal Serial Bus) connections. The Southbridge 135 includes a USB controller 140 that provides USB connectivity to USB-connected devices. These devices include a webcam (camera) 150, an infrared (IR) receiver 148, a keyboard and trackpad 144, and a Bluetooth device 146 that provides a wireless personal area network (PAN). The USB controller 140 also provides USB connectivity to other miscellaneous USB-connected devices 142, such as a mouse, a removable non-volatile storage device 145, a modem, a network card, an Integrated Services Digital Network (ISDN) connector, a fax machine, a printer, a USB hub, and many other types of USB-connected devices. Although the removable non-volatile storage device 145 is shown as a USB-connected device, it may be connected using different interfaces, such as a Firewire interface.

[0022] The wireless local area network (LAN) device 175 connects to the southbridge 135 via PCI or PCI Express bus 172. The LAN device 175 implements one of the wireless modulation techniques of the IEEE 802.11 standard, which typically uses the same protocol for wireless communication between the information handling system 100 and another computer system or device. The optical storage device 190 connects to the southbridge 135 using the Serial ATA (SATA) bus 188. The Serial ATA adapter and devices communicate over a high-speed serial link. The Serial ATA bus also connects the southbridge 135 to other forms of storage devices, such as hard disk drives. An audio circuit 160, such as a sound card, connects to the southbridge 135 via bus 158. The audio circuit 160 also provides functionality related to audio hardware, such as audio line-in and optical digital audio input ports 162, optical digital output and headphone jack 164, an internal speaker 166, and an internal microphone 168. The Ethernet controller 170 connects to the Southbridge 135 using a bus such as a PCI bus or PCI Express bus. The Ethernet controller 170 connects the information handling system 100 to computer networks such as a local area network (LAN), the Internet, and other public and private computer networks.

[0023] Figure 1 shows one information handling system, but information handling systems can take many forms. For example, an information handling system can take the form of a desktop, server, portable, laptop, notebook, or other form factor computer or data processing system. Furthermore, an information handling system can take other form factors, such as a personal digital assistant (PDA), a gaming device, an automated teller machine (ATM), a mobile phone device, a communication device, or other devices containing a processor and memory.

[0024] Figure 2 provides an extension of the information handling system environment shown in Figure 1, illustrating that the methods described herein can be implemented with a wide variety of information handling systems operating in a network environment. The types of information handling systems range from small handheld devices such as handheld computers / mobile phones 210 to large mainframe systems such as mainframe computers 270. Examples of handheld computers 210 include personal digital assistants (PDAs), personal entertainment devices such as Moving Picture Experts Group Layer-3 Audio (MP3) players, portable televisions, and compact disc players. Other examples of information handling systems include pens or tablets, computers 220, laptops or notebooks, computers 230, workstations 240, personal computer systems 250, and servers 260. Other types of information handling systems not individually shown in Figure 2 are represented by information handling systems 280. As illustrated, the various information handling systems can be networked using computer networks 200. The types of computer networks that can be used to interconnect various information handling systems include local area networks (LANs), wireless local area networks (WLANs), the Internet, public switched telephone networks (PSTNs), other wireless networks, and any other network topologies that can be used to interconnect information handling systems. Many information handling systems include non-volatile data stores such as hard disks, non-volatile memory, or both. The embodiment of the information handling system shown in Figure 2 includes separate non-volatile data stores (more specifically, server 260 utilizes non-volatile data store 265, mainframe computer 270 utilizes non-volatile data store 275, and information handling system 280 utilizes non-volatile data store 285).The non-volatile data store may be an external component of various information handling systems, or it may be built into one of the information handling systems. Furthermore, the removable non-volatile storage device 145 can be shared between two or more information handling systems using various techniques, such as connecting it to a USB port or other connector of an information handling system.

