Data security control method and system based on large language model

Abstract

Multiple embodiments of this specification relate to the field of information technology, and specifically to a data security management method and system based on a large language model. The method comprises the following steps: reading target data, extracting the attributes and content keywords of the target data using a pre-connected large language model; obtaining the data type of the target data according to the attributes and content keywords; obtaining the association between the target data and the historical data according to the attributes and content keywords and the attributes and content keywords of the historical data; obtaining the security level of the target data according to the attributes, content keywords, data types, associations and the security level of the corresponding historical data.

Description

Technical Field

[0001] Multiple embodiments of this specification relate to the field of information technology, and specifically to a data security management method and system based on a large language model. Background Art

[0002] With the development of information technology, enterprises have generated a large amount of sensitive information in their daily operations, including customer information, financial data, trade secrets, etc. The security of this information is crucial to enterprises. Therefore, enterprises need to formulate and implement strict data protection policies, such as data encryption, access control, backup and recovery strategies, etc., to prevent data leakage, loss or illegal tampering. Enterprise data materials are rich in variety and have different security levels and confidentiality requirements. For example, the company's equity structure, business license, etc., these documents are public materials that clarify the nature of the company, the scope of business, and the rights and obligations between shareholders. As the company grows and develops, more professional documents will be generated, such as financial reports, tax returns, audit reports, etc., and in daily operations, enterprises need to prepare various contract texts, such as procurement contracts, sales contracts, lease contracts, labor contracts, etc. In addition, a large number of management documents will be formed within the enterprise, such as employee handbooks, performance appraisal standards, training materials, etc., to guide employee behavior and improve team work efficiency. In order to adapt to market changes and technological advances, enterprises also need to regularly update relevant documents for products or services, such as product manuals, technical white papers, user guides, etc. The increase in enterprise data materials has brought new challenges to data security management. Therefore, it is necessary to study data security management and control technologies that have both high security and high efficiency. Summary of the invention

[0003] Multiple embodiments of this specification describe a data security management method and system based on a large language model.

[0004] In a first aspect, an embodiment of the present specification provides a data security management method based on a large language model, characterized in that it includes the steps of:

[0005] Read the target data and extract the attributes and content keywords of the target data using the pre-connected large language model;

[0006] According to the attributes and content keywords, obtaining the data type of the target data;

[0007] Obtaining an association relationship between the target data and the historical data according to the attributes and content keywords and the attributes and content keywords of the historical data;

[0008] The security level of the target data is obtained according to the attributes, content keywords, data types, association relationships and the security level of corresponding historical data.

[0009] In a second aspect, the embodiments of this specification provide a data security management and control system based on a large language model, characterized in that:

[0010] A reading module reads the target data and uses a pre-connected large language model to extract the attributes and content keywords of the target data;

[0011] A type module, which obtains the data type of the target data according to the attribute and content keyword;

[0012] An association module, which obtains an association relationship between the target data and the historical data according to the attributes and content keywords and the attributes and content keywords of the historical data;

[0013] The level module obtains the security level of the target data according to the attributes, content keywords, data types, association relationships and the security level of corresponding historical data.

[0014] In a third aspect, an embodiment of this specification provides an electronic device, including a processor and a memory;

[0015] The processor is connected to the memory;

[0016] The memory is used to store executable program code;

[0017] The processor runs a program corresponding to the executable program code by reading the executable program code stored in the memory, so as to execute the method described in any one of the above aspects.

[0018] In a fourth aspect, an embodiment of the present specification provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program implements the method described in any of the above aspects when executed by a processor.

[0019] In a fifth aspect, an embodiment of this specification provides a computer program product, including a computer program, which implements the method described in any of the above aspects when executed by a processor.

