An Archive Record Tracing Method and System Based on Multi-Agent Collaboration

Abstract

The present invention discloses a method and system for tracing archive records based on multi-agent collaboration, including collecting archive data and record data of multiple agents, and preprocessing the archive data and the record data; using a blockchain to encrypt the information of the record data to obtain encrypted data, and classifying and storing the encrypted data according to the identity level for node operations to obtain classified data; performing subject marking on the classified data according to the identity information to obtain marked data, and performing optimized comparison on the marked data and the archive data to obtain a matching degree; constructing a multi-agent collaborative archive record tracing model according to the matching degree, inputting the data to be traced into the multi-agent collaborative archive record tracing model, and outputting a tracing result. This method can not only improve the accuracy of archive record tracing, but also has good interpretability and can be directly applied to the archive record tracing system.

Description

Technical Field

[0001] The present invention relates to the field of record traceability, and particularly to a method and system for file record traceability based on multi-agent collaboration. Background Art

[0002] In today's information-based and digital social environment, various organizations, institutions, and individuals generate a vast amount of file record data in their daily operations and activities. These data not only record historical information but also serve as important bases for key scenarios such as decision-making support, compliance review, and dispute resolution. However, with the rapid increase in data volume and the diversification of data sources, how to efficiently and accurately manage and trace these file records has become a huge challenge. Especially in scenarios involving multi-agent collaboration, such as cross-departmental cooperation among governments, supply chain collaboration among enterprises, and multi-party participation in scientific research projects, file records often span different systems, formats, and even regions, making data integration, query, and verification more complex.

[0003] File records are important carriers for recording history and inheriting civilization, and their authenticity and integrity are crucial. However, traditional file record traceability methods have many problems. File records are scattered and stored in various departments or institutions, lacking an effective sharing mechanism, forming information silos, and making it difficult to conduct cross-departmental and cross-institutional traceability. File records involve sensitive information, and traditional data storage and transmission methods have security risks, making it easy to cause data leakage and tampering. Traditional file record traceability mainly relies on manual operations, with low efficiency and difficulty in meeting the traceability requirements in large-scale and complex scenarios. Traditional file record traceability lacks an effective verification mechanism and is difficult to ensure the credibility of traceability results.

[0004] In modern information-based society, the management and traceability of file record data face more and more challenges. File record data is not only huge in quantity but also involves multiple agents, with wide data sources and complex structures. To ensure the security, integrity, and traceability of file record data, an efficient multi-agent collaborative traceability method is urgently needed. Summary of the Invention

[0005] The purpose of the present invention is to provide a method for file record traceability based on multi-agent collaboration.

[0006] To achieve the above object, the present invention is implemented according to the following technical solution:

[0007] The present invention includes the following steps:

[0008] Collect the file data and record data of multiple agents, and preprocess the file data and the record data;

[0009] Use blockchain to encrypt the recorded data to obtain encrypted data, and classify and store the encrypted data according to the identity level to obtain classified data; including:

[0010] Statistically classify the recorded data and obtain the revised quantity, and sort the recorded data according to the revised quantity;

[0011] Match the sorted recorded data based on the archive database to obtain the similarity. When the similarity is greater than 0.358, generate a private key, and the expression is:

[0012] ,

[0013] where the private key of the matching request w is ,the private key coefficient is K, the prime number is p, and the private key exponent is m;

[0014] Adjust the matching structure, use blockchain technology to set up the information storage architecture, encrypt the recorded data with the public key, and calculate the importance of the recorded data:

[0015] ,

[0016] where the total number of revisions is M, the i-th recorded data is ,the recorded data The probability of revision is ,the number of revisions of the i-th recorded data is ,the importance of the i-th recorded data is ,the weight coefficient of the i-th recorded data is ,and the probability of obtaining the recorded data in the case where the retrieval E occurs is ,and the number of recorded data is ;

[0017] Detect whether the encrypted key information is in the original state, regenerate the key, obtain the decryption key according to the storage location, and restore the matching plaintext information. The expression is:

[0018] ,

[0019] ,

[0020] where the key is ,the recorded data structure information with a similarity greater than 0.358 is ,the parameter for restoring the key decryption method to the initial state is ,the important component index of the ciphertext is Q, and the decryption information parameter is ,

[0021] Insert the subject identity into the keywords of the record data to obtain the inserted data, randomly generate characters of length m, encrypt the inserted data through the generation method of the logical key to obtain the key, and convert the key into a hexadecimal number;

[0022] According to the elliptic curve encryption algorithm process, use the public key to encrypt the inserted data into ciphertext. The expression is:

[0023] ,

[0024] where the ciphertext is , the public key is F, the record data is K, the random number is , and the base point is H;

[0025] Output the encrypted inserted data as encrypted data;

[0026] Perform subject marking on the classified data according to the identity information to obtain marked data, and perform optimized comparison on the marked data and the file data to obtain the matching degree, including:

[0027] Convert the marked data and the file data into word vectors and input them into the optimized comparison model. Use the convolutional neural network to strengthen the adjacent text information of the marked data, and use the long short-term memory neural network to increase the timing information. The expression is:

[0028] ,

[0029] ,

[0030] ,

[0031] ,

[0032] where the forgetting information at the s-th moment is , the memory information at the s-th moment is , the privacy state at the s-th moment is , the cell state at the s-th moment is , the cell state at the (s - 1)-th moment is , the transpose of the forgetting information is , the transpose of the memory information is , the transpose of the cell state is , the privacy state at the (s - 1)-th moment is , the activation function is , the input vector at the s-th moment is ;

[0033] Take the privacy state as the timing information, perform one-by-one comparison of the timing information according to the time interval, and obtain the adjacent relationship between the timing information and the word vectors of the file data. The expression is:

[0034] ,

[0035] ,

[0036] where the constraint coefficient is ,the sampling interval is ,the first timing information is ,the second timing information is ,the third timing information is ,the nth timing information is ,the (n + 1)th timing information is ,the first word vector is ,the second word vector is ,the third word vector is ,the nth word vector is ,the (n + 1)th word vector is ,the adjacent relationship of the timing information is ,the adjacent relationship of the word vectors is ;

[0037] Perform semantic analysis on the timing information and the word vectors of the archival data to obtain a knowledge representation language, and calculate the matching degree between the timing information and the word vectors of the archival data according to the knowledge representation language:

[0038] ,

[0039] where the (a + 1)th timing information is ,the ath timing information is ,the (a + 1)th word vector is ,the ath word vector is ,the word vector of the archival data is k, and the matching degree between the timing information f and the word vector k is ,the timing information and the distance length between the knowledge representation languages of the word vector is ,and the distance between the knowledge representation languages of the timing information and the word vectors is ;

[0040] The long short-term memory neural network includes a forget gate and a cell state;

[0041] Construct a multi-agent collaborative archival record tracing model according to the matching degree, input the data to be traced into the multi-agent collaborative archival record tracing model, and output the tracing result.

[0042] Furthermore, a method for classifying and storing node operations on the encrypted data according to the identity level to obtain classified data includes:

[0043] Obtain the operation text fragment and the operator identity according to the recorded data, segment the operation text fragment to obtain the segmented operation text fragment, use the segmented operation text fragment as the operation node, match the corresponding encrypted data according to the operation node to obtain the node encrypted data, and the support vector machine classifies the node encrypted data based on the identity level by finding the optimal separation hyperplane in the feature space, and stores the classified node encrypted data to obtain the classified data.

[0044] Furthermore, a method for subject marking the classified data according to the identity information, including:

[0045] Mark the classified data according to the text annotation based on the operator identity information in the recorded data; the operator identity information includes name and work unit;

[0046] The annotation includes the operator identity information and the operation time.

[0047] Furthermore, the knowledge representation language includes word vectors with multi-semantic expressions and ambiguous word vectors.

