Data security risk assessment and control method and related device

Through the improved Apriori algorithm and NSGA-II algorithm, the network data security risk assessment and control problems of multiple systems in complex organizations are solved, and fast and accurate risk assessment and control are achieved, providing diversified strategy selection and resource optimization , improves the applicability and scalability of the system.

CN118944924APending Publication Date: 2024-11-12GUANGXI POWER GRID CORP
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410984259.8
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-22
Publication Date
2024-11-12

AI Technical Summary

Technical Problem

The prior art is difficult to effectively analyze and control network data security risks in multiple systems in complex organizations, especially in large enterprises or organizations, where the applicability and accuracy of existing methods are limited.

Method used

Using the improved Apriori algorithm and NSGA-II algorithm, the transaction item set is extracted from network data, frequent item sets are generated, and the dual-objective optimal risk control model is constructed. The NSGA-II algorithm is used to solve it and select the best through the maximum satisfaction method. A compromise solution to achieve the assessment and control of data security risks.

Benefits of technology

It significantly reduces the number of intermediate item sets during data processing, reduces memory consumption and computing time, provides fast risk assessment and control, can find the best balance point between multiple goals, generate diversified risk control strategies, and improves The system is applied and scalable, and the resource allocation is optimized to achieve the security and stability of network data.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118944924A_ABST
    Figure CN118944924A_ABST
Patent Text Reader

Abstract

The invention discloses a data security risk assessment and control method and a related device, and the method comprises the steps: extracting a transaction item set from to-be-assessed network data, and generating all possible single item sets based on the transaction item set; calculating the support degree of each single item set, and deleting the item set with the support degree smaller than a preset minimum support degree threshold value to obtain a frequent item set; iteratively performing various combinations on items in the frequent item set to generate a new candidate set and calculate the support degree, and continuously deleting item sets which do not meet a minimum support degree threshold to obtain a maximum frequent item set; mapping the transaction item set into a Boolean matrix by using an improved Apriori algorithm, and performing column item partitioning on the Boolean matrix to obtain a plurality of sub-matrixes; respectively calculating the frequent item set of each sub-matrix, and splicing the frequent item sets of each sub-matrix into the maximum frequent item set; based on the maximum frequent item set, constructing a dual-target optimization risk control model; and solving the dual-target optimization risk control model by adopting an NSGA-II algorithm, and solving an optimal compromise solution by adopting a maximum satisfaction method after a Pareto optimal solution set is obtained. The objective of the invention is to solve the problems of network data security risk analysis and control under the conditions of complex organizations and a large number of systems.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention belongs to the field of data security, and in particular relates to a data security risk assessment and control method and related devices. Background Art

[0002] With the rapid development of the Internet and information technology, network data security has become one of the major challenges faced by enterprises and organizations. In an increasingly complex network environment, security incidents such as data leakage, illegal access and malicious attacks occur frequently, which not only cause huge economic losses to enterprises, but also may damage their reputation and customer trust, thus having an adverse impact on the long-term development of enterprises.

[0003] At present, there are three main analysis methods for network data security risks: qualitative analysis, quantitative analysis and comprehensive analysis. Qualitative analysis mainly relies on the analyst's experience and intuition, determines the value of risk indicators through expert discussions and scoring systems, and outputs risk assessment results based on fixed mathematical models. However, this method is highly subjective and highly dependent on analysts, and it is often difficult to fully reflect the risks faced by the system. The quantitative analysis method is to establish an evaluation system framework, accurately quantify each indicator in a single evaluation, and calculate the risk value through mathematical modeling. Although this method can provide more accurate risk assessment results, it is often difficult to quantitatively evaluate system collections or large-scale information systems, and the reliability of the results is subject to certain limitations. The comprehensive analysis method is a combination of qualitative and quantitative methods. It flexibly selects different combinations of qualitative and quantitative methods according to the actual application situation, so as to obtain risk assessment results suitable for enterprises or systems. This method is more complicated in operation and needs to build a specific analysis method based on actual and historical experience, but its advantage is that it can more comprehensively reflect the risks faced by the system.

[0004] However, the above three methods are mainly applicable to a single system or a smaller system cluster. Their applicability is limited for large enterprises or organizations with complex organizational structures and a large number of systems. Summary of the invention

[0005] In view of the problems existing in the prior art, the present invention provides a data security risk assessment and control method and related devices, which aim to solve the network data security risk analysis and control problems in situations where the organization is complex and the number of systems is large.