[0025] As mentioned above, the fundamental aspect of a node / sensor in a nuanced cognitive multi-agent network is its axioms, and nodes in a cognitive multi-agent network influence each other by assigning weights based on the data they receive from neighboring nodes. The challenge in existing systems is that, while they have mechanisms for ranking nodes, they do not take into account the node's axioms regarding data points. Furthermore, while existing systems may evaluate the properties of nodes, they do not consider the nuances based on the cognitive ability of each node to independently possess axioms. In other words, existing systems do not consider how a node's cognitive process differs from that of other nodes with respect to different data points that the node receives.

[0026] To detect malfunctioning sensors / devices, it's necessary to analyze their behavior when detecting the required phenomena. Numerical readings from malfunctioning sensors often don't match those of most sensors in the subnetwork or nearby areas. Such persistent discrepancies in readings can obscure the nuances of which node is malfunctioning.

[0027] In cognitive multi-agent networks, the fundamental aspect where nodes / sensors have nuances is the node / sensor axiom. A node axiom is a property inherent to the node, representing its local world view for every data point it receives. This axiom is close to an established truth for the node and is considered the basis of its decision-making process. For example, in the weather processing node in the scenario described above, the node's axiom could be "the boiler temperature can never go below 0°C." A node takes this belief or axiom at face value either because it is programmed that way or because that is how its experience has unfolded.

[0028] In a cognitive multi-agent network, nodes influence each other by assigning weights to neighboring nodes based on the data they receive from those neighbors. These weights are a measure of importance that a node assigns to the data it receives from its neighbors. The weights assigned to other nodes depend on (a) the node's own axioms and (b) the data that the node itself generates. If a particular node is given a high weight by its neighbors, it is unlikely that the node is malfunctioning because the data points sent by that node are consistent with (a) the axioms of its neighbors and (b) the data points generated by its neighbors.

[0029] Figures 3-8 illustrate an approach feasible for an information handling system that maintains data integrity in a large-scale, autonomous cognitive multi-agent system. This approach evaluates the nuances of nodes in the cognitive multi-agent system based on axioms. Then, it calculates the reliability of nodes in the cognitive multi-agent system based on the axioms and nuances. On the other hand, this approach iteratively updates the axioms of the nodes in the cognitive multi-agent system based on the nuances.

[0030] Figure 3 shows an example of a data integrity system that maintains the integrity of a multi-agent system. The data integrity system 350 provides a mechanism for calculating nuances of node 310 in a cognitive multi-agent system 300 using node axioms for different input data points. In one embodiment, the data integrity system 350 assigns nuances based on node axioms on input data points in applications such as i) reliable weather monitoring and forecasting over large areas, reliable maintenance and monitoring of the temperature of heat-sensitive equipment, reliable navigation and global positioning using satellite data, and reliable large-scale deployment of edge devices.

[0031] In one embodiment, the network of a cognitive multi-agent system includes two distinct types of embedded structures, such as an infrastructure network and an interaction network. In this embodiment, the infrastructure network includes the actual connections between nodes (e.g., a satellite constellation used in a global positioning system) that essentially constitute the infrastructure for further data exchange. The interaction network is an overlay network of interactions between nodes, using the underlying infrastructure network (e.g., weights assigned to each other by the nodes as a result of data exchange).

[0032] The data integrity system 350 employs algorithms that define measures such as axioms, inconsistencies, and nuances on the interaction network underlying the network of the cognitive multi-agent system 300. When the data integrity system 350 evaluates the nuances of node 310 in the cognitive multi-agent system 300 based on the node's axioms, the data integrity system 350 provides a mechanism for calculating the reliability of node 310 in maintaining data integrity based on nuances. Node reliability is an indicator of the independence of its cognitive processes based on the axioms it maintains for various data points. Furthermore, reliability is an indicator of data integrity and is iteratively defined based on the reliability of neighboring nodes and the nuances of the node itself.