[0048] Furthermore, a method for constructing a multi-subject collaborative archive record traceability model according to the matching degree, including:

[0049] Construct an objective function by weighted summation according to the matching degree and the loss function. The multi-subject collaborative archive record traceability model includes a random forest algorithm, a graph theory algorithm, a disambiguation correction algorithm, and a deep machine learning algorithm;

[0050] The random forest algorithm divides the input data into training data and test data according to a ratio of 6:1;

[0051] The graph theory algorithm constructs an operation source graph according to the data flow edges formed by the archive operation content, operator identity information, and archive operation time of the training data.

[0052] The disambiguation correction algorithm eliminates and corrects the ambiguous words in the operation source graph by analyzing and using the context information of the training data and combining machine learning to obtain an optimized operation source graph, and extracts node information from the optimized operation source graph to obtain node operation data;

[0053] The deep machine learning algorithm learns the archive traceability law of the objective function through a multi-layer non-linear transformation model, and performs archive record traceability on the node operation data according to the traceability law.

[0054] In a second aspect, an archive record traceability system based on multi-subject collaboration includes:

[0055] Data acquisition module: used to acquire the file data and record data of multiple entities, and preprocess the file data and the record data;

[0056] Encryption and classification module: used to encrypt the record data using a blockchain to obtain encrypted data, and classify and store the encrypted data according to the identity level to obtain classified data;

[0057] Label matching module: used to perform entity labeling on the classified data according to the identity information to obtain labeled data, and perform optimized comparison on the labeled data and the file data to obtain the matching degree;

[0058] Construction and output module: used to construct a multi-entity collaborative file record traceability model according to the matching degree, input the data to be traced into the multi-entity collaborative file record traceability model, and output the traceability result.

[0059] In a third aspect, an embodiment of the present application further provides an electronic device, including:

[0060] A processor; and a memory arranged to store computer-executable instructions, the executable instructions, when executed, cause the processor to execute the method steps described in the first aspect.

[0061] In a fourth aspect, an embodiment of the present application further provides a computer-readable storage medium, the computer-readable storage medium stores one or more programs, and when the one or more programs are executed by an electronic device including a plurality of application programs, the electronic device is caused to execute the method steps described in the first aspect.

[0062] The beneficial effects of the present invention are:

[0063] The present invention is a file record traceability method and system based on multi-entity collaboration. Compared with the prior art, the present invention has the following technical effects:

[0064] Through the steps of preprocessing, obtaining a satisfactory deviation, obtaining a fuzzy intention, obtaining a corrected intention, obtaining a target intention, and model construction, the present invention can improve the accuracy of user intention recognition in multi-dimensional human-computer interaction scenarios, thereby improving the precision of user intention recognition in multi-dimensional human-computer interaction scenarios. Optimizing the user intention recognition in multi-dimensional human-computer interaction scenarios can greatly save resources, improve work efficiency, can realize the automatic recognition of user intentions in multi-dimensional human-computer interaction scenarios, and can correct intentions and enhance knowledge in real time for the user intention recognition in multi-dimensional human-computer interaction scenarios. It is of great significance for the user intention recognition in multi-dimensional human-computer interaction scenarios, and can adapt to the user intention recognition in multi-dimensional human-computer interaction scenarios with different standards and the user intention recognition requirements in different multi-dimensional human-computer interaction scenarios, and has a certain universality. Description of the Drawings

[0065] Figure 1 This is the step flowchart of a method for tracing archive records based on multi-agent collaboration according to the present invention;

[0066] Figure 2 This is the structural schematic diagram of an electronic device in the embodiments of this specification. Detailed implementation manners

[0067] The present invention will be further described below through specific embodiments. The illustrative embodiments and descriptions of the present invention are used to explain the present invention, but do not limit the present invention.

