Security demand recommendation method and device, electronic equipment, medium and program product

Abstract

The invention provides a security demand recommendation method which can be applied to the technical field of big data and the technical field of artificial intelligence. The method comprises the following steps: acquiring demand information input by a user, wherein the demand information comprises application demand information and grade demand information; based on the application demand information, generating a preliminary recommendation result by utilizing a generative artificial intelligence content algorithm in combination with a pre-constructed security demand knowledge graph library; based on the application demand information, the grade demand information and the preliminary recommendation result, performing multi-level matching by using a security demand grading model to generate an intermediate recommendation result; and displaying the intermediate recommendation result through a user interface, adjusting the intermediate recommendation result in response to an interactive operation of a user, and generating a security demand recommendation result.

Description

Technical Field

[0001] The present disclosure relates to the fields of big data technology and artificial intelligence, and more specifically to a security requirement recommendation method, device, equipment, medium and program product. Background Art

[0002] With the rapid development of modern information technology, the security requirements of information systems and applications have become increasingly important. Traditional security requirements analysis usually relies on manual experience, which is time-consuming and labor-intensive, and requires high professional knowledge of analysts, making it difficult to meet the needs of modern information technology scenarios for high efficiency and high accuracy.

[0003] At present, the existing technology proposes recommendations for security demand analysis based on knowledge graphs. By establishing a knowledge graph library containing security knowledge, laws and regulations, and application security scenarios, the search and recommendation of security requirements are realized by using reasoning operations. However, although this static rule design has a certain effect in common application scenarios, it cannot be adjusted and optimized according to the personalized needs input by users and the dynamic changes of the actual environment. It may lack adaptability in complex scenarios and it is difficult to meet users' needs for high precision and high relevance. Summary of the invention

[0004] In view of the above problems, the present disclosure provides a security requirement recommendation method, apparatus, device, medium and program product.

[0005] According to the first aspect of the present disclosure, a security requirement recommendation method is provided, including: obtaining requirement information input by a user, the requirement information including application requirement information and level requirement information; based on the application requirement information, generating a preliminary recommendation result using a generative artificial intelligence content algorithm in combination with a pre-built security requirement knowledge graph library; based on the application requirement information, the level requirement information and the preliminary recommendation result, performing multi-level matching using a security requirement grading model to generate an intermediate recommendation result; and displaying the intermediate recommendation result through a user interface, adjusting the intermediate recommendation result in response to the user's interactive operation, and generating a security requirement recommendation result.

[0006] According to an embodiment of the present disclosure, the preliminary recommendation results are generated based on the application requirement information using a generative artificial intelligence content algorithm combined with a pre-built security requirement knowledge graph library, specifically including: using a generative artificial intelligence algorithm to perform semantic analysis and contextual understanding of the application requirement information to obtain user requirement semantic features; based on the security requirement knowledge graph library, retrieving security requirement nodes related to the application requirement information to generate a node feature vector; and generating the preliminary recommendation results based on interactive modeling of the user requirement semantic features and the node feature vectors.

[0007] According to an embodiment of the present disclosure, the use of a generative artificial intelligence algorithm to perform semantic analysis and contextual understanding on the application requirement information to obtain semantic features of user requirements specifically includes: performing data preprocessing on the application requirement information; generating text embeddings for the preprocessed application requirement information to obtain a semantic feature vector; analyzing the semantic feature vector through a self-attention mechanism to obtain global semantic information; and aggregating the semantic feature vector and the global semantic information to obtain the semantic features of the user requirements.

[0008] According to an embodiment of the present disclosure, the generating of the preliminary recommendation result through interactive modeling of the user demand semantic feature and the node feature vector specifically includes: calculating a first similarity between the user demand semantic feature and the node feature vector; weighting the first similarity based on the context information of the application demand information to obtain a second similarity; and sorting the security demand nodes based on the second similarity, and obtaining the preliminary recommendation result based on the sorting result.

[0009] According to an embodiment of the present disclosure, the preliminary recommendation result includes multiple recommended items, and the multi-level matching is performed based on the application requirement information, the level requirement information and the preliminary recommendation result using a security requirement grading model to generate an intermediate recommendation result, specifically including: based on the level requirement information, defining a recommendation target level in the security requirement grading model, and using the security requirement grading model to divide the recommended items in the preliminary recommendation result into multiple level sets; filtering the recommended items layer by layer based on the multiple level sets to generate a candidate item set that meets the recommendation target level; and calculating the degree of fitness of each item in the candidate item set with the application requirement information, adjusting the priority order of the items in the candidate item set based on the fitness, and generating an intermediate recommendation result.

[0010] According to an embodiment of the present disclosure, the degree of fit between each entry in the candidate entry set and the application requirement information is calculated, specifically including: extracting a security requirement characteristic vector for each entry in the candidate entry set, generating a target requirement vector based on the application requirement information; and calculating the degree of fit between the security requirement characteristic vector and the target requirement vector through vector similarity.

[0011] According to an embodiment of the present disclosure, the intermediate recommendation result is displayed through a user interface, and the intermediate recommendation result is adjusted in response to the user's interactive operation, specifically including: displaying the recommended items of the intermediate recommendation result through a user interface, wherein the order of the recommended items of the intermediate recommendation result follows the order of priority of the items in the candidate item set; and updating the user interface in real time in response to the user adjusting the order of the recommended items of the intermediate recommendation result through a drag operation.

[0012] According to an embodiment of the present disclosure, the method also includes: obtaining user feedback information on the security requirement recommendation results based on a preset time interval; and updating the security requirement knowledge graph and / or adjusting the recommendation strategy of the generative artificial intelligence content algorithm in combination with the feedback information.

[0013] According to an embodiment of the present disclosure, the updating of the security requirement knowledge graph specifically includes: extracting the modification frequency and modification pattern of the user's security requirement recommendation results based on the feedback information; and dynamically adjusting the node association weights in the knowledge graph based on the statistical results of the modification frequency and modification pattern.

[0014] According to an embodiment of the present disclosure, the method also includes: constructing an expert knowledge base containing security expert knowledge and experience, wherein the expert knowledge base is updated and maintained based on a preset time interval; and training and optimizing the generative artificial intelligence content algorithm based on expert rules defined in the expert knowledge base.

[0015] The second aspect of the present disclosure provides a security requirement recommendation device, including: an input acquisition module, used to: obtain requirement information input by a user, the requirement information including application requirement information and level requirement information; a preliminary recommendation result generation module, used to: generate preliminary recommendation results based on the application requirement information, using a generative artificial intelligence content algorithm combined with a pre-built security requirement knowledge graph library; an intermediate recommendation result generation module, used to: based on the application requirement information, the level requirement information and the preliminary recommendation results, using a security requirement grading model for multi-level matching to generate intermediate recommendation results; and a recommendation result generation module, used to: display the intermediate recommendation results through a user interface, adjust the intermediate recommendation results in response to user interaction operations, and generate security requirement recommendation results.

[0016] According to an embodiment of the present disclosure, the security requirement analysis result generation module may include a preliminary recommendation module, and the preliminary recommendation module may include a semantic analysis unit, a node retrieval unit and an interaction modeling unit.