[0033] In one embodiment, the data integrity system 350 scores the nodes 310 based on the convergence of reliability. The reliability score determines which node is most reliable in maintaining the data integrity of the network. The more reliable a node is, the less likely it is to corrupt data and the more likely it is to maintain data integrity. On the other hand, the data integrity system 350 iteratively updates the axioms of the node 310 in the cognitive multi-agent system 300 based on its nuances and reliability, thereby allowing the node 310 in the multi-agent system 300 to evolve over time.

[0034] Figure 4 is a flowchart illustrating an example of the procedure taken to evaluate the nuances of a node in a cognitive multi-agent system based on its axioms, calculate its reliability, and iteratively update the node's axioms accordingly.

[0035] The process in Figure 4 begins at 400, and in step 420, the process evaluates the nuance of node 310 in the cognitive multi-agent system 300 based on the node's axioms. In one embodiment, a data integrity system 350 addresses the problem of calculating the node's nuance in the cognitive multi-agent system 300 based on its axioms regarding the various input data points received by the node. Specifically, the data integrity system 350 correctly analyzes the data points received by the node and draws attention to the node's tendency to include them in the node's decision-making process based on the node's axioms. This ability to analyze data points is referred to herein as the node's nuance. As defined herein, nuance is the variance that the node exhibits in its ability to analyze data points. In one embodiment, the variance is in the range [0,1] (see Figure 6 and the corresponding text for details).

[0036] In one embodiment, to evaluate nuances and help characterize the analytical capabilities of a node based on axioms, the data integrity system 350 determines whether a node has a tendency to i) have the same axioms regardless of the data points it receives, ii) a tendency to be forced to mismatch with other nodes regardless of the data points it receives, or iii) a tendency to match or mismatch with other nodes without following a particular pattern across the set of data points.

[0037] In another embodiment, the data integrity system 350 defines discrepancies as differences in the axioms of each node compared to other nodes. Discrepancies are modeled in two ways. The first discrepancy is a discrepancy with an aggregated overall assessment. In the first discrepancy, for a given data point, the data integrity system 350 calculates an aggregated overall assessment of the population and then calculates the difference between the axioms of each node and the overall assessment. The second discrepancy is a discrepancy with individual nodes. In the second discrepancy, for a batch of data points, the data integrity system 350 calculates the overall difference in the node's axioms compared to other nodes. In either of the two discrepancies, the data integrity system 350 defines the nuance of a node as an aggregate (statistical moment) of the distribution of discrepancies.

[0038] In step 440, the process calculates the reliability of a node in the cognitive multi-agent system 300 based on the node's axioms and corresponding nuances. The data integrity system 350 defines the reliability of a node based on the axioms of a particular node and the corresponding nuances of the node regarding data points. (See Figure 7 and the corresponding text for further details).

[0039] In step 460, the process iteratively updates the node axioms in the cognitive multi-agent system based on the corresponding nuances and reliability of the nodes. In one embodiment, the data integrity system 350 iteratively updates the node axioms using one of the three models described herein and simultaneously calculates the nuances and reliability (see Figures 7, 8 and the corresponding text for further details). The data integrity system 350 continues to score the nodes and determine which nodes are most reliable in maintaining data integrity in the cognitive multi-agent system 300. The process in Figure 4 then terminates at 495.

[0040] Figure 5 shows an example of a data integrity system 350 that determines inconsistencies within nodes over time and calculates nuances of the nodes based on these inconsistencies.

[0041] Data point 500 of a neighboring node is input to both node A510 and node B540. Axiom A515 of node A510 applies an interpretation to the input and outputs the number "A". Similarly, axiom B545 of node B540 applies an interpretation to the input and outputs the number "B".

[0042] The discrepancy analyzer 520 of node A510 evaluates the discrepancy between its value A and value B, as well as between values ​​C, D, and E from neighboring nodes C, D, and E, and stores the discrepancy (e.g., A=4, B=5, discrepancy=2) in the data store 525. Over time, the discrepancy analyzer 520 stores multiple discrepancies in the data store 525. The nuance calculation module 530 analyzes the discrepancies in the data store 525 and calculates a nuance A535, which in one embodiment ranges from 0 to 1, where a nuance closer to 0 is a low nuance and a nuance closer to 1 is a high nuance.