[0020] The beneficial effects brought by the technical solutions provided by some embodiments of this specification include at least:

[0021] In multiple embodiments of this specification, the data security management method provided is combined with content keywords and historical data when determining the security level of target data. By combining content keywords, the security level of target data can be more accurately divided, and by combining associated historical data, the security level of target data can be more reasonably determined based on the association between target data and historical data. The data security management method provided in this specification realizes dynamic management of security levels by setting a confidentiality period, which helps to reduce the cost of data security management and more reasonably determine the security level of target data.

[0022] Other features and advantages of the various embodiments of the present specification will be further disclosed in the following detailed description and drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] In order to more clearly illustrate the technical solutions in the embodiments of this specification, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this specification. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0024] Figure 1 A schematic diagram of an application scenario of the data security management and control method provided in the embodiments of this specification.

[0025] Figure 2 A schematic diagram of the application framework of the data security management and control method provided in the embodiments of this specification.

[0026] Figure 3 A schematic diagram of the operating interface of an example of a data security management and control method provided in an embodiment of this specification.

[0027] Figure 4 A flowchart of a data security management method provided in an embodiment of this specification.

[0028] Figure 5 A schematic diagram of the data type acquisition method provided in the embodiments of this specification.

[0029] Figure 6 This is a flow chart of a method for obtaining a degree of association provided in an embodiment of this specification.

[0030] Figure 7 A schematic diagram of a method for obtaining a security level according to an embodiment of the present disclosure.

[0031] Figure 8 A schematic diagram of a data security management and control system provided in accordance with an embodiment of this specification.

[0032] Fig. 9 A schematic diagram of an electronic device provided in an embodiment of this specification. DETAILED DESCRIPTION

[0033] The following is an explanation and description of the technical solutions of the embodiments of this specification in conjunction with the drawings of the embodiments of this specification, but the following embodiments are only preferred embodiments of this specification, not all. Based on the embodiments in the implementation mode, other embodiments obtained by those skilled in the art without creative work are all within the scope of protection of this specification.

[0034] The terms "first", "second", "third", etc. in the description and claims of this specification and the above-mentioned drawings are used to distinguish different objects, rather than to describe a specific order. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not limited to the listed steps or units, but optionally includes steps or units that are not listed, or optionally includes other steps or units inherent to these processes, methods, products or devices.

[0035] In the following description, terms such as "inside", "outside", "up", "down", "left", "right", etc. that indicate directions or positional relationships are only used to facilitate the description of the embodiments and simplify the description, and do not indicate or imply that the device or element referred to must have a specific direction, be constructed and operated in a specific direction, and therefore should not be understood as a limitation of this specification.

[0036] The data involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection of relevant data complies with the relevant laws, regulations and standards of relevant countries and regions.

[0037] Before describing the technical solution in this specification, an introduction is given to the application scenarios and related technologies of the technical solution.

[0038] There are many types of corporate documents, covering documents from company establishment to operation and management. According to different needs and uses, these documents can be classified from multiple perspectives, such as legal documents, financial documents, personnel documents, marketing and sales documents, production and technical documents, administrative documents and R&D documents. There are three categories based on importance and confidentiality level, namely public documents, internal documents and confidential documents. Public documents include annual reports, white papers, press releases, etc. Internal documents are documents that are only available to internal company personnel, such as employee handbooks, internal training materials, etc. Confidential documents are data involving the company's core competitiveness or sensitive information, which need to be kept strictly confidential, such as strategic planning, financial data, etc.