[0017] According to an embodiment of the present disclosure, the semantic analysis unit can be used to perform semantic analysis and context understanding on the application requirement information using a generative artificial intelligence algorithm to obtain semantic features of user requirements.

[0018] According to an embodiment of the present disclosure, the node retrieval unit may be used to retrieve security requirement nodes related to the application requirement information based on the security requirement knowledge graph library, and generate a node feature vector.

[0019] According to an embodiment of the present disclosure, the interaction modeling unit may be used to generate the preliminary recommendation result based on interaction modeling of user demand semantic features and node feature vectors.

[0020] According to an embodiment of the present disclosure, the preliminary recommendation result generation module can also be used to perform data preprocessing on the application requirement information; generate text embedding for the preprocessed application requirement information to obtain a semantic feature vector; analyze the semantic feature vector through a self-attention mechanism to obtain global semantic information; and aggregate the semantic feature vector and the global semantic information to obtain the user requirement semantic feature.

[0021] According to an embodiment of the present disclosure, the preliminary recommendation result generation module can also be used to calculate a first similarity between the user demand semantic feature and the node feature vector; based on the context information of the application demand information, adjust the weight of the first similarity to obtain a second similarity; and sort the security demand nodes based on the second similarity, and obtain the preliminary recommendation result based on the sorting result.

[0022] According to an embodiment of the present disclosure, the preliminary recommendation result generation module can also be used to obtain user feedback information on the security requirement recommendation results based on a preset time interval; and based on the feedback information, update the security requirement knowledge graph and adjust the recommendation strategy of the generative artificial intelligence content algorithm.

[0023] According to an embodiment of the present disclosure, the preliminary recommendation result generation module can also be used to extract the user's modification frequency and modification pattern of the security requirement recommendation results based on the feedback information; and dynamically adjust the node association weights in the knowledge graph based on the statistical results of the modification frequency and modification pattern.

[0024] According to an embodiment of the present disclosure, the preliminary recommendation result generation module can also be used to build an expert knowledge base containing security expert knowledge and experience, and the expert knowledge base is updated and maintained based on preset time intervals; and based on the expert rules defined in the expert knowledge base, the artificial intelligence content algorithm is trained and optimized.

[0025] According to an embodiment of the present disclosure, the intermediate recommendation result generation module may include an intermediate recommendation result generation module, and the intermediate recommendation result generation module may include a level definition unit, a filtering unit, and a calculation unit.

[0026] According to an embodiment of the present disclosure, the level definition unit can be used to define a recommended target level in the security requirement grading model based on the level requirement information, and use the security requirement grading model to divide the recommended items in the preliminary recommendation result into multiple hierarchical sets.

[0027] According to an embodiment of the present disclosure, the filtering unit may be used to filter the recommended items layer by layer based on multiple levels to generate a set of candidate items that meet the target level.

[0028] According to an embodiment of the present disclosure, the calculation unit may be used to calculate the degree of fit between each entry in the candidate entry set and the application requirement information, adjust the priority order of the entries in the candidate entry set based on the degree of fit, and generate an intermediate recommendation result.

[0029] According to an embodiment of the present disclosure, the intermediate recommendation result generation module can also be used to extract the security requirement characteristic vector of each entry in the candidate entry set, generate a target requirement vector based on the application requirement information; and calculate the degree of fit between the security requirement characteristic vector and the target requirement vector through vector similarity.

[0030] According to an embodiment of the present disclosure, the recommendation result generation module can also be used to display the recommendation items of the intermediate recommendation results through a user interface, wherein the order of the recommendation items of the intermediate recommendation results follows the priority order of the items in the candidate item set; and in response to the user adjusting the order of the recommendation items of the intermediate recommendation results through a drag operation, the user interface is updated in real time.

[0031] A third aspect of the present disclosure provides an electronic device, comprising: one or more processors; and a memory for storing one or more computer programs, wherein the one or more processors execute the one or more computer programs to implement the steps of the above method.

[0032] The fourth aspect of the present disclosure further provides a computer-readable storage medium having a computer program or instructions stored thereon, which implements the steps of the above method when the above computer program or instructions are executed by a processor.

[0033] The fifth aspect of the present disclosure further provides a computer program product, including a computer program or instructions, which implement the steps of the above method when the above computer program or instructions are executed by a processor.

[0034] According to the embodiments of the present disclosure, by combining the generative artificial intelligence algorithm with the security demand knowledge graph, it is possible to dynamically analyze user needs in combination with artificial intelligence technology, reduce the reliance on manually configured rules and data processing in traditional rule matching methods, and improve computing efficiency; introduce a multi-level matching mechanism, reduce the computing burden of irrelevant items through hierarchical screening and optimization processes, and optimize the allocation and utilization of computing resources. At the same time, a real-time adjustment function based on user interaction is provided, so that the recommendation results dynamically respond to user needs, and improve the controllability and satisfaction of users with the recommendation system. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] The above contents and other objects, features and advantages of the present disclosure will become more apparent through the following description of the embodiments of the present disclosure with reference to the accompanying drawings, in which:

[0036] Figure 1 The application scenario diagram of the security requirement recommendation method, apparatus, device, medium and program product according to the embodiment of the present disclosure is schematically shown;

[0037] Figure 2 A flowchart of a security requirement recommendation method according to an embodiment of the present disclosure is schematically shown;

[0038] Figure 3 A flowchart of a method for generating preliminary recommendation results according to an embodiment of the present disclosure is schematically shown;

[0039] Figure 4 A flowchart of a method for generating an intermediate recommendation result according to an embodiment of the present disclosure is schematically shown;

[0040] Figure 5 A structural block diagram of a security requirement recommendation device according to an embodiment of the present disclosure is schematically shown; and

[0041] Figure 6 A block diagram of an electronic device suitable for implementing a security requirement recommendation method according to an embodiment of the present disclosure is schematically shown. DETAILED DESCRIPTION

[0042] Hereinafter, embodiments of the present disclosure will be described with reference to the accompanying drawings. However, it should be understood that these descriptions are exemplary only and are not intended to limit the scope of the present disclosure. In the following detailed description, for ease of explanation, many specific details are set forth to provide a comprehensive understanding of the embodiments of the present disclosure. However, it is apparent that one or more embodiments may also be implemented without these specific details. In addition, in the following description, descriptions of known structures and technologies are omitted to avoid unnecessary confusion of the concepts of the present disclosure.

[0043] The terms used herein are only for describing specific embodiments and are not intended to limit the present disclosure. The terms "comprise", "include", etc. used herein indicate the existence of the features, steps, operations and / or components, but do not exclude the existence or addition of one or more other features, steps, operations or components.

[0044] All terms (including technical and scientific terms) used herein have the meanings commonly understood by those skilled in the art unless otherwise defined. It should be noted that the terms used herein should be interpreted as having a meaning consistent with the context of this specification and should not be interpreted in an idealized or overly rigid manner.