[0043] Similarly, the discrepancy analyzer 550 of node B540 evaluates the discrepancies between its value B and value A, as well as between values ​​F, G, and H from neighboring nodes F, G, and H. The discrepancy analyzer 550 stores the discrepancies in the data store 555, and over time, the discrepancy analyzer 550 stores multiple discrepancies in the data store 555. The nuance calculation module 560 analyzes the discrepancies in the data store 555 and calculates a nuance B565, which in one embodiment ranges from 0 to 1, where a nuance closer to 0 is a low nuance and a nuance closer to 1 is a high nuance.

[0044] As discussed herein, the data integrity system 350 uses computed nuances to evaluate nodes and to adjust its axioms as necessary to improve the overall reliability of the cognitive multi-agent system 300.

[0045] Figure 6 shows an example of the relationship between node axioms and nuances. The data integrity system 350 uses a novel approach to scoring nodes 310 in the cognitive multi-agent system 300 by considering the axioms that node 310 has for different data points and the nuances that node 310 exercises. As will be described later, the framework of the data integrity system 350 includes a signed network that captures the relationships between upvotes and downvotes between nodes as factors in calculating the corresponding nuances of the nodes.

[0046] Figure 600 shows node 610, axiom 620, and nuance 630. As previously explained, axiom 620 is an inherent property of node 610 and represents node 610's local world view of any data point it receives. Axiom 620 is an established truth for node 610 and is considered the basis of node 610's decision-making process. Each node 310 has an autonomous cognitive process based on the axiom corresponding to the node on a different data point. When node 610 receives a data point, node 610 applies axiom 620 and generates an cognition of the data point to be included in its decision-making process. Thus, for each data point, node 610 generates a new cognition using axiom 620. As discussed herein, axiom 620 can be modified by the influence of neighboring nodes (see Figure 8 and corresponding text for further details).

[0047] Nuance 630 is the ability of node 610 to analyze different data points it receives based on axiom 620. The output of node 610's cognitive process changes depending on the data points the node receives. Thus, the data integrity system 350 represents a cognitive system in which independent nodes 310 in a multi-agent system 300 interact with each other to exchange data points and use their nuances as a scoring mechanism for node 310.

[0048] The data integrity system 350 presents a model for scoring the nuances of a node 310 in a cognitive multi-agent system 300 based on the axioms of the node. The data integrity system 350 instantaneously takes the axioms of the node on the data points as input, presents a basic mathematical model for calculating discrepancies, and explains the calculation of nuances using discrepancies. In one embodiment, the data integrity system 350 is a directed unsigned graph G u to G u =(V,E,w u ) is defined as follows. Here, V is a set of nodes in a graph. E⊆V×V is an ordered set of pairs of nodes that represent edges in a graph. w u :E→[0,1] indicates the edge weight.

[0049] Next, the data integrity system 350 defines G=(V,E,w) as a directed signed graph where the edge weights are given by w:E→[-1,1]. Function w u The range of is a subset of the range of the function w, and therefore, a signed network (graph) is a generalization of an unsigned network.

[0050] In one embodiment, the data integrity system 350 uses the following as a basis for determining the nuance 630: (i) An interaction network modeled as a directed graph represented by an adjacency matrix. The edges may be signed. For example, the weight of an edge represents a vote in favor (+1) or against (-1) placed by the source node of the edge at the destination node. (ii) A set of data points exchanged through the network. The data integrity system 350 assumes that these are injected into the network in batches. (iii) An axiom that each node has for all data points exchanged through the network. The axiom is represented as a matrix, and each column represents a vector of the node's axioms over all data points. X n Taking (i) as the axiom of node i for data point n, the data integrity system 350 is X n assuming (i) ∈ [0,1], and the value quantifies the axiom of the node held on data point n.