[0039] However, it is not a simple task to classify confidentiality levels. For example, if the same financial data involves the R&D and procurement of new projects of the enterprise, it needs to be strictly kept confidential and is a confidential document. However, if it is only the financial data of the annual public welfare donation of the enterprise, it not only does not need to be kept confidential, but also needs to be actively disclosed. That is, the same financial data will be divided into different confidentiality levels due to different contents involved. Similarly, internal training materials are internal documents, which are generally accessible to internal personnel of the enterprise. For training on conference room reservation and use, training on the use of corporate buses, and training on travel expense reimbursement, internal training materials can be used. However, if it involves the use of training materials for new products of the enterprise, especially new products in the testing phase, because there are usually many defects and problems that the enterprise is unwilling to disclose, it is necessary to use confidential documents to keep it confidential. Similar situations also exist in meeting minutes. It can be seen that the confidentiality level needs to be determined not only according to the type of data, but also according to the content of the data. However, the current method of manually viewing enterprise data and determining the confidentiality level is not only inefficient but also easy to determine the wrong level. For this reason, it is necessary to study technologies that can improve the confidentiality level classification of enterprise data.

[0040] Please see attached Figure 1 The technical solution of this embodiment is applied in an enterprise and runs on a server 20. The server 20 receives data generated by the enterprise, namely the target data 10, and marks the confidentiality level classification for the target data 10, which is then managed by the enterprise according to the defined confidentiality level classification. Since the higher the confidentiality level, the higher the management cost, and the confidentiality requirements of some target data 10 are not static, the target data 10 that originally needed to be kept confidential may need to continue to be kept confidential, or may no longer need to be kept confidential as time changes. Therefore, the technical solution provided in this specification can dynamically classify the confidentiality level classification of the target data 10 according to time. Please refer to the attached Figure 2 , which is the scenario for the application of the technical solution recorded in this specification. The staff or the enterprise's management system submits the target data 10 to the server 20 by calling the interface. The staff can use a terminal device 11 such as a PC, a smart phone, a tablet computer, etc. to submit the target data 10 to the server 20. And obtain feedback from the server 20 on the confidentiality level classification of the target data 10. The terminal device 11 can run a client, or at least run an operation interface 12. Please refer to the attached Figure 3 The client or operation interface 12 can be run on the terminal device 11 in the form of a browser, or in the form of an independent application (APP), etc. The specific display form of the client is not limited here.

[0041] The server 20 involved in this application can be an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery networks (CDN), and big data and artificial intelligence platforms. The terminal device 11 can be a smart phone, a tablet computer, a laptop, a PDA, a personal computer, a smart speaker, a smart TV, a smart watch, a car device, a wearable device, etc., but is not limited to this. The terminal device 11 and the server 20 can be directly or indirectly connected via wired or wireless communication, and this application does not limit this. The number of servers 20 and terminal devices 11 is also not limited. The solution provided in this application can be completed independently by the terminal device 11, or by the server 20, or by the terminal device 11 and the server 20 in cooperation, and this application does not make specific limitations on this.

[0042] Since this application involves some professional terms, these professional terms will be introduced below.

[0043] Security Level

[0044] The security level in this manual refers to the classification of confidentiality levels. There are three types of confidentiality levels for enterprises, namely public documents, internal documents, and confidential documents. The information contained in public documents is open to the public and does not require any special protection measures. Anyone can access this information, which may include company profiles, product introductions, annual reports, etc. Internal documents refer to documents that are limited to internal employees of the company. This type of information is not suitable for public disclosure because it may contain sensitive content such as the company's operating details, strategic planning, customer lists, etc. For such files, there are usually certain access control measures, such as logging into the company's internal network or using specific permissions to view them. Confidential files are the highest level of confidential information and are usually only open to a very limited number of people. Confidential files may include highly sensitive information such as the company's core business secrets, financial data, research and development results, etc. When handling such files, the company will take strict security measures, such as encrypted storage, physical isolation, and signing of confidentiality agreements.

[0045] Target data

[0046] The target data in this specification refers to data that needs to be classified into security levels, and the data forms include documents, recordings, videos, etc. The main purpose of classifying the target data 10 into security levels is to ensure the security and privacy of the data and prevent unauthorized access, leakage or abuse. The security management measures for the target data 10 can be carried out in the manner disclosed in the prior art. Exemplary, access control, encryption technology, physical security, auditing and monitoring can be used. When using access control technology, different access rights are set for data of different security levels to ensure that only authorized personnel can access data of the corresponding level. Confidential data is encrypted to ensure that it cannot be easily read even if the data is stolen. For particularly important data, physical isolation measures can be taken, such as storing it in a dedicated server 20 or storage device. Regularly check data access logs, monitor abnormal access behavior, and promptly detect and deal with potential security threats.