[0068] A method and system for tracing archive records based on multi-agent collaboration according to the present invention include the following steps:

[0069] As Figure 1 shown, in this embodiment, the following steps are included:

[0070] Collect the archive data and record data of multiple agents, and preprocess the archive data and the record data;

[0071] In an actual evaluation, Hospital A, Insurance Company B, Government Department C, and Research Institution D are used as research objects to obtain the archive data and record data of Person E;

[0072] The record data of Hospital A is: Record number HA013, medical record update, update the treatment plan of Patient E, add physical therapy, operation time: 2024-06-01 14:32:54, operator identity doctor, revision times 2, importance score 0.851;

[0073] The record data of Insurance Company B is: Record number IB001, claim settlement review, operation content: review the medical expense claim application of Patient E, approve the claim amount of 5000 yuan, operation time: 2024-06-02 09:47:09, operator identity claim settlement specialist, revision times 1, importance score 0.703;

[0074] The record data of Government Department C is: Record number CA001, policy query, query the medical subsidy policy for the area where Patient E is located, operation time 2024-06-03 11:20:00, operator identity: policy analyst, work unit: Government Department C, revision times 0, importance score 0.498;

[0075] The record data of Research Institution D is: Record number DA001, research data collection, operation content: collect the treatment data of Patient E for research analysis, operation time 2024-06-04 10:00:00, operator identity researcher, revision times 1, importance score 0.632;

[0076] Use blockchain to encrypt the recorded data to obtain encrypted data, and classify and store the encrypted data according to the identity level to obtain classified data; including:

[0077] Statistically classify the recorded data and obtain the revision quantity, and sort the recorded data according to the revision quantity;

[0078] Match the sorted recorded data based on the archive database to obtain the similarity. When the similarity is greater than 0.358, generate a private key, and the expression is:

[0079] ,

[0080] where the private key of the matching request w is , the private key coefficient is K, the prime number is p, and the private key exponent is m;

[0081] Adjust the matching structure, use blockchain technology to set up the information storage architecture, encrypt the recorded data with the public key, and calculate the importance of the recorded data:

[0082] ,

[0083] where the total number of revisions is M, the i-th recorded data is , the recorded data The probability of revision is , the number of revisions of the i-th recorded data is , the importance of the i-th recorded data is , the weight coefficient of the i-th recorded data is , and the probability of obtaining the recorded data in the case where the retrieval E occurs is , and the number of recorded data is ;

[0084] Detect whether the encrypted key information is in the original state, regenerate the key, obtain the decryption key according to the storage location, and restore the matching plaintext information. The expression is:

[0085] ,

[0086] ,

[0087] where the key is , the recorded data structure information with a similarity greater than 0.358 is , the parameter for restoring the key decryption method to the initial state is , the important component index of the ciphertext is Q, and the decrypted information parameter is ,

[0088] Insert the subject identity into the keywords of the record data to obtain the inserted data, randomly generate characters of length m, encrypt the inserted data through the generation method of the logical key to obtain the key, and convert the key into a hexadecimal number;

[0089] According to the elliptic curve encryption algorithm process, use the public key to encrypt the inserted data into ciphertext, and the expression is:

[0090] ,

[0091] where the ciphertext is , the public key is F, the record data is K, the random number is , and the base point is H;

[0092] Output the encrypted inserted data as encrypted data;

[0093] In the actual evaluation, the encrypted data is: Hospital A: 3b2453435658e7969ae58fade69a91e697895f57534051596f5d4849424552465f476d51504d5fe58d9ee795bc5e57586f455d4c42; Insurance Company B: 3a2753425458e79199e8b4bfe5af84e6a1815f57534051596f5d48494145534b5f40685155405fe791a3e8b4b7e4b9a1e590bd586e475557445550; Government Department C: 302453425458e695a0e7acbde69e80e8ae9b5f57534051596f5d4849404552435f466f5155495fe6959ae7acb5e589b4e69fb5e5b9bc735b49495d515a4a; Research Institution D: 372453425458e7a18be7a89de69495e68c97e69585e99aa34f4055466b46554f5e5557525444655b554343554fe7a1a6e7a893e590ac735a49495d535040;

[0094] Perform subject marking on the classified data according to the identity information to obtain marked data, and perform optimized comparison on the marked data and the file data to obtain a matching degree; including:

[0095] Convert the labeled data and archive data into word vectors and input them into the optimized comparison model. Use the convolutional neural network to strengthen the adjacent text information of the labeled data, and use the long short-term memory neural network to increase the temporal information. The expression is:

[0096] ,

[0097] ,

[0098] ,

[0099] ,

[0100] where the forgetting information at the s-th moment is , the memory information at the s-th moment is , the privacy state at the s-th moment is , the cell state at the s-th moment is , the cell state at the (s - 1)-th moment is , the transpose of the forgetting information is , the transpose of the memory information is , the transpose of the cell state is , the privacy state at the (s - 1)-th moment is , the activation function is , the input vector at the s-th moment is ;

[0101] Take the privacy state as the temporal information, and perform pairwise comparison of the temporal information according to the time interval to obtain the adjacent relationship between the temporal information and the word vectors of the archive data. The expression is:

[0102] ,

[0103] ,

[0104] where the constraint coefficient is , the sampling interval is , the first temporal information is , the second temporal information is , the third temporal information is , the n-th temporal information is , the (n + 1)-th temporal information is , the first word vector is , the second word vector is , the third word vector is , the n-th word vector is , the (n + 1)-th word vector is , the adjacent relationship of the temporal information is , the adjacent relationship of the word vectors is ;

[0105] Semantically analyze the word vectors of the timing information and archival data to obtain a knowledge representation language, and calculate the matching degree of the word vectors of the timing information and archival data according to the knowledge representation language:

[0106] ,

[0107] where the (a + 1)-th timing information is , the a-th timing information is , the (a + 1)-th word vector is , the a-th word vector is , the word vector of the archival data is k, and the matching degree between the timing information f and the word vector k is , the timing information and the distance length between the knowledge representation languages of the word vector is , and the distance between the knowledge representation languages of the timing information and the word vector is ;

[0108] The long short-term memory neural network includes a forgetting gate and a cell state;

[0109] In the actual evaluation, the matching degrees of Hospital A, Insurance Company B, Government Department C, and Research Institution D are 0.37, 0.863, 0.507, and 0.439 respectively;

[0110] Construct a multi-agent collaborative archival record tracing model according to the matching degree, input the data to be traced into the multi-agent collaborative archival record tracing model, and output the tracing result.

[0111] In this embodiment, the method for classifying and storing the encrypted data into classified data according to the identity level includes:

[0112] Obtain the operation text segment and the operator identity according to the recorded data, segment the operation text segment to obtain the segmented operation text segment, use the segmented operation text segment as the operation node, match the corresponding encrypted data according to the operation node to obtain the node encrypted data, and the support vector machine classifies the node encrypted data based on the identity level by finding the optimal separation hyperplane in the feature space, and stores the classified node encrypted data to obtain the classified data.

[0113] In this embodiment, the method for subject-marking the classified data according to the identity information to obtain the marked data includes:

[0114] Mark the classified data according to the operator identity information of the recorded data based on the text annotation according to the operator identity information; the operator identity information includes the name and the work unit.

[0115] The annotation includes operator identity information and operation time.

[0116] In this embodiment, the knowledge representation language includes word vectors with multiple semantics and ambiguous word vectors.

[0117] In this embodiment, the method for constructing a multi-agent collaborative archive record traceability model according to the matching degree includes:

[0118] Construct a target function by weighted summation according to the matching degree and the loss function. The multi-agent collaborative archive record traceability model includes a random forest algorithm, a graph theory algorithm, a disambiguation correction algorithm, and a deep machine learning algorithm;

[0119] The random forest algorithm divides the input data into training data and test data according to a ratio of 6:1;

[0120] The graph theory algorithm constructs an operation source graph according to the data flow edges formed by the archive operation content, operator identity information, and archive operation time of the training data.