[0045] When using expressions such as "at least one of A, B, and C, etc.", they should generally be interpreted according to the meaning of the expression commonly understood by those skilled in the art (for example, "a system having at least one of A, B, and C" should include but is not limited to a system having A alone, B alone, C alone, A and B, A and C, B and C, and / or A, B, C, etc.).

[0046] First, the technical terms described in this article are explained and described as follows.

[0047] Knowledge Graph is a structured knowledge representation method that organizes information through a graph structure consisting of nodes (entities) and edges (relationships). It is widely used in the fields of knowledge management and information retrieval.

[0048] Generative Artificial Intelligence Content Algorithm is a technology that generates content that meets input requirements through deep learning models. It is often used in scenarios such as natural language generation and image generation.

[0049] Self-Attention Mechanism is a deep learning technique used to capture the correlation between elements in sequence data and is widely used in natural language processing (NLP) tasks.

[0050] Multi-level matching is a hierarchical screening and optimization recommendation algorithm that aims to gradually optimize the recommendation results according to different levels of user needs (such as basic, intermediate, and advanced).

[0051] In the development process of modern information systems and applications, the identification and recommendation of security requirements is a key link in ensuring system security. Faced with a complex and ever-changing security environment, how to accurately and quickly analyze and recommend appropriate security requirements has become one of the key challenges in the development and operation of information systems. Traditional security requirements analysis usually relies on manual experience, which is time-consuming and labor-intensive, and requires a high level of professional knowledge from analysts, making it difficult to meet the needs of modern information technology scenarios for high efficiency and high accuracy.

[0052] In the existing technology, a typical implementation method is to use the application security requirements analysis recommendation method based on knowledge graph: by building a security requirements knowledge graph, the application scenario is quickly analyzed and matched, thereby providing reference and suggestions for security requirements analysts. This method relies on legal and regulatory knowledge and expert knowledge in the field of application security, and uses the reasoning and query capabilities of the knowledge graph to reduce the reliance of security requirements analysis on manual experience to a certain extent and improve the analysis speed.

[0053] However, the existing technology still has some shortcomings and limitations, which are as follows: First, the recommendation results of the existing technology are heavily dependent on the predefined knowledge graph structure and search rules. Once the user needs do not match the preset rules of the graph, it is difficult for the system to provide optimization suggestions for the user. This rigid rule leads to the lack of flexibility and dynamic tuning capabilities of the system, and it is difficult to adapt to complex and changing application scenarios; secondly, due to the lack of a complete update mechanism, it is difficult for the knowledge graph to reflect the latest laws, regulations and safety standards in a timely manner. This makes the security requirements output by the system likely to deviate from the current latest industry specifications, reducing the timeliness and accuracy of the recommendation results; finally, the existing methods often only provide a single level of recommendation, lacking a multi-level matching strategy for different security levels and refined demand levels. Therefore, for users who require higher security scenarios or need more fine-grained matching, the system's recommendation results may not meet their accuracy and adaptability requirements.

[0054] Based on this, an embodiment of the present disclosure provides a security requirement recommendation method, the method comprising: obtaining the requirement information input by the user, the requirement information comprising application requirement information and level requirement information; based on the application requirement information, using a generative artificial intelligence content algorithm in combination with a pre-built security requirement knowledge graph library to generate a preliminary recommendation result; based on the application requirement information, the level requirement information and the preliminary recommendation result, using a security requirement grading model to perform multi-level matching to generate an intermediate recommendation result; and displaying the intermediate recommendation result through a user interface, adjusting the intermediate recommendation result in response to the user's interactive operation, and generating a security requirement recommendation result. The security requirement recommendation method provided by the present disclosure, through the combination of a generative artificial intelligence algorithm and a security requirement knowledge graph, can dynamically analyze user requirements in combination with artificial intelligence technology, reduce the dependence on manually configured rules and data processing in traditional rule matching methods, and improve computing efficiency; introduce a multi-level matching mechanism, reduce the computing burden of irrelevant items through hierarchical screening and optimization processes, and optimize the allocation and utilization of computing resources. At the same time, a real-time adjustment function based on user interaction is provided, so that the recommendation results dynamically respond to user needs, and improve the controllability and satisfaction of users with the recommendation system.

[0055] It should be noted that the security requirement recommendation method, device, equipment, medium and program product determined in the present disclosure can be used in the fields of big data technology and artificial intelligence technology, and can also be used in various fields other than the fields of big data technology and artificial intelligence technology. The application fields of the security requirement recommendation method, device, equipment, medium and program product provided by the embodiments of the present disclosure are not limited.

[0056] In the technical solution of the present disclosure, the user information (including but not limited to user personal information, user image information, user device information, such as location information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved are all information and data authorized by the user or fully authorized by all parties, and the collection, storage, use, processing, transmission, provision, disclosure and application of the relevant data comply with relevant laws, regulations and standards, take necessary confidentiality measures, do not violate public order and good morals, and provide corresponding operation entrances for users to choose to authorize or refuse.

[0057] In the scenario of using personal information for automated decision-making, the methods, devices, and systems provided by the embodiments of the present disclosure provide users with corresponding operation portals for users to choose to agree or reject the automated decision-making results; if the user chooses to reject, the expert decision-making process will be entered. The expression "automated decision-making" here refers to the activity of automatically analyzing and evaluating a person's behavioral habits, interests and hobbies, or economic, health, credit status, etc. through computer programs, and making decisions. The expression "expert decision-making" here refers to the activity of making decisions by people who specialize in a certain field, have specialized experience, knowledge and skills, and have reached a certain level of professionalism.

[0058] Figure 1 The application scenario diagram of the security requirement recommendation method, apparatus, device, medium and program product according to the embodiments of the present disclosure is schematically shown.

[0059] like Figure 1 As shown, the application scenario 100 according to this embodiment may include a first terminal device 101, a second terminal device 102, a third terminal device 103, a network 104, and a server 105. The network 104 is used to provide a medium for a communication link between the first terminal device 101, the second terminal device 102, the third terminal device 103, and the server 105. The network 104 may include various connection types, such as wired, wireless communication links, or optical fiber cables, etc.

[0060] The user can use the first terminal device 101, the second terminal device 102, and the third terminal device 103 to interact with the server 105 through the network 104 to receive or send messages, etc. Various communication client applications can be installed on the first terminal device 101, the second terminal device 102, and the third terminal device 103, such as shopping applications, web browser applications, search applications, instant messaging tools, email clients, social platform software, etc. (only for example).

[0061] The first terminal device 101, the second terminal device 102, and the third terminal device 103 may be various electronic devices having display screens and supporting web browsing, including but not limited to smart phones, tablet computers, laptop computers, desktop computers, and the like.

[0062] The server 105 may be a server that provides various services, such as a background management server (only as an example) that provides support for websites browsed by users using the first terminal device 101, the second terminal device 102, and the third terminal device 103. The background management server may analyze and process the received data such as user requests, and feed back the processing results (such as web pages, information, or data obtained or generated according to user requests) to the terminal device.