[0047] Large Language Model

[0048] Large language models are deep learning models with large parameter counts that are designed to understand and generate natural language. These models learn patterns and structures in language by training on large amounts of text data, enabling them to perform a variety of natural language tasks such as machine translation, text summarization, sentiment analysis, question-answering systems, etc. Large language models have millions or even billions of parameters, which enables them to capture very complex patterns in language.

[0049] Natural language tasks

[0050] Natural language tasks refer to natural language processing. Natural language processing (NLP) is an interdisciplinary subject in the fields of computer science, artificial intelligence, and linguistics, which aims to enable computers to understand, interpret, and generate human language. Natural language tasks (NLP tasks) refer to a series of language-related tasks completed through algorithms and technologies.

[0051] This manual first provides a data security management method based on a large language model. Figure 4 , including the steps of:

[0052] Step S101) Read the target data 10, and use the pre-connected large language model to extract the attributes and content keywords of the target data 10.

[0053] By using the method of constructing natural language tasks, the attributes and content keywords of the target data 10 can be completed with the help of a large language model. The attributes include one or more of the time when the target data 10 was created, the data format, the data creator, and the data size. Among them, the attributes should at least include the time when the target data 10 was created. It is worth noting that the attributes of the target data 10 itself are usually recorded in the metadata of the file where the target data 10 is located, but it does not rule out that some attributes are not recorded in the metadata of some files, and the corresponding attributes are recorded in the content of the target data 10, so they can be extracted by the large language model.

[0054] Step S102) According to the attributes and content keywords, the data type of the target data 10 is obtained. The data type is different from the file type. The data type is meeting minutes, training materials, purchase and sales contracts, shareholder agreements, memoranda, archived materials, user manuals, personnel tables, financial statements, etc. The file types of these data types are all documents. The specific types of documents can also be text, word, pdf, etc.

[0055] For details, please refer to the attached Figure 5 , including: step S1021) reading a preset type table, the type table records the data type, a number of reference keywords and type description. Please refer to Table 1, which is the type table used in this embodiment. When implementing, the type table can be set according to the specific data conditions of the enterprise. The reference keywords and type descriptions can be set manually, or automatically generated using the technology disclosed in the field, or directly copied and used from the information disclosed in the field. Or generated using a large language model.

[0056] Table 1 Type table

[0057]

[0058] Step S1022) Obtain the matching degree of the attribute and content keywords with the reference keywords and type description according to the large language model. The attribute and content keywords record some key information, which is recorded by text (including attribute values ​​and content keywords). The matching degree can be obtained by comparing the text.

[0059] Step S1023 ) Select the data type corresponding to the highest matching degree as the data type of the target data 10 .

[0060] Step S103) Obtain the association relationship between the target data 10 and the historical data according to the attributes and content keywords and the attributes and content keywords of the historical data.

[0061] The historical data is associated with an attribute set, and the attribute set records the attributes that the historical data can be associated with. Exemplarily, if the historical data is a sales contract, the attributes that can be associated with the sales contract include: meeting minutes, contracts, agreements, memoranda, user manuals, and financial statements.

[0062] For details, please refer to the attached Figure 5 , the method for obtaining the association relationship between the target data 10 and the historical data includes: when the attribute of the target data 10 is in the attribute set, calculating the association between the content keywords of the target data 10 and the content keywords of the historical data; when the association is higher than a preset threshold, determining that the target data 10 has an association relationship with the corresponding historical data.