[0051] For data point n, the data integrity system 350 calculates the disagreement of the node with other nodes. First, the data integrity system 350 explains the calculation of the aggregated overall evaluation and the node's disagreement. For all data points, the data integrity system 350 defines the overall evaluation O n as the average of the axioms of all nodes that received data point n. TIFF0007847506000001.tif17122

[0052] Here |S n | is the total number of nodes that received the data point. Next, the data integrity system 350 calculates the difference between the node's axiom and the overall evaluation for all data points and calls it the disagreement dn(i). TIFF0007847506000002.tif8124

[0053] In one embodiment, the data integrity system 350 calculates node inconsistencies with individual nodes instead of an overall assessment. For each node, the data integrity system 350 defines the inconsistency as a pairwise cosine dissimilarity or Euclidean distance between its own axiom vector and the axiom vectors of other nodes. Alternatively, any other arbitrary measure of distance (normalized to the interval [0,1]) can be used instead of cosine or Euclidean distance. TIFF0007847506000003.tif9131

[0054] Here, X(i) is the axiom vector of node i for a batch of data points. TIFF0007847506000004.tif14162

[0055] Here, the sum is calculated over the batch of data points.

[0056] Next, the data integrity system 350 calculates a mismatch vector. In one embodiment, the data integrity system 350 determines the degree of mismatch using equation (2) above, where the degree of mismatch vector for a node is defined as the set of degrees of mismatch for that node for all data points received by that node. TIFF0007847506000005.tif9118

[0057] In another embodiment, the data integrity system 350 uses equations 3 and 4 above to determine a mismatch, where the node mismatch vector is defined as a set of mismatches of that node with other nodes in the network. TIFF0007847506000006.tif10125

[0058] For each node (node ​​610), the data integrity system 350 calculates its corresponding nuance, where the nuance 630 is defined as the variance of its mismatch vector. TIFF0007847506000007.tif8119

[0059] Nuance 630 represents node 610's ability to analyze different data points it receives based on axiom 620.

[0060] The output of the cognitive process of a nuance node changes depending on the data points it receives. In one embodiment, nuance has a boundary between [0,1], where values ​​closer to 0 indicate low nuance for the node, and values ​​closer to 1 indicate high nuance for the node. Node 610 exhibits conforming behavior if it consistently matches the general population for all data points. Node 610 exhibits contrarian behavior if it consistently opposes the general population for all data points. In both cases, the mismatch vector of node 610 has a small variance, so these behaviors are not classified as nuance.

[0061] On the other hand, if node 610 does not exhibit a particular pattern in matching or mismatching with other nodes, node 610 is considered to have nuance (nuance 630), and as a result, the variance of node 610's mismatch vector is high. Nodes in the network are scored based on their corresponding nuance 630. In one embodiment, to compute a generalization of mismatches for a signed network, the data integrity system 350 defines w' as edge weights scaled to the range [0,1]. TIFF0007847506000008.tif18139

[0062] Here, t is the edge weight in graph G. The data integrity system 350 defines discrepancy as follows: TIFF0007847506000009.tif9159d'(i,j): Weighted mismatch. w'(i,j): The linearly transformed vote value t belongs to [-1,1] -> t belongs to [0,1]. d(i,j): The mismatch value defined in (ii).

[0063] Figure 7 shows an example of the relationship between nodes, axioms, nuances, and reliability. Figure 700 shows node 610, axiom 620, nuance 630, and reliability 710. For data point n, the data integrity system 350 calculates the nuance 630 as described above. Using the nuance 630 and edge weights, the data integrity system 350 calculates the reliability 710 for each node, and in one embodiment, the reliability of a node is defined as the eigenvector centrality of the adjacency matrix of the graph modified by the nuance values ​​(unsigned). TIFF0007847506000010.tif9116

[0064] Here, A is the edge weight w ij This is a modified adjacency matrix of the graph, defined by and scaled by the nuances of the destination node j. ηj is defined by one of the following equations: TIFF0007847506000011.tif32149

[0065] The data integrity system 350 iteratively calculates the eigenvectors, and after each iteration (720), the data integrity system 350 scores the nodes based on reliability. The data integrity system 350 reaches the convergence condition when the Spearman correlation coefficient of the node reliability values ​​exceeds 1-ε for any small ε over several iterations. After the iterations have converged, the data integrity system 350 scores the nodes based on reliability.