[0063] It should be noted that the security requirement recommendation method provided in the embodiment of the present disclosure can generally be executed by the server 105. Accordingly, the security requirement recommendation device provided in the embodiment of the present disclosure can generally be set in the server 105. The security requirement recommendation method provided in the embodiment of the present disclosure can also be executed by a server or server cluster that is different from the server 105 and can communicate with the first terminal device 101, the second terminal device 102, the third terminal device 103 and / or the server 105. Accordingly, the security requirement recommendation device provided in the embodiment of the present disclosure can also be set in a server or server cluster that is different from the server 105 and can communicate with the first terminal device 101, the second terminal device 102, the third terminal device 103 and / or the server 105.

[0064] It should be understood that Figure 1 The number of terminal devices, networks and servers in the embodiment is only for illustration. Any number of terminal devices, networks and servers may be provided according to implementation requirements.

[0065] The following will be based on Figure 1 The scene described by Figure 2 to Figure 4 The security requirement recommendation method of the disclosed embodiment is described in detail.

[0066] Figure 2 The flowchart of the security requirement recommendation method according to the embodiment of the present disclosure is schematically shown.

[0067] like Figure 2 As shown, the security requirement recommendation method of this embodiment includes operations S210 to S240, and the transaction processing method can be executed by the server 105.

[0068] In operation S210, the requirement information input by the user is obtained, and the requirement information includes application requirement information and level requirement information. Specifically, the application requirement information can be represented as functional description information about the target requirement input by the user. For example, it can include information such as application function, operating environment and running environment.

[0069] Application functions can represent the business functions that the application developed or deployed by the user needs to implement, and are used to identify the core requirements of the application, thereby inferring the security risks it may face and the security technical support it needs. For example, it can include: user identity management, whether the application requires user registration, login, permission management and other functions, such as user authentication, role allocation, etc.; data storage and processing, whether the application involves the storage or processing of sensitive data, such as payment data, medical data, personal privacy information, etc.; external interface interaction, whether the application needs to interact with other systems or services, such as calling third-party APIs and accessing external databases; business logic functions, the core functional modules of the application, such as the order processing function of the e-commerce platform and the messaging function of the social platform; multimedia processing, whether the application involves the upload, processing and display of multimedia data such as images.

[0070] The operating environment and runtime environment describe the deployment location of the application, the runtime environment, and the external conditions it faces, which affect the security measures that the application needs to adapt to and determine the adaptability of security requirements. For example, cloud deployment requires extra attention to virtualization isolation and multi-tenant data security, while mobile terminals need to focus on mobile operating system vulnerabilities and data transmission encryption.

[0071] Level requirement information can indicate the user's expected level of security requirements and is an important parameter that affects the recommendation of security technology items. Level requirement information can be related to the risk level, sensitivity and compliance requirements of the application. For example, it can include: basic level, which is suitable for low-sensitivity applications and provides basic security measures, such as simple identity authentication and basic firewalls; intermediate level, which is suitable for medium-sensitivity applications and adds more security features, such as data encryption and access control lists; advanced level, which is suitable for high-sensitivity or high-risk scenarios, such as finance, medical care, etc., and includes advanced security measures such as multi-factor authentication, behavioral analysis and zero-trust architecture. In addition, compliance levels can also be divided according to specific industry regulatory requirements.

[0072] In operation S220, based on the application requirement information, a preliminary recommendation result is generated by utilizing a generative artificial intelligence content algorithm in combination with a pre-built security requirement knowledge graph library.

[0073] Figure 3 The flowchart of the method for generating preliminary recommendation results according to an embodiment of the present disclosure is schematically shown.

[0074] like Figure 3 As shown, the method for generating preliminary recommendation results in this embodiment may include operations S310 to S330.

[0075] In operation S310, a generative artificial intelligence algorithm is used to perform semantic analysis and context understanding on the application requirement information to obtain semantic features of user requirements.

[0076] Generative AI algorithms achieve efficient semantic analysis and understanding of user needs by integrating a variety of deep learning technologies and natural language processing methods. Specifically, the core components of generative AI algorithms can include deep learning models, self-attention mechanisms, and contextual semantic aggregation mechanisms.

[0077] In an embodiment of the present disclosure, the process of using a generative artificial intelligence algorithm to perform semantic analysis and context understanding on the application requirement information may include:

[0078] 1) Perform data preprocessing on the application requirement information to ensure the integrity and consistency of the input data. Exemplarily, data preprocessing may include format standardization of user input information, removal of redundant information and noise data, and word segmentation, stop word removal, and morphological restoration of the input text. These operations can effectively reduce the redundancy of the input data and facilitate subsequent semantic analysis.

[0079] 2) Generate text embedding for the preprocessed application requirement information to obtain a semantic feature vector. Specifically, as the core component of the generative artificial intelligence algorithm, the deep learning model can map the input text into a high-dimensional vector representation. The semantic feature vector can capture the explicit and implicit semantic features of the user input and include context-dependent information.

[0080] 3) Analyze the semantic feature vector through the self-attention mechanism to obtain global semantic information. Specifically, the self-attention mechanism is an important part of the generative artificial intelligence algorithm, which is used to calculate the association weights between different parts of the input text and capture key semantic relationships and contextual dependencies. This process ensures that the system focuses on the most relevant information when processing long texts or complex requirements, while retaining the global structure of the input data. By integrating local and global information, the generative artificial intelligence algorithm can extract semantic features more accurately and improve the comprehensiveness and depth of demand analysis.

[0081] 4) Aggregating the semantic feature vector and the global semantic information to obtain the user demand semantic feature. The aggregation process can use weighted summation, hierarchical attention or other aggregation techniques to effectively combine local and global features to ensure that the output user demand semantic feature can fully reflect the context information and semantic structure of the application demand.

[0082] In operation S320, based on the security requirement knowledge graph library, security requirement nodes related to the application requirement information are retrieved to generate a node feature vector.

[0083] In an embodiment of the present disclosure, a security requirement knowledge graph library can be constructed from security requirement information extracted from laws, regulations, security standards, and industry best practices, and includes nodes (such as security items) and relationships between them (such as dependencies or applicable conditions). Specifically, by parsing the key semantic features in user requirements, the graph retrieval algorithm can be used to locate security requirement nodes related to application requirement information. The retrieval results are represented in the form of feature vectors of the nodes, which contain the semantic information, contextual relationships, and weight information of the nodes.

[0084] In operation S330, the preliminary recommendation result is generated based on interaction modeling between the user demand semantic feature and the node feature vector.

[0085] In an embodiment of the present disclosure, the process of generating preliminary recommendation results based on interactive modeling of user demand semantic features and node feature vectors may include:

[0086] 1) Calculate the first similarity between the user demand semantic feature and the node feature vector. Specifically, mathematical methods such as cosine similarity, Euclidean distance or dot product can be used to quantify the similarity between the user demand semantic feature vector and the security demand node feature vector. This process can preliminarily identify nodes that have a high correlation with user demand.

[0087] 2) Based on the context information of the application requirement information, the similarity is weighted to obtain a second similarity. The context information includes but is not limited to the specific scenario, operation and operating environment of the application. By introducing a weight adjustment mechanism, the similarity value can be dynamically corrected so that the recommendation results can more accurately reflect the actual scenario of the user's needs.