[0063] Please see the attached Figure 6 The method for calculating the relevance between the content keywords of the target data 10 and the content keywords of the historical data includes:

[0064] Step S1031) traverses each content keyword of the target data 10. The content keywords are extracted by a large language model.

[0065] Step S1032) Calculate the relevance between the content keyword of the target data 10 currently traversed and each content keyword of the historical data, and obtain the maximum and average values ​​of the relevance. The relevance between each content keyword and the content keyword of the historical data is also obtained through the large language model.

[0066] For example, the historical data is a purchase and sales contract for a laser cutting machine. The keywords of the content of the purchase and sales contract include: YSK-203 laser cutting machine, Company A, Company B, national quality standards for laser cutting machines, Wang bought and Liu sold, actual cutting test, cutting accuracy, cutting speed, maximum cutting thickness, cutting end face quality, etc.

[0067] A subsequent communication meeting between the two parties produced a meeting minutes, which was used as target data 10. The keywords of the meeting minutes were: cutting speed, cutting accuracy, Wang Maimai, and Liu Maimai. The correlation between the cutting speed and each keyword in {YSK-203 laser cutting machine, Company A, Company B, national quality standard for laser cutting machines, Wang Maimai, Liu Maimai, actual cutting test, cutting accuracy, cutting speed, maximum cutting thickness, and cutting end face quality} was calculated respectively. For example, the correlation between the cutting speed and the YSK-203 laser cutting machine is 0.9, the correlation between the cutting speed and Company A is 0.1, the correlation between the cutting speed and Wang Maimai is 0.1, the correlation between the cutting speed and the actual cutting test is 0.6, and the correlation between the cutting speed and the cutting speed is 1.0, etc. After obtaining all the correlations, the maximum and average values ​​of the correlations are obtained.

[0068] Step S1033) According to the maximum value and the average value, obtain the correlation between the content keyword of the target data 10 currently traversed and the content keyword of the historical data, which is recorded as a single correlation. Calculate the weighted average of the maximum value and the average value, where the weight of the maximum value should be relatively larger.

[0069] Step S1034) Obtain the maximum value and average value of all single correlations.

[0070] Step S1035) Obtain the relevance between the content keyword of the target data 10 and the content keyword of the historical data according to the maximum value and the average value of the single relevance.

[0071] Step S104) Obtain the security level of the target data 10 according to the attributes, content keywords, data types, association relationships and security levels of corresponding historical data.

[0072] The attribute records the time when the target data 10 is generated. Figure 7 According to the attributes, content keywords, data types, association relationships and security levels of corresponding historical data, the method for obtaining the security level of the target data 10 includes:

[0073] Step S1041) Obtaining the initial security level of the target data 10 according to the association relationship and the security level of the corresponding historical data.

[0074] For example, when the historical data corresponding to a meeting minutes is a communication and feedback meeting on the execution of specific design indicators in a new product development contract, since the security level of the new product development contract is relatively high and confidential, the meeting minutes will obtain an initial security level based on the security level of the new product development contract, that is, the initial security level of the meeting minutes is confidential. When the new product development is completed and launched on the market. There is no need for the company to continue to manage the meeting minutes as confidential, because the cost of managing data as confidential is high. Therefore, by setting a confidentiality period, dynamically managing the security level of the target data 10 can help reduce the cost of confidentiality management. When a meeting minutes is a communication and discussion meeting about the destination of a department's team-building activity, since the security level of the team-building activity is to issue a press release to showcase the corporate culture, the initial security level of the meeting minutes is also public.

[0075] Step S1042) Obtain a confidentiality period based on the content keywords and data type. For a communication and feedback meeting on new product development, if the content keywords involved in the meeting minutes are confidential, the confidentiality period is set to be relatively long or long-term. On the contrary, if the meeting minutes do not involve relevant content keywords that need to be kept confidential, the confidentiality period is set to a preset duration, for example, a confidentiality period of 2 years.