[0006] In order to solve the above technical problems, the present invention is implemented by the following technical solutions:

[0007] A data security risk assessment and control method, comprising:

[0008] Extract transaction item sets from the network data to be evaluated, and generate all possible single item sets based on the transaction item sets;

[0009] Calculate the support of each single item set, delete the item sets whose support is less than the preset minimum support threshold, and obtain frequent item sets;

[0010] Iteratively combine the items in the frequent itemsets to generate new candidate sets and calculate the support. By continuously deleting the itemsets that do not meet the minimum support threshold, the maximum frequent itemset is obtained.

[0011] The improved Apriori algorithm is used to map the transaction item set into a Boolean matrix, and the Boolean matrix is ​​divided into columns to obtain multiple sub-matrices.

[0012] Calculate the frequent item sets of each submatrix respectively, and concatenate the frequent item sets of each submatrix into the maximum frequent item set;

[0013] Based on the maximum frequent itemset, a dual-objective optimization risk control model is constructed;

[0014] The NSGA-II algorithm is used to solve the dual-objective optimization risk control model. After obtaining the Pareto optimal solution set, the maximum satisfaction method is used to solve the optimal compromise solution.

[0015] Furthermore, the dual-objective optimization risk control model includes:

[0016] Objective function for lowest data security risk level:

[0017]

[0018] P k =f(x i )

[0019] The objective function for minimizing the cost of data security risks is:

[0020]

[0021] Constraints:

[0022]

[0023] Where: P represents the overall risk status value of network data; k is the total number of risk types; P k represents the k risk state value; n represents the total number of risk sources corresponding to k risk; C k represents the cost of controlling k risk; represents the cost of managing all risk sources of risk k; c i Represents the cost of controlling the ith risk source relative to other risk sources; α(pi ) represents the status value of the i-th risk source before control; m represents the total number of transactions; x i Represents the risk status value of the i-th risk source after control.

[0024] Furthermore, the NSGA-II algorithm is used to solve the dual-objective optimization risk control model, specifically including:

[0025] Initialize the individual crowding distance of the same layer, and the calculation formula is:

[0026] H[i] d =0

[0027] The individuals in the same layer are arranged in ascending order according to the mth objective function value. The calculation formula is:

[0028] H=sort(H,m)

[0029] Given a large number M to avoid individuals on the edge from being abandoned, the calculation formula is:

[0030] H[0] d =H[l] d =M

[0031] The crowding distance of the individual in the middle is calculated using the following formula:

[0032] H[i] d =H[l] d +(H[i+1] m -H[i-1] m )

[0033] Calculate the crowding distance of the bi-objective optimization problem, and the calculation formula is:

[0034]

[0035] For an optimization problem with q objectives, the calculation formula for the crowding distance of individual i is:

[0036]

[0037] Where: i d It is represented as a minimum rectangle that can only contain individual i and no other individuals; l represents the marginal individual; H[i] m represents the mth objective function value of the ith individual; f represents the objective function; f max With f min They represent the maximum and minimum values ​​of the objective function respectively.

[0038] Furthermore, the column-by-column partitioning of the Boolean matrix is ​​specifically as follows:

[0039] The block calculation method is used to divide the Boolean matrix into blocks.

[0040] Furthermore, after obtaining the Pareto optimal solution set, the maximum satisfaction method is used to solve the optimal compromise solution, specifically:

[0041] By calculating the satisfaction of each objective function corresponding to each Pareto solution, the solution with the highest comprehensive satisfaction is selected as the optimal compromise solution.

[0042] A data security risk assessment and control device, comprising:

[0043] An extraction module is used to extract transaction item sets from the network data to be evaluated, and generate all possible single item sets based on the transaction item sets;

[0044] The filtering module is used to calculate the support of each single item set, delete the item sets whose support is less than the preset minimum support threshold, and obtain frequent item sets;

[0045] Iteration module, used to iteratively combine the items in the frequent itemset, generate new candidate sets and calculate the support, and obtain the maximum frequent itemset by continuously deleting the itemsets that do not meet the minimum support threshold;

[0046] A mapping module is used to map the transaction item set into a Boolean matrix using the improved Apriori algorithm, and to divide the Boolean matrix into columns to obtain multiple sub-matrices;

[0047] A concatenation module, used to calculate the frequent item sets of each submatrix respectively, and concatenate the frequent item sets of each submatrix into the maximum frequent item set;

[0048] A construction module, used for constructing a dual-objective optimization risk control model based on the maximum frequent itemset;

[0049] The solution module is used to solve the dual-objective optimization risk control model using the NSGA-II algorithm. After obtaining the Pareto optimal solution set, the maximum satisfaction method is used to solve the optimal compromise solution.