[0088] 3) Sort the security requirement nodes based on the second similarity, and obtain the preliminary recommendation result based on the sorting result. The preliminary recommendation result may include multiple recommendation items.

[0089] Furthermore, by obtaining user feedback on security requirement recommendation results based on preset time intervals, support can be provided for optimizing recommendation algorithms and knowledge graphs. Specifically, during operation, the user's interaction with the recommendation results can be regularly monitored, including the acceptance, rejection, and modification of the recommended items, as well as the user's specific feedback on the recommended content. These feedback information are stored in a structured or unstructured form, and are sorted and standardized through a preprocessing module to extract key data features, such as the user's modification frequency, the directionality and acceptance of the modified content, etc. Furthermore, these feedback information can also be used to analyze the user's preference patterns and demand change trends for the recommendation results, and generate feedback feature data.

[0090] In combination with the collected feedback feature data, the security requirements knowledge graph can be dynamically updated. Specifically, the user's modification frequency and modification mode of the security requirement recommendation results can be extracted, and the node association weights can be adjusted based on the modification frequency and modification mode. For example, if a user modifies or deletes a certain type of item multiple times, it may indicate that its matching degree is low. In this case, the weight of the item node and the related requirement node can be reduced accordingly to optimize the association logic in the graph. In addition, for items that are frequently selected or accepted by users, their priority in the knowledge graph can be increased, so that subsequent recommendations are closer to the user's actual needs. For new user needs or security items not covered in the feedback, the graph nodes can also be supplemented and updated through knowledge extraction technology to ensure that the knowledge graph can reflect the latest needs in a timely manner.

[0091] In the embodiments of the present disclosure, feedback information can be used to adjust the recommendation strategy of the generative artificial intelligence content algorithm. For example, the recommendation logic can be dynamically optimized based on the feedback data, such as giving priority to recommending item categories with high user acceptance and reducing the recommended content that deviates greatly from user needs.

[0092] Preferably, for changes in demand preferences that occur multiple times, the generative AI content algorithm can be dynamically adjusted through a reinforcement learning algorithm, so that the recommendation results continue to adapt to the changing trends of user demand. By combining feedback information to coordinate the optimization of the knowledge graph and recommendation strategy, dynamic adjustment, enhanced personalization, and improved timeliness of recommended content can be achieved, thereby improving the accuracy of recommendation results and user satisfaction.

[0093] In the embodiments of the present disclosure, authoritative support and professional guidance can be provided for security requirement recommendations by constructing an expert knowledge base that contains the knowledge and experience of security experts. The expert knowledge base can include a series of security rules and logic defined by experts in the field, and stored in a structured manner to be directly applied to the analysis and recommendation process of security requirements. In order to ensure the timeliness and accuracy of the knowledge base, a preset time interval can also be set to regularly update and maintain the knowledge base. The updated content mainly comes from the experts' newly added rules, the latest developments in the security field, and user feedback data to ensure that the knowledge base always reflects the current security needs and technological developments of the industry.

[0094] According to the embodiments of the present disclosure, the rules in the expert knowledge base can be used to train and optimize the generative artificial intelligence content algorithm. Specifically, in the training stage of the artificial intelligence algorithm, the expert rules can be used as additional supervisory signals to guide the model to be closer to the actual application needs in the process of semantic analysis, feature extraction and matching calculations. For example, expert rules can provide the algorithm with key semantic relationships or dependency information between security requirements, thereby improving the model's ability to adapt to complex scenarios. In the model optimization stage, the algorithm's parameters and logic can be dynamically adjusted according to the newly added rules in the knowledge base, so that the algorithm can continuously improve the accuracy and professionalism of the recommendation results.

[0095] According to an embodiment of the present disclosure, the expert knowledge base that is regularly updated and maintained can also be accessed by the user at any time during use to obtain professional advice and guidance.

[0096] Return to reference Figure 2 In operation S230, based on the application requirement information, the level requirement information and the preliminary recommendation result, a security requirement grading model is used to perform multi-level matching to generate an intermediate recommendation result.

[0097] Figure 4 The flowchart of the method for generating an intermediate recommendation result according to an embodiment of the present disclosure is schematically shown.

[0098] like Figure 4 As shown, the method for generating an intermediate recommendation result in this embodiment may include operations S410 to S430.

[0099] In operation S410, based on the level requirement information, a recommendation target level is defined in the security requirement grading model, and the recommended items in the preliminary recommendation result are divided into a plurality of level sets using the security requirement grading model.

[0100] In an embodiment of the present disclosure, the security requirement grading model can be a structured hierarchical model, which divides the security requirement items into multiple levels (e.g., basic level, intermediate level, and advanced level) in advance according to the security requirement complexity, compliance requirements, and risk level of the application scenario. Specifically, based on the level requirement information input by the user, the user's expectations for the strictness of the security requirements, that is, the recommended target level, can be extracted. Exemplarily, the level requirement information can be obtained through explicit level labels (such as "intermediate", "advanced") or implicit security requirement descriptions (such as "need to meet financial industry regulations").

[0101] After the recommended target level is determined, the security requirement items in the preliminary recommendation results can be hierarchically divided according to the security requirement grading model. Each recommended item is matched with the grade standard in the security requirement grading model through its characteristic label (such as scope of application, security level, complexity index, etc.) and classified into the corresponding hierarchical set. For example, a simple identity authentication item may be classified as a "basic level", while a multi-factor authentication scheme may belong to an "advanced level". Through this hierarchical division mechanism, the preliminary recommendation results can be organized into a structured hierarchical set, ensuring that the subsequent screening process can quickly focus on items related to the target level.

[0102] In operation S420, the recommended items are filtered layer by layer based on the multiple level sets to generate a set of candidate items that meet the target level. The core of layer by layer filtering is to gradually eliminate items that do not match the recommended target level, while retaining those items that meet the level requirements and may supplement the diversified recommendations.

[0103] In an embodiment of the present disclosure, the layer-by-layer filtering process can start from a set of lower-level entries: for each entry, its security characteristics (such as applicable scenarios, covered security requirement categories, and technical complexity) can be checked to see if they meet the minimum requirements of the recommended target level. If the entry attributes are lower than the target level standard, it can be marked as a mismatched entry and removed from the recommended entry set. For example, if the recommended target level is "intermediate", all entries classified as "basic level" will be removed to ensure that the security requirement entries in the candidate set meet at least the "intermediate" standard or higher.

[0104] Furthermore, for items above the target level (such as advanced items), they can be retained or prioritized according to their relevance and applicability to the application requirement information.

[0105] In operation S430, the degree of compatibility between each item in the candidate item set and the application requirement information is calculated, and the priority order of items in the candidate item set is adjusted based on the degree of compatibility to generate an intermediate recommendation result.