[0076] Specifically, according to the content keywords and data types, the method of obtaining the confidentiality period includes:

[0077] Setting the confidentiality period to a preset initial duration;

[0078] Reading a pre-configured confidentiality table, which records the types of data that need to be kept confidential for a long time;

[0079] When the data type of the target data 10 matches the confidentiality table, updating the confidentiality period to long-term;

[0080] Reading a pre-configured secret word list, wherein the secret word list records screening words that need to be kept confidential for a long time;

[0081] When the content keyword of the target data 10 matches the screening word, the confidentiality period is updated to long term.

[0082] Exemplarily, the confidentiality table records data types such as financial statements and shareholder agreements as data types that need to be kept confidential for a long time. Therefore, when the data type of the target data 10 is financial statements and shareholder agreements, the confidentiality period is set to long-term. Exemplarily, the confidentiality table records profit composition, cost structure, customer information, channels, and department organization. When the content keywords extracted from the target data 10 match the screening words recorded in the confidentiality table, the confidentiality period of the target data 10 is set to long-term.

[0083] Step S1043) When the target data 10 is within the confidentiality period, the security level of the target data 10 is set to the initial security level, and when the target data 10 is outside the confidentiality period, the security level of the target data 10 is lowered, that is, the confidential target data 10 is lowered to internal, and the internal target data 10 is lowered to public.

[0084] On the other hand, this specification provides a data security management system based on a large language model. Figure 8 ,include:

[0085] The reading module 100 reads the target data 10 and uses the pre-connected large language model to extract the attributes and content keywords of the target data 10;

[0086] A type module 200, which obtains the data type of the target data 10 according to the attribute and content keyword;

[0087] An association module 300 obtains an association relationship between the target data 10 and the historical data according to the attributes and content keywords and the attributes and content keywords of the historical data;

[0088] The level module 400 obtains the security level of the target data 10 according to the attributes, content keywords, data types, association relationships and security levels of corresponding historical data.

[0089] See also Fig. 9 A schematic diagram of the structure of an electronic device provided in an embodiment of this specification is shown.

[0090] like Fig. 9 As shown, the electronic device 1100 may include: at least one processor 1101, at least one network interface 1104, a user interface 1103, a memory 1105 and at least one communication bus 1102. Among them, the communication bus 1102 can be used to realize the connection and communication of the above-mentioned components. Among them, the user interface 1103 may include a button, and the optional user interface may also include a standard wired interface and a wireless interface. Among them, the network interface 1104 may include but is not limited to a Bluetooth module, an NFC module, a Wi-Fi module, etc. Among them, the processor 1101 may include one or more processing cores. The processor 1101 uses various interfaces and lines to connect various parts in the entire electronic device 1100, and executes various functions and processes data of the routing device 1100 by running or executing instructions, programs, code sets or instruction sets stored in the memory 1105, and calling data stored in the memory 1105. Optionally, the processor 1101 can be implemented in at least one hardware form of DSP, FPGA, and PLA. The processor 1101 may integrate one or a combination of CPU, GPU and modem, etc. Among them, the CPU mainly processes the operating system, user interface and application programs, etc.; the GPU is responsible for rendering and drawing the content to be displayed on the display screen; the modem is used to process wireless communication.

[0091] It is understandable that the above-mentioned modem may not be integrated into the processor 1101, but may be implemented by a separate chip.

[0092] Among them, the memory 1105 may include RAM or ROM. Optionally, the memory 1105 includes a non-transitory computer-readable medium. The memory 1105 can be used to store instructions, programs, codes, code sets or instruction sets. The memory 1105 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for at least one function (such as a touch function, a sound playback function, an image playback function, etc.), instructions for implementing the above-mentioned various method embodiments, etc.; the data storage area may store data involved in the above-mentioned various method embodiments, etc. The memory 1105 may also be at least one storage device located away from the aforementioned processor 1101. The memory 1105 as a computer storage medium may include an operating system, a network communication module, a user interface module and an application. The processor 1101 may be used to call the application stored in the memory 1105 and execute the methods in the above-mentioned multiple embodiments.