[0121] The disambiguation correction algorithm eliminates and corrects the ambiguous words in the operation source graph by analyzing and using the context information of the training data and combining machine learning to obtain an optimized operation source graph, and extracts node information from the optimized operation source graph to obtain node operation data;

[0122] The deep machine learning algorithm learns the archive traceability rules of the target function through a multi-layer non-linear transformation model, and performs archive record traceability on the node operation data according to the traceability rules.

[0123] Second, an archive record traceability system based on multi-agent collaboration includes:

[0124] Data acquisition module: used to acquire archive data and record data of multiple agents, and preprocess the archive data and the record data;

[0125] Encryption and classification module: used to encrypt the record data using blockchain to obtain encrypted data, and classify and store the encrypted data according to the identity level to obtain classified data;

[0126] Label matching module: used to perform subject labeling on the classified data according to the identity information to obtain labeled data, and perform optimized comparison on the labeled data and the archive data to obtain a matching degree;

[0127] Construction and output module: used to construct a multi-agent collaborative archive record traceability model according to the matching degree, input the data to be traced into the multi-agent collaborative archive record traceability model, and output the traceability result.

[0128] Figure 2 is a schematic structural diagram of an electronic device according to an embodiment of the present application. Please refer to Figure 2 , at the hardware level, the electronic device includes a processor, and optionally also includes an internal bus, a network interface, and a memory. Among them, the memory may include a memory, such as a high-speed random access memory (Random-Access Memory, RAM), and may also include a non-volatile memory, such as at least one disk memory, etc. Of course, the electronic device may also include other hardware required for other services.

[0129] The processor, network interface, and memory can be interconnected through an internal bus, and the internal bus can be an ISA (Industry Standard Architecture) bus, a PCI (Peripheral Component Interconnect) bus, or an EISA (Extended Industry Standard Architecture) bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc. For the sake of simplicity of representation, Figure 2 only a bidirectional arrow is used in

[0130] the figure, but it does not mean that there is only one bus or one type of bus.

[0131] The memory is used to store programs. Specifically, the program may include program code, and the program code includes computer operation instructions. The memory may include a memory and a non-volatile memory, and provide instructions and data to the processor.

[0132] As described above in the present application Figure 1The illustrated embodiment discloses a method for archival record tracing based on multi-agent collaboration that can be applied to or implemented by a processor. The processor may be an integrated circuit chip with signal processing capabilities. During implementation, each step of the method described above can be performed by hardware integrated logic circuits within the processor or by software instructions. The processor described above can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. The methods, steps, and logic block diagrams disclosed in the embodiments of this application can be implemented or executed. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the method disclosed in the embodiments of this application can be directly executed by a hardware decoding processor or by a combination of hardware and software modules within the decoding processor. The software module can be located in a storage medium mature in the art such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory, or an electrically erasable programmable memory, a register, etc. The storage medium is located in the memory, and the processor reads the information in the memory and completes the steps of the above method in combination with its hardware.

[0133] The electronic device may also perform Figure 1 A multi-agent collaborative archival record tracing method is implemented in Figure 1 The functions of the illustrated embodiment will not be described in detail in the embodiments of the present application.

[0134] An embodiment of the present application also proposes a computer-readable storage medium, which stores one or more programs, and the one or more programs include instructions. When the instructions are executed by an electronic device including multiple applications, any one of the aforementioned archival record tracing methods based on multi-agent collaboration is executed.

[0135] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0136] The present application is described with reference to the flowcharts and / or block diagrams of methods, apparatuses (systems), and computer program products according to the embodiments of the present application. It should be understood that each flow and / or block in the flowchart and / or block diagram, as well as the combination of flows and / or blocks in the flowchart and / or block diagram, can be realized by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, such that the instructions executed by the processor of the computer or other programmable data processing devices generate means for implementing the functions specified in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0137] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, such that the instructions stored in the computer-readable memory generate a manufactured article including instruction means that implement the functions specified in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0138] These computer program instructions can also be loaded onto a computer or other programmable data processing device, such that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process, and thus the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0139] In a typical configuration, a computing device includes one or more processors (CPUs), an input / output interface, a network interface, and memory.