[0106] In an embodiment of the present disclosure, when calculating the degree of fit between each entry in a set of candidate entries and the application requirement information, the security requirement feature vector of each entry in the set of candidate entries can be first extracted. The security requirement feature vector can be extracted from the security requirement knowledge graph, and contains, for example, the semantic features of the entry itself, security function attributes, applicable scenarios, and related context information. The construction of the security requirement feature vector is based on the node representation of the entry in the knowledge graph and its association with other nodes. The nodes in the knowledge graph are embedded through a deep learning model (such as a neural network in the figure) to generate a high-dimensional feature vector. These vectors comprehensively characterize the technical characteristics of the security requirement items and their semantic position in the security requirement system.

[0107] Meanwhile, a target requirement vector may be generated according to the application requirement information provided by the user. Exemplarily, the target requirement vector may be generated by text embedding technology to capture the semantic features and contextual relationships of the application requirements.

[0108] According to the embodiments of the present disclosure, the calculation of the adaptability not only takes into account the semantic consistency between the item and the demand, but also combines the application priority of the actual scenario, so that the sorting of recommended items is more in line with user needs, further optimizing the accuracy and adaptability of the recommendation results.

[0109] In an embodiment of the present disclosure, the degree of fit between the security requirement characteristic vector and the target requirement vector may be calculated by vector similarity. For example, the cosine similarity formula may be used to specifically reflect the degree of similarity between two vectors by calculating the cosine value of the angle between the two vectors.

[0110] Return to reference Figure 2 In operation S240, the intermediate recommendation result is displayed through a user interface, and a security requirement recommendation result is generated in response to a user's interactive operation.

[0111] For example, the recommended items can be displayed in the user interface in the form of a list or card according to the priority order of the items in the candidate item set described above, and key attribute information such as applicable scenarios, security levels, implementation complexity, and reasons for recommendation can be attached to each item to facilitate users to quickly understand and select. In this way, users can obtain recommended content at a glance and quickly locate high-priority items according to their needs.

[0112] Furthermore, in order to improve the flexibility of recommendation results and user experience, users can also adjust the order of recommended items by dragging and dropping. Users can drag or touch the screen to move the items to the desired position to rearrange the priority of the recommended items. After each adjustment, the user's operation can be captured in real time and the order of the recommendation list can be dynamically updated, and the new arrangement results can be instantly reflected in the user interface. This interactive mechanism allows users to adjust the system's recommendation results according to specific needs, enhancing controllability and personalization capabilities.

[0113] Based on the above security requirement recommendation method, the present disclosure also provides a security requirement recommendation device. Figure 5 The device is described in detail.

[0114] Figure 5 The structural block diagram of the security requirement recommendation device according to an embodiment of the present disclosure is schematically shown.

[0115] like Figure 5 As shown, the security requirement recommendation device 500 of this embodiment includes an input acquisition module 510 , a preliminary recommendation result generation module 520 , an intermediate recommendation result generation module 530 and a recommendation result generation module 540 .

[0116] The input acquisition module 510 may be used to acquire the requirement information input by the user, wherein the requirement information includes application requirement information and level requirement information. In one embodiment, the input acquisition module 510 may be used to perform the operation S210 described above, which will not be described in detail here.

[0117] The preliminary recommendation result generation module 520 can be used to generate preliminary recommendation results based on the application requirement information using a generative artificial intelligence content algorithm combined with a pre-built security requirement knowledge graph library. In one embodiment, the preliminary recommendation result generation module 520 can be used to perform the operation S220 described above, which will not be repeated here.

[0118] The intermediate recommendation result generation module 530 can be used to generate an intermediate recommendation result by performing multi-level matching based on the application requirement information, the level requirement information and the preliminary recommendation result using the security requirement classification model. In one embodiment, the intermediate recommendation result generation module 530 can be used to perform the operation S230 described above, which will not be described in detail here.

[0119] The recommendation result generation module 540 may be used to display the intermediate recommendation result through a user interface, adjust the intermediate recommendation result in response to a user's interactive operation, and generate a security requirement recommendation result. In one embodiment, the recommendation result generation module 540 may be used to perform the operation S240 described above, which will not be described in detail here.

[0120] According to an embodiment of the present disclosure, the security requirement analysis result generating module 520 may include a preliminary recommendation module, which may include a semantic analysis unit, a node retrieval unit, and an interaction modeling unit.

[0121] The semantic analysis unit may be used to perform semantic analysis and context understanding on the application requirement information using a generative artificial intelligence algorithm to obtain semantic features of user requirements. In one embodiment, the semantic analysis unit may be used to perform the operation S310 described above, which will not be described in detail herein.

[0122] The node retrieval unit may be used to retrieve security requirement nodes related to the application requirement information based on the security requirement knowledge graph library to generate a node feature vector. In one embodiment, the node retrieval unit may be used to perform the operation S320 described above, which will not be described in detail here.

[0123] The interactive modeling unit may be used to generate the preliminary recommendation result based on the interactive modeling of the user demand semantic features and the node feature vectors. In one embodiment, the interactive modeling unit may be used to perform the operation S330 described above, which will not be described in detail here.

[0124] According to an embodiment of the present disclosure, the preliminary recommendation result generation module 520 can also be used to perform data preprocessing on the application requirement information; generate text embedding for the preprocessed application requirement information to obtain a semantic feature vector; analyze the semantic feature vector through a self-attention mechanism to obtain global semantic information; and aggregate the semantic feature vector and the global semantic information to obtain the user requirement semantic feature.

[0125] According to an embodiment of the present disclosure, the preliminary recommendation result generation module 520 can also be used to calculate a first similarity between the user demand semantic feature and the node feature vector; based on the context information of the application demand information, adjust the weight of the first similarity to obtain a second similarity; and sort the security demand nodes based on the second similarity, and obtain the preliminary recommendation result based on the sorting result.

[0126] According to an embodiment of the present disclosure, the preliminary recommendation result generation module 520 can also be used to obtain user feedback information on the security requirement recommendation results based on a preset time interval; and based on the feedback information, update the security requirement knowledge graph and / or adjust the recommendation strategy of the generative artificial intelligence content algorithm.

[0127] According to an embodiment of the present disclosure, the preliminary recommendation result generation module 520 can also be used to extract the user's modification frequency and modification pattern of the security requirement recommendation result based on the feedback information; and dynamically adjust the node association weights in the knowledge graph based on the statistical results of the modification frequency and modification pattern.

[0128] According to an embodiment of the present disclosure, the preliminary recommendation result generation module 520 can also be used to build an expert knowledge base containing security expert knowledge and experience, and the expert knowledge base is updated and maintained based on a preset time interval; and based on the expert rules defined in the expert knowledge base, the generative artificial intelligence content algorithm is trained and optimized.

[0129] According to an embodiment of the present disclosure, the intermediate recommendation result generation module 530 may include an intermediate recommendation result generation module, and the intermediate recommendation result generation module may include a level definition unit, a filtering unit, and a calculation unit.

[0130] The level definition unit may be used to define a recommended target level in the security requirement grading model based on the level requirement information, and divide the recommended items in the preliminary recommendation result into a plurality of level sets using the security requirement grading model. In one embodiment, the level definition unit may be used to perform the operation S410 described above, which will not be described in detail here.