[0093] The embodiments of this specification also provide a computer-readable storage medium, which stores instructions, and when the instructions are executed on a computer or a processor, the computer or the processor executes the multiple steps in the above embodiments. If the components of the above electronic device are implemented in the form of software functional units and sold or used as independent products, they can be stored in the computer-readable storage medium.

[0094] The embodiments of this specification also provide a computer program product, including a computer program, which implements multiple steps in the above embodiments when executed by a processor.

[0095] In the absence of conflict, the technical features in this embodiment and implementation scheme can be combined arbitrarily.

[0096] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented by software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes a plurality of computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function described in the embodiment of this specification is generated in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions may be stored in a computer-readable storage medium or transmitted through the computer-readable storage medium. The computer instructions may be transmitted from a website site, a computer, a server 20 or a data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (Digital Subscriber Line, DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) mode to another website site, computer, server 20 or data center. The computer-readable storage medium may be any available medium that a computer can access or a data storage device such as a server 20, a data center, etc. that includes multiple available media integrated. The available medium may be a magnetic medium (eg, a floppy disk, a hard disk, a magnetic tape), an optical medium (eg, a digital versatile disc (DVD)), or a semiconductor medium (eg, a solid state disk (SSD)), etc.

[0097] When implemented by hardware or firmware, the aforementioned method flow is programmed into the hardware circuit to obtain the corresponding hardware circuit structure and realize the corresponding function. For example, a programmable logic device (PLD) (such as a field programmable gate array (FPGA)) is such an integrated circuit, and its logic function is determined by the user programming the device. The designer programs by himself to "integrate" a digital system on a PLD, without asking a chip manufacturer to design and make a dedicated integrated circuit chip. Moreover, nowadays, instead of manually making integrated circuit chips, this programming is mostly implemented by "logic compiler" software, which is similar to the software compiler used when writing program development, and the original code before compilation must also be written in a specific programming language, which is called hardware description language (HDL), and HDL is not just one, but many. Those skilled in the art should also be aware that it is only necessary to program the method flow slightly in the above-mentioned hardware description languages ​​and program it into the integrated circuit to easily obtain the hardware circuit that implements the logic method flow.

[0098] The embodiments described above are merely preferred embodiments of this specification and are not intended to limit the scope of this specification. Without departing from the design spirit of this specification, various modifications and improvements made to the technical solutions of this specification by ordinary technicians in this field should fall within the scope of protection determined by the claims of this specification.

Claims

1. A data security management and control method based on a large language model, characterized in that: Includes steps: Read the target data and extract the attributes and content keywords of the target data using the pre-connected large language model; According to the attributes and content keywords, obtaining the data type of the target data; Obtaining the association relationship between the target data and the historical data according to the attributes and content keywords and the attributes and content keywords of the historical data; Obtaining the security level of the target data according to the attributes, content keywords, data types, association relationships and security levels of corresponding historical data; The historical data is associated with an attribute set, and the attribute set records the attributes that the historical data can be associated with. The method of obtaining the association relationship between the target data and the historical data according to the attributes and content keywords and the attributes and content keywords of the historical data includes: When the attribute of the target data is in the attribute set, calculating the relevance between the content keyword of the target data and the content keyword of the historical data; When the correlation degree is higher than a preset threshold, it is determined that the target data has a correlation relationship with the corresponding historical data; The method for calculating the relevance between the content keywords of the target data and the content keywords of the historical data includes: Traversing each content keyword of the target data; Calculating the relevance between the content keyword of the target data currently traversed and each content keyword of the historical data, and obtaining the maximum value and average value of the relevance; According to the maximum value and the average value, the correlation between the content keyword of the target data currently traversed and the content keyword of the historical data is obtained, which is recorded as a single correlation; Get the maximum and average values ​​of all single correlations; Obtaining the relevance between the content keyword of the target data and the content keyword of the historical data according to the maximum value and the average value of the single relevance; The attribute records the time when the target data is generated. The method of obtaining the security level of the target data according to the attributes, content keywords, data types, association relationships and security levels of corresponding historical data includes: Obtaining an initial security level of the target data according to the association relationship and the security level of the corresponding historical data; Obtain confidentiality period based on the content keywords and data types; When the target data is within the confidentiality period, the security level of the target data is set to the initial security level; when the target data is outside the confidentiality period, the security level of the target data is lowered.