[0066] As described above, the convergence criterion is defined by the Spearman correlation coefficient of the reliability values. After all iterative calculations have converged, the data integrity system 350 scores the network nodes based on reliability to determine which nodes have the greatest impact on maintaining data integrity.

[0067] Figure 8 is an illustrative diagram showing two embodiments of iteratively updating the node axioms based on nuance and reliability in a cognitive multi-agent system. As described above, for data point n, the data integrity system 350 calculates the nuance and reliability of each node using nuance and edge weights. After each iteration, the data integrity system 350 scores the nodes based on reliability and reaches the convergence condition when the Spearman correlation coefficient of the node reliability values ​​exceeds 1-ε for any small ε over several iterations.

[0068] After all iterative calculations have converged, the data integrity system 350 scores the nodes 310 based on the corresponding reliability and nuance. The scoring has two purposes: i) to determine the convergence of the iterative calculations of reliability and nuance, and ii) to determine which node is most reliable in maintaining data integrity.

[0069] Model 800 updates axiom 620 using the following formula, which uses the weighted sum of the influences of the axiom values ​​in the previous iteration and their neighbors. TIFF0007847506000012.tif22116X k Axioms for the k-th iteration. i: The node where the axioms are updated. N in : Neighboring nodes in this node. δ: The weight assigned to the influence of neighboring node in at the kth iteration. η: The nuance of a node.

[0070] The magnitude of the impact depends on the axioms and nuances of neighboring nodes, and the nuances of the axioms themselves. Model 800 has two iterative processes. The first iterative process brings about the convergence of nuances and axioms (810). Once the axioms converge, the second iterative process calculates reliability (720). The convergence of nuances always precedes the convergence of reliability. The data completeness system 350 defines this behavior as a binomial convergence of axioms and nuances followed by a unary convergence of reliability.

[0071] Model 850 is similar to Model 800, except that the influential terms now depend on the reliability of neighboring nodes instead of nuance 630. TIFF0007847506000013.tif21117X k Axioms for the k-th iteration. i: The node where the axioms are updated. N in : Neighboring nodes in this node. δ: The weight assigned to the influence of neighboring node in at the kth iteration. R: Node reliability.

[0072] Model 850 has two nested iterative processes. In the inner iterative process, reliability is calculated from nuances (720). After the convergence of reliability, the axioms are updated (860), which is the outer iterative process. Meanwhile, the newly calculated nuances 630 return to the iterative process for calculating reliability. The data completeness system 350 calls this operation a trinomial convergence of axioms, nuances, and reliability.

[0073] In one embodiment, the data integrity system 350 uses a third model, which is a time-based model, to iteratively update the axioms of node 310. The third model tracks how the axioms change over a period of time. TIFF0007847506000014.tif9123

[0074] Here, X' is the first derivative of the node's axiom vector with respect to time. To find the consistency of the node, the data completeness system 350 tracks how often the node's axioms change. The data completeness system 350 finds the second derivative of the axioms with respect to time and calculates the root mean square error (RMSE) with respect to the zero vector (representing the origin), which is given by the following equation. TIFF0007847506000015.tif8114

[0075] In this embodiment, the data integrity system 350 defines node consistency C as follows: TIFF0007847506000016.tif8114

[0076] Here, the smaller the RMSE, the higher the node consistency. The data integrity system 350 uses node consistency as an additional parameter for the axiom update process of the data integrity system 350, using the following equation. TIFF0007847506000017.tif11135X t : The axiom of time t. k: Number of repetitions. i: The node where the axioms are updated. N in : Neighboring nodes in this node. η j : The nuance of node j. P t : Dependency of node j. δ: The weight assigned to the influence of neighboring node in at the kth iteration.