[0131] The filtering unit may be used to filter the recommended items layer by layer based on multiple level sets to generate a set of candidate items that meet the recommendation target level. In one embodiment, the filtering unit may be used to perform the operation S420 described above, which will not be described in detail here.

[0132] The calculation unit may be used to calculate the degree of compatibility of each entry in the candidate entry set with the application requirement information, adjust the priority order of the entries in the candidate entry set based on the degree of compatibility, and generate an intermediate recommendation result. In one embodiment, the calculation unit may be used to perform the operation S430 described above, which will not be described in detail here.

[0133] According to an embodiment of the present disclosure, the intermediate recommendation result generation module 530 can also be used to extract the security requirement characteristic vector of each entry in the candidate entry set, generate a target requirement vector based on the application requirement information; and calculate the degree of fit between the security requirement characteristic vector and the target requirement vector through vector similarity.

[0134] According to an embodiment of the present disclosure, the recommendation result generation module 540 can also be used to display the recommendation items of the intermediate recommendation results through a user interface, wherein the order of the recommendation items of the intermediate recommendation results follows the priority order of the items in the candidate item set; and in response to the user adjusting the order of the recommendation items of the intermediate recommendation results through a drag operation, the user interface is updated in real time.

[0135] According to an embodiment of the present disclosure, any multiple modules of the input acquisition module 510, the preliminary recommendation result generation module 520, the intermediate recommendation result generation module 530, and the recommendation result generation module 540 can be combined into one module for implementation, or any one of the modules can be split into multiple modules. Alternatively, at least part of the functions of one or more of these modules can be combined with at least part of the functions of other modules and implemented in one module. According to an embodiment of the present disclosure, at least one of the input acquisition module 510, the preliminary recommendation result generation module 520, the intermediate recommendation result generation module 530, and the recommendation result generation module 540 can be at least partially implemented as a hardware circuit, such as a field programmable gate array (FPGA), a programmable logic array (PLA), a system on a chip, a system on a substrate, a system on a package, an application specific integrated circuit (ASIC), or can be implemented by hardware or firmware such as any other reasonable way of integrating or packaging the circuit, or implemented in any one of the three implementation methods of software, hardware, and firmware, or in any appropriate combination of any of them. Alternatively, at least one of the input acquisition module 510, the preliminary recommendation result generation module 520, the intermediate recommendation result generation module 530 and the recommendation result generation module 540 may be at least partially implemented as a computer program module, which may perform a corresponding function when executed.

[0136] Figure 6 A block diagram of an electronic device suitable for implementing a security requirement recommendation method according to an embodiment of the present disclosure is schematically shown.

[0137] like Figure 6 As shown, the electronic device 600 according to an embodiment of the present disclosure includes a processor 601, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 602 or a program loaded from a storage part 608 to a random access memory (RAM) 603. The processor 601 may include, for example, a general-purpose microprocessor (e.g., a CPU), an instruction set processor and / or a related chipset and / or a dedicated microprocessor (e.g., an application-specific integrated circuit (ASIC)), etc. The processor 601 may also include an onboard memory for caching purposes. The processor 601 may include a single processing unit or multiple processing units for performing different actions of the method flow according to an embodiment of the present disclosure.

[0138] In RAM 603, various programs and data required for the operation of electronic device 600 are stored. Processor 601, ROM 602 and RAM 603 are connected to each other through bus 604. Processor 601 performs various operations of the method flow according to the embodiment of the present disclosure by executing the program in ROM 602 and / or RAM 603. It should be noted that the program can also be stored in one or more memories other than ROM 602 and RAM 603. Processor 601 can also perform various operations of the method flow according to the embodiment of the present disclosure by executing the program stored in the one or more memories.

[0139] According to an embodiment of the present disclosure, the electronic device 600 may further include an input / output (I / O) interface 605, which is also connected to the bus 604. The electronic device 600 may further include one or more of the following components connected to the input / output (I / O) interface 605: an input portion 606 including a keyboard, a mouse, etc.; an output portion 607 including a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker, etc.; a storage portion 608 including a hard disk, etc.; and a communication portion 609 including a network interface card such as a LAN card, a modem, etc. The communication portion 609 performs communication processing via a network such as the Internet. A drive 610 is also connected to the input / output (I / O) interface 605 as needed. A removable medium 611, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 610 as needed, so that a computer program read therefrom is installed into the storage portion 608 as needed.

[0140] The present disclosure also provides a computer-readable storage medium, which may be included in the device / apparatus / system described in the above embodiments; or may exist independently without being assembled into the device / apparatus / system. The above computer-readable storage medium carries one or more programs, and when the above one or more programs are executed, the method according to the embodiment of the present disclosure is implemented.

[0141] According to an embodiment of the present disclosure, the computer-readable storage medium may be a non-volatile computer-readable storage medium, for example, it may include but is not limited to: a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In the present disclosure, a computer-readable storage medium may be any tangible medium containing or storing a program, which may be used by or in combination with an instruction execution system, an apparatus or a device. For example, according to an embodiment of the present disclosure, the computer-readable storage medium may include the ROM 602 and / or RAM 603 described above and / or one or more memories other than ROM 602 and RAM 603.

[0142] The embodiment of the present disclosure also includes a computer program product, which includes a computer program, and the computer program contains program code for executing the method shown in the flowchart. When the computer program product is run in a computer system, the program code is used to enable the computer system to implement the security requirement recommendation method provided by the embodiment of the present disclosure.

[0143] The above functions defined in the system / device of the embodiment of the present disclosure are performed when the computer program is executed by the processor 601. According to the embodiment of the present disclosure, the system, device, module, unit, etc. described above can be implemented by a computer program module.

[0144] In one embodiment, the computer program may rely on tangible storage media such as optical storage devices, magnetic storage devices, etc. In another embodiment, the computer program may also be transmitted and distributed in the form of signals on a network medium, and downloaded and installed through the communication part 609, and / or installed from a removable medium 611. The program code contained in the computer program may be transmitted using any appropriate network medium, including but not limited to: wireless, wired, etc., or any suitable combination of the above.

[0145] In such an embodiment, the computer program can be downloaded and installed from the network through the communication part 609, and / or installed from the removable medium 611. When the computer program is executed by the processor 601, the above functions defined in the system of the embodiment of the present disclosure are performed. According to the embodiment of the present disclosure, the system, device, apparatus, module, unit, etc. described above can be implemented by a computer program module.

[0146] According to an embodiment of the present disclosure, the program code for executing the computer program provided by the embodiment of the present disclosure can be written in any combination of one or more programming languages. Specifically, these computing programs can be implemented using high-level process and / or object-oriented programming languages, and / or assembly / machine languages. Programming languages ​​include, but are not limited to, Java, C++, python, "C" language or similar programming languages. The program code can be executed entirely on the user computing device, partially on the user device, partially on the remote computing device, or entirely on the remote computing device or server. In the case of a remote computing device, the remote computing device can be connected to the user computing device through any type of network, including a local area network (LAN) or a wide area network (WAN), or can be connected to an external computing device (for example, using an Internet service provider to connect through the Internet).