[0140] The memory may include non-permanent memory in the form of computer-readable media, random access memory (RAM), and / or non-volatile memory such as read-only memory (ROM) or flash RAM. The memory is an example of computer-readable media.

[0141] A computer-readable medium includes permanent and non-permanent, removable and non-removable media that can implement information storage by any method or technology. The information can be computer-readable instructions, data structures, program modules, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassette tapes, magnetic tape magnetic disk storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information that can be accessed by a computing device. As defined herein, a computer-readable medium does not include transitory computer-readable media, such as modulated data signals and carrier waves.

[0142] It should also be noted that the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, such that a process, method, article or device comprising a series of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article or device. Without further limitation, an element defined by the statement "comprising an..." does not exclude the presence of additional identical elements in the process, method, article or device comprising the element.

[0143] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system or a computer program product. Therefore, the present application can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk memory, CD-ROM, optical memory, etc.) containing computer-usable program code.

[0144] The above are only the preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present invention shall be included within the protection scope of the present invention.

Claims

1. A method for tracing archive records based on multi-agent collaboration, characterized in that, Including the following steps: Collect the file data and record data of multiple subjects, and preprocess the file data and the record data; Use blockchain to encrypt the information of the record data to obtain encrypted data, and classify and store the encrypted data according to the identity level to obtain classified data; including: Statistically classify the record data and obtain the revised quantity, and sort the record data according to the revised quantity; Based on the file database, match the sorted record data to obtain the similarity. When the similarity is greater than 0.358, generate a private key, and the expression is: ， The private key for matching request w is , the private key coefficient is K, the prime number is Z, and the private key exponent is m; Adjust the matching structure, use blockchain technology to set up the information storage architecture, encrypt the record data with the public key, and calculate the importance of the record data: ， where the total number of revisions is M, and the i-th recorded data is , the recorded data the probability of revision is , the number of revisions of the i-th recorded data is , the importance of the i-th recorded data is , the weight coefficient of the i-th recorded data is , the probability of obtaining the recorded data in the case where the retrieval E occurs is , the number of recorded data is ; Detect whether the encrypted key information is in the original state, regenerate the key, obtain the decryption key according to the storage location, and restore the matching plaintext information, and the expression is: ， ， where the secret key is , the record data structure information with a similarity greater than 0.358 is , the parameter for restoring the secret key decryption method to the initial state is , the important component index of the ciphertext is Q, and the decryption information parameter is , Insert the subject identity into the keywords of the record data to obtain inserted data, randomly generate characters with a length of m, encrypt the inserted data through the generation method of the logical key to obtain the key, and convert the key into a hexadecimal number; According to the elliptic curve encryption algorithm process, encrypt the inserted data into ciphertext with the public key, and the expression is: ， where the ciphertext is , the public key is F, the recorded data is K, and the random number is , and the base point is H; Output the encrypted inserted data as encrypted data; Mark the classified data according to the identity information to obtain marked data, and perform an optimized comparison between the marked data and the file data to obtain the matching degree; including: Convert the marked data and the file data into word vectors and input them into the optimized comparison model, use the convolutional neural network to strengthen the adjacent text information of the marked data, and use the long short-term memory neural network to increase the timing information, and the expression is: ， ， ， ， where the forgotten information at the s-th moment is , the memory information at the s-th moment is , the privacy state at the s-th moment is , the cell state at the s-th moment is , the cell state at the (s - 1)-th moment is , the transpose of the forgotten information is , the transpose of the memory information is , the transpose of the cell state is , the privacy state at the (s - 1)-th moment is , the activation function is , the input vector at the s-th moment is ; Take the privacy status as the timing information, perform a one-by-one timing information comparison according to the time interval, and obtain the adjacent relationship between the timing information and the word vectors of the file data, and the expression is: ， ， where the constraint coefficient is , the sampling interval is , the first timing information is , the second timing information is , the third timing information is , the nth timing information is , the (n + 1)th timing information is , the first word vector is , the second word vector is , the third word vector is , the nth word vector is , the (n + 1)th word vector is , the adjacent relationship of the timing information is , the adjacent relationship of the word vectors is ; Perform semantic analysis on the word vectors of the timing information and the file data to obtain the knowledge representation language, and calculate the matching degree between the word vectors of the timing information and the file data according to the knowledge representation language: ， where the (a + 1)-th timing information is , the a-th timing information is , the (a + 1)-th word vector is , the a-th word vector is , the word vector of the file data is k, and the matching degree between the timing information f and the word vector k is , the timing information 's knowledge representation language and the word vector 's knowledge representation language has a distance length of , and the distance between the knowledge representation languages of the timing information and the word vector is ; The long short-term memory neural network includes a forgetting gate and a cell state; Construct a multi-subject collaborative file record traceability model according to the matching degree, input the data to be traced into the multi-subject collaborative file record traceability model, and output the traceability result.