[0077] The data integrity system 350 can maintain the data integrity of node 310 in the cognitive multi-agent system 300 using Model 800, Model 850, a time-based model, or other models or combinations thereof.

[0078] While specific embodiments of the Disclosure have been shown and described, it will be apparent to those skilled in the art that changes and modifications can be made based on the teachings herein without departing from the Disclosure and its broader aspects. Accordingly, the attached claims encompass all such changes and modifications within their scope as they fall within the true spirit and scope of the Disclosure. Furthermore, it will be understood that the Disclosure is defined solely by the attached claims. Where a particular number of elements of a claim is intended, such intention is explicitly stated in the claim; where there is no such statement, it will be understood to those skilled in the art that no such limitation exists. As a non-restrictive example, for the sake of understanding, the following attached claims include the use of the introductory phrases “at least one” and “one or more” to introduce elements of a claim. However, the use of such phrases should not be interpreted as meaning that any particular claim containing the introduced element is limited to a disclosure containing only one such element, even if the same claim also contains the introductory phrase "one or more" or "at least one" and an indefinite article such as "a" or "an," and the same applies to the use of definite articles in the claims.

Claims

1. The first node applies the first axiom to a set of data points in order to generate the first set of outputs, The second node applies a second axiom to the set of data points in order to generate a second set of outputs, wherein the first and second nodes are part of a computer network comprising multiple nodes. Calculating a first nuance based on the set of discrepancies between the first set of outputs and the second set of outputs, Based on the first nuance described above, the reliability, which is an indicator of data integrity of the first node in the computer network, Computer implementation methods, including those mentioned above.

2. Capture the first set of outputs and the second set of outputs over a certain period of time, The calculation of the set of discrepancies between the first set of outputs and the second set of outputs over the aforementioned period, wherein the first nuance is the variance of the set of discrepancies between the first set of outputs and the second set of outputs over the aforementioned period, The computer implementation method according to claim 1, further comprising:

3. The reliability of the first node is iteratively calculated based on the set of edge weights corresponding to the set of neighboring nodes and the first nuance, until the reliability reaches a convergence condition based on the set of edge weights. The computer implementation method according to claim 1, further comprising:

4. Before iteratively calculating the reliability of the first node, the first axiom is iteratively adjusted based on a weighted sum of a set of iterative nuance calculations in order to converge the first axiom and the first nuance. The computer implementation method according to claim 3, further comprising:

5. In response to iteratively calculating the reliability of the first node, the first axiom is iteratively adjusted to converge the first axiom, the first nuance, and the reliability. The computer implementation method according to claim 3, further comprising:

6. Tracking the set of adjustments to the first axiom over a certain period of time, Based on the set of adjustments, the consistency of the first node is determined, The first axiom mentioned above is to include consistency as a factor in the iterative adjustments, The computer implementation method according to claim 5, further comprising:

7. The computer network is a cognitive multi-agent system including a set of independent nodes, and the method is The set of independent nodes, including the first node and the second node, enables the maintenance of an autonomous cognitive process based on a set of axioms, including the first axiom and the second axiom. The set of independent nodes interacts with each other and allows for the exchange of the set of data points, The set of networks, including the set of independent nodes in the cognitive multi-agent system, enables the capture of sets of vote relationships and sets of vote relationships between the sets of independent nodes, The computer implementation method according to claim 1, further comprising:

8. An information handling system, One or more processors, A memory connected to at least one of the aforementioned processors, The set of computer program instructions stored in the memory and executed by one or more of the processors to perform an operation, the operation is The first node applies the first axiom to a set of data points in order to generate the first set of outputs, The second node applies a second axiom to the set of data points in order to generate a second set of outputs, wherein the first and second nodes are part of a computer network comprising multiple nodes. Calculating a first nuance based on the set of discrepancies between the first set of outputs and the second set of outputs, Based on the first nuance described above, the reliability, which is an indicator of data integrity of the first node in the computer network, An information handling system, including [the following].