[0147] The flow charts and block diagrams in the accompanying drawings illustrate the possible architecture, functions and operations of the systems, methods and computer program products according to various embodiments of the present disclosure. In this regard, each box in the flow chart or block diagram can represent a module, a program segment, or a part of a code, and the above-mentioned module, program segment, or a part of a code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram or flow chart, and the combination of the boxes in the block diagram or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0148] It will be appreciated by those skilled in the art that the features described in the various embodiments of the present disclosure may be combined and / or combined in a variety of ways, even if such combinations or combinations are not explicitly described in the present disclosure. In particular, without departing from the spirit and teachings of the present disclosure, the features described in the various embodiments of the present disclosure may be combined and / or combined in a variety of ways. All of these combinations and / or combinations fall within the scope of the present disclosure.

[0149] The embodiments of the present disclosure are described above. However, these embodiments are only for illustrative purposes and are not intended to limit the scope of the present disclosure. Although the embodiments are described above, this does not mean that the measures in the various embodiments cannot be used in combination to advantage. Without departing from the scope of the present disclosure, those skilled in the art may make a variety of substitutions and modifications, which should all fall within the scope of the present disclosure.

Claims

1. A security requirement recommendation method, characterized in that: The method comprises: Acquire demand information input by a user, wherein the demand information includes application demand information and level demand information; Based on the application requirement information, a generative artificial intelligence content algorithm is used in combination with a pre-built security requirement knowledge graph library to generate preliminary recommendation results; Based on the application requirement information, the level requirement information and the preliminary recommendation result, a security requirement grading model is used to perform multi-level matching to generate an intermediate recommendation result; and The intermediate recommendation result is displayed through a user interface, and the intermediate recommendation result is adjusted in response to a user's interactive operation to generate a security requirement recommendation result.

2. The method according to claim 1, characterized in that Based on the application requirement information, the generative artificial intelligence content algorithm is combined with a pre-built security requirement knowledge graph library to generate preliminary recommendation results, specifically including: Using a generative artificial intelligence algorithm to perform semantic analysis and context understanding on the application requirement information to obtain semantic features of user requirements; Based on the security requirement knowledge graph library, searching for security requirement nodes related to the application requirement information and generating a node feature vector; and The preliminary recommendation result is generated based on the interaction modeling of the user demand semantic feature and the node feature vector.

3. The method according to claim 2, characterized in that The method of using a generative artificial intelligence algorithm to perform semantic analysis and context understanding on the application requirement information to obtain semantic features of user requirements specifically includes: Performing data preprocessing on the application requirement information; Generate text embedding for the preprocessed application requirement information to obtain a semantic feature vector; Analyzing the semantic feature vector through a self-attention mechanism to obtain global semantic information; and The semantic feature vector and the global semantic information are aggregated to obtain the user demand semantic feature.

4. The method according to claim 2 or 3, characterized in that: The generating the preliminary recommendation result through interactive modeling of the user demand semantic feature and the node feature vector specifically includes: Calculating a first similarity between the user demand semantic feature and the node feature vector; Based on the context information of the application requirement information, weighting the first similarity to obtain a second similarity; and The security requirement nodes are sorted based on the second similarity, and the preliminary recommendation result is generated based on the sorting result.

5. The method according to claim 4, characterized in that The preliminary recommendation result includes a plurality of recommendation items. Based on the application requirement information, the level requirement information and the preliminary recommendation result, a multi-level matching is performed using a security requirement grading model to generate an intermediate recommendation result, which specifically includes: Based on the level requirement information, defining a recommended target level in the security requirement grading model, and using the security requirement grading model to divide the recommended items in the preliminary recommendation result into a plurality of level sets; Filter the recommended items layer by layer based on the multiple hierarchical sets to generate a set of candidate items that meet the recommendation target level; and The degree of compatibility between each entry in the candidate entry set and the application requirement information is calculated, and the priority order of the entries in the candidate entry set is adjusted based on the degree of compatibility to generate an intermediate recommendation result.

6. The method according to claim 5, characterized in that The step of calculating the degree of compatibility of each entry in the candidate entry set with the application requirement information specifically includes: extracting a security requirement characteristic vector of each entry in the candidate entry set, and generating a target requirement vector based on the application requirement information; and The degree of adaptation between the security requirement characteristic vector and the target requirement vector is calculated through vector similarity.

7. The method according to claim 5, characterized in that Displaying the intermediate recommendation result through a user interface and adjusting the intermediate recommendation result in response to a user's interactive operation specifically includes: Displaying the recommended items of the intermediate recommendation result through a user interface, wherein the order of the recommended items of the intermediate recommendation result follows the priority order of the items in the candidate item set; and In response to the user adjusting the order of the recommendation items of the intermediate recommendation result through a drag operation, the user interface is updated in real time.

8. The method according to any one of claims 1 to 3, 5 to 7, characterized in that: The method further comprises: Obtaining user feedback on the security requirement recommendation result based on a preset time interval; and In combination with the feedback information, the security requirement knowledge graph is updated and / or the recommendation strategy of the generative artificial intelligence content algorithm is adjusted.

9. The method according to claim 8, characterized in that The updating of the security requirement knowledge graph specifically includes: Based on the feedback information, extracting the user's modification frequency and modification mode of the security requirement recommendation result; and Based on the statistical results of modification frequency and modification mode, the node association weights in the knowledge graph are dynamically adjusted.

10. The method according to any one of claims 1 to 3 and 5 to 7, characterized in that: The method further comprises: Building an expert knowledge base containing security expert knowledge and experience, wherein the expert knowledge base is updated and maintained based on a preset time interval; and The generative artificial intelligence content algorithm is trained and optimized based on the expert rules defined in the expert knowledge base.

11. A security requirement recommendation device, characterized in that: The device comprises: An input acquisition module is used to: acquire demand information input by a user, wherein the demand information includes application demand information and level demand information; A preliminary recommendation result generation module is used to: generate preliminary recommendation results based on the application requirement information by using a generative artificial intelligence content algorithm combined with a pre-built security requirement knowledge graph library; An intermediate recommendation result generating module, used to: generate an intermediate recommendation result by performing multi-level matching using a security requirement grading model based on the application requirement information, the level requirement information and the preliminary recommendation result; and The recommendation result generation module is used to: display the intermediate recommendation result through a user interface, adjust the intermediate recommendation result in response to the user's interactive operation, and generate a security requirement recommendation result.

12. An electronic device comprising: one or more processors; a memory for storing one or more computer programs, It is characterized in that the one or more processors execute the one or more computer programs to implement the steps of the method according to any one of claims 1 to 10.

13. A computer-readable storage medium having a computer program or instruction stored thereon, characterized in that: When the computer program or instruction is executed by a processor, the steps of the method according to any one of claims 1 to 10 are implemented.

14. A computer program product comprising a computer program or instructions, characterized in that When the computer program or instruction is executed by a processor, the steps of the method according to any one of claims 1 to 10 are implemented.

Images