2. The archival record traceability method based on multi-agent collaboration according to claim 1, wherein The method for classifying and storing the encrypted data according to the identity level to obtain classified data includes: Obtain the operation text segment and the operator identity according to the record data, segment the operation text segment to obtain the segmented operation text segment, use the segmented operation text segment as the operation node, match the corresponding encrypted data according to the operation node to obtain the node encrypted data, and the support vector machine finds the optimal separation hyperplane in the feature space, classifies the node encrypted data based on the identity level, and stores the classified node encrypted data to obtain the classified data.

3. The method for tracing archive records based on multi-agent collaboration according to claim 1, wherein The method for marking the classified data according to the identity information to obtain marked data includes: Based on the operator identity information of the record data, mark the classified data according to the text annotation based on the operator identity information; the operator identity information includes the name and the work unit; The annotation includes operator identity information and operation time.

4. The archival record traceability method based on multi-agent collaboration according to claim 1, wherein The knowledge representation language includes word vectors with multiple semantics and ambiguous word vectors.

5. The method for tracing archive records based on multi-agent collaboration according to claim 1, characterized in that The method for constructing a multi-agent collaborative archive record traceability model according to the matching degree includes: Constructing an objective function by weighted summation according to the matching degree and the loss function. The multi-agent collaborative archive record traceability model includes a random forest algorithm, a graph theory algorithm, a disambiguation correction algorithm, and a deep machine learning algorithm; The random forest algorithm divides the input data into training data and test data according to a ratio of 6:1; The graph theory algorithm constructs an operation source graph according to the data flow edges formed by the archive operation content, operator identity information, and archive operation time of the training data; The disambiguation correction algorithm eliminates and corrects the ambiguous words in the operation source graph by analyzing and using the context information of the training data and combining machine learning to obtain an optimized operation source graph, and extracts node information from the optimized operation source graph to obtain node operation data; The deep machine learning algorithm learns the archive traceability rules of the objective function through a multi-layer non-linear transformation model, and performs archive record traceability on the node operation data according to the traceability rules.

6. A file record traceability system based on multi-agent collaboration for implementing the method according to any one of claims 1-5, characterized in that, Including: A data acquisition module: used to acquire archive data and record data of multiple agents, and preprocess the archive data and the record data; An encryption classification module: used to encrypt the record data using blockchain to obtain encrypted data, and classify and store the encrypted data according to the identity level to obtain classified data; A marker matching module: used to perform subject marking on the classified data according to the identity information to obtain marked data, and perform optimized comparison between the marked data and the archive data to obtain a matching degree; A construction output module: used to construct a multi-agent collaborative archive record traceability model according to the matching degree, input the data to be traced into the multi-agent collaborative archive record traceability model, and output the traceability result.

7. An electronic device, comprising: A processor; And A memory arranged to store computer-executable instructions, the executable instructions when executed cause the processor to execute the method according to any one of claims 1 to 5.

8. A computer-readable storage medium, the computer-readable storage medium stores one or more programs, when the one or more programs are executed by an electronic device including a plurality of application programs, the electronic device is caused to execute the method according to any one of claims 1 to 5.

Images