9. The aforementioned processor, Capture the first set of outputs and the second set of outputs over a certain period of time, The calculation of the set of discrepancies between the first set of outputs and the second set of outputs over the aforementioned period, wherein the first nuance is the variance of the set of discrepancies between the first set of outputs and the second set of outputs over the aforementioned period, The information handling system according to claim 8, which performs additional operations including the following:

10. The aforementioned processor, The reliability of the first node is iteratively calculated based on the set of edge weights corresponding to the set of neighboring nodes and the first nuance, until the reliability reaches a convergence condition based on the set of edge weights. The information handling system according to claim 8, which performs additional operations including the following:

11. The aforementioned processor, Before iteratively calculating the reliability of the first node, the first axiom is iteratively adjusted based on a weighted sum of a set of iterative nuance calculations in order to converge the first axiom and the first nuance. The information handling system according to claim 10, which performs additional operations including the following:

12. The aforementioned processor, In response to iteratively calculating the reliability of the first node, the first axiom is iteratively adjusted to converge the first axiom, the first nuance, and the reliability. The information handling system according to claim 10, which performs additional operations including the following:

13. The aforementioned processor, Tracking the set of adjustments to the first axiom over a certain period of time, Based on the set of adjustments, the consistency of the first node is determined, The first axiom mentioned above is to include consistency as a factor in the iterative adjustments, The information handling system according to claim 12, which performs additional operations including the following:

14. The computer network is a cognitive multi-agent system including a set of independent nodes, and the processor is The set of independent nodes, including the first node and the second node, enables the maintenance of an autonomous cognitive process based on a set of axioms, including the first axiom and the second axiom. The set of independent nodes interacts with each other and allows for the exchange of the set of data points, The set of networks, including the set of independent nodes in the cognitive multi-agent system, enables the capture of sets of vote relationships and sets of vote relationships between the sets of independent nodes, The information handling system according to claim 8, which performs additional operations including the following:

15. A computer, The first node applies the first axiom to a set of data points in order to generate the first set of outputs, The second node applies a second axiom to the set of data points in order to generate a second set of outputs, wherein the first and second nodes are part of a computer network comprising multiple nodes. Calculating a first nuance based on the set of discrepancies between the first set of outputs and the second set of outputs, Based on the first nuance described above, the reliability, which is an indicator of data integrity of the first node in the computer network, A computer program that executes something.

16. To the aforementioned computer, Capture the first set of outputs and the second set of outputs over a certain period of time, The calculation of the set of discrepancies between the first set of outputs and the second set of outputs over the aforementioned period, wherein the first nuance is the variance of the set of discrepancies between the first set of outputs and the second set of outputs over the aforementioned period, To make it run further, The computer program according to claim 15.

17. To the aforementioned computer, The reliability of the first node is iteratively calculated based on the set of edge weights corresponding to the set of neighboring nodes and the first nuance, until the reliability reaches a convergence condition based on the set of edge weights. To make it run further, The computer program according to claim 15.

18. To the aforementioned computer, Before iteratively calculating the reliability of the first node, the first axiom is iteratively adjusted based on a weighted sum of a set of iterative nuance calculations in order to converge the first axiom and the first nuance. To make it run further, The computer program according to claim 17.

19. To the aforementioned computer, In response to iteratively calculating the reliability of the first node, the first axiom is iteratively adjusted to converge the first axiom, the first nuance, and the reliability. To make it run further, The computer program according to claim 17.

20. To the aforementioned computer, Tracking the set of adjustments to the first axiom over a certain period of time, Based on the set of adjustments, the consistency of the first node is determined, The first axiom mentioned above is to include consistency as a factor in the iterative adjustments, To make it run further, The computer program according to claim 19.

Images