2. The data security management method based on a large language model according to claim 1 is characterized in that: The method of obtaining the data type of the target data according to the attribute and the content keyword includes: Reading a preset type table, wherein the type table records a data type, a number of reference keywords and a type description; Obtaining a matching degree between the attribute and content keywords and the reference keywords and type description according to the large language model; The data type corresponding to the highest matching degree is selected as the data type of the target data.

3. The data security management method based on a large language model according to claim 1 is characterized in that: According to the content keywords and data types, the method of obtaining the confidentiality period includes: Setting the confidentiality period to a preset initial duration; Reading a pre-configured confidentiality table, which records the types of data that need to be kept confidential for a long time; When the data type of the target data matches the confidentiality period table, updating the confidentiality period to long-term; Reading a pre-configured secret word list, wherein the secret word list records screening words that need to be kept confidential for a long time; When the content keyword of the target data matches the screening word, the confidentiality period is updated to long-term.

4. Data security management and control system based on large language model, characterized by: include: A reading module reads the target data and uses a pre-connected large language model to extract attributes and content keywords of the target data; A type module, which obtains the data type of the target data according to the attribute and content keyword; An association module, which obtains an association relationship between the target data and the historical data according to the attributes and content keywords and the attributes and content keywords of the historical data; A level module, which obtains the security level of the target data according to the attributes, content keywords, data types, association relationships and security levels of corresponding historical data; The historical data is associated with an attribute set, and the attribute set records the attributes that the historical data can be associated with. The method of obtaining the association relationship between the target data and the historical data according to the attributes and content keywords and the attributes and content keywords of the historical data includes: When the attribute of the target data is in the attribute set, calculating the relevance between the content keyword of the target data and the content keyword of the historical data; When the correlation degree is higher than a preset threshold, it is determined that the target data has a correlation relationship with the corresponding historical data; The method for calculating the relevance between the content keywords of the target data and the content keywords of the historical data includes: Traversing each content keyword of the target data; Calculating the relevance between the content keyword of the target data currently traversed and each content keyword of the historical data, and obtaining the maximum value and average value of the relevance; According to the maximum value and the average value, the correlation between the content keyword of the target data currently traversed and the content keyword of the historical data is obtained, which is recorded as a single correlation; Get the maximum and average values ​​of all single correlations; Obtaining the relevance between the content keyword of the target data and the content keyword of the historical data according to the maximum value and the average value of the single relevance; The attribute records the time when the target data is generated. The method of obtaining the security level of the target data according to the attributes, content keywords, data types, association relationships and security levels of corresponding historical data includes: Obtaining an initial security level of the target data according to the association relationship and the security level of the corresponding historical data; Obtain confidentiality period based on the content keywords and data types; When the target data is within the confidentiality period, the security level of the target data is set to the initial security level; when the target data is outside the confidentiality period, the security level of the target data is lowered.

5. An electronic device, characterized in that including a processor and a memory; The processor is connected to the memory; The memory is used to store executable program code; The processor runs a program corresponding to the executable program code by reading the executable program code stored in the memory, so as to execute the method according to any one of claims 1 to 3.

6. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 3 is implemented.

7. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the method according to any one of claims 1 to 3 is implemented.

Images