[0050] A device comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of a data security risk assessment and control method when executing the computer program.

[0051] A computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of a data security risk assessment and control method are implemented.

[0052] Compared with the prior art, the present invention has at least the following beneficial effects:

[0053] The present invention provides a data security risk assessment and control method. By improving the Apriori algorithm, the present invention significantly reduces the number of intermediate item sets generated in the data processing process, thereby greatly reducing memory consumption and calculation time. This improvement enables the algorithm to process large-scale data sets more quickly, providing the possibility for real-time risk assessment and control. The application of the NSGA-II algorithm enables the present invention to find the best balance point between multiple objectives and generate a series of Pareto optimal solutions. These solutions provide decision makers with a variety of risk control strategy options, which helps to achieve the security and stability of network data in practical applications. At the same time, multi-objective optimization can also consider the trade-offs between multiple risk factors, making the risk control strategy more comprehensive and reasonable. The risk control strategy generated based on the NSGA-II algorithm can be adjusted and selected according to actual needs to meet different risk control needs. This flexibility enables the present invention to adapt to risk assessment and control needs in different scenarios, and improves the applicability and scalability of the system. Through multi-objective optimization, the present invention can optimize the allocation of resources while meeting risk control needs. For example, when the risk level is high, priority can be given to reducing the risk level; and when the cost budget is limited, the strategy can be adjusted to reduce cost consumption. This optimization helps to achieve reasonable allocation and efficient use of resources. In summary, the present invention realizes efficient assessment and accurate control of network data security risks by combining the improved Apriori algorithm and the NSGA-II algorithm, which not only improves the accuracy and efficiency of risk assessment, but also provides a scientific and reasonable strategy selection for risk control, which is of great significance for ensuring the security and stability of network data.

[0054] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, preferred embodiments are given below and described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] In order to more clearly illustrate the technical solutions in the specific implementation modes of the present invention, the drawings required for use in the description of the specific implementation modes will be briefly introduced below. Obviously, the drawings described below are some implementation modes of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0056] Figure 1 A flowchart of a data security risk assessment and control method according to an embodiment of the present invention;

[0057] Figure 2 This is the specific flow chart of the Apriori algorithm.

[0058] Figure 3 This is the flowchart of the improved Apriori algorithm.

[0059] Figure 4 Schematic diagram of population crowding.

[0060] Figure 5 This is the flow chart of the NSGA-II algorithm.

[0061] Figure 6 This is a schematic diagram of the small-scale satisfaction function curve.

[0062] Figure 7 Comparison of the temporary storage space required for the Apriori algorithm before and after improvement.

[0063] Figure 8 This is a comparison chart of the running time of the Apriori algorithm and the improved Apriori algorithm.

[0064] Fig. 9 The top 10 risk factors and confidence levels for network security in a railway unit.

[0065] Fig.10 The Pareto front result diagram based on different test functions.

[0066] Fig.11 This is the Pareto frontier result diagram of dual-objective risk control based on different algorithms.

[0067] Fig.12 Graph of the Pareto optimal solution and four control schemes. DETAILED DESCRIPTION

[0068] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0069] like Figure 1As shown, the embodiment of the present application provides a data security risk assessment and control method. Aiming at the problem of data security risk assessment and formulation of effective control strategies, the present invention captures the correlation between data and performs corresponding risk assessment by introducing an improved Apriori algorithm, and at the same time adopts a non-dominated sorting genetic algorithm with an elite strategy (NSGA-II algorithm) to control data security risks. In addition, considering the risk state value and control cost, the present invention adopts a Pareto frontier solution with a risk value close to 0.5 to control application system and data security risks. Specifically, the following steps are included:

[0070] S1. Extract transaction item sets from the network data to be evaluated, and generate all possible single item sets based on the transaction item sets.

[0071] S2. Calculate the support of each single item set, delete the item sets whose support is less than the preset minimum support threshold, and obtain frequent item sets (i.e., frequent item sets whose support is not less than the minimum support threshold).

[0072] S3. Iteratively combine the items in the frequent itemset to generate new candidate sets and calculate the support. By continuously deleting the itemsets that do not meet the minimum support threshold, the maximum frequent itemset is obtained.

[0073] Regarding this step, specifically, Figure 2 As shown in the figure, first, the items in the frequent itemset are combined in pairs to generate a candidate set; then, the support of each item set in the candidate set is calculated, and the item sets with support less than the minimum support threshold are deleted to obtain the second frequent item set; then, the candidate items in the second frequent item set are split into element items, and then any three element items are combined to exclude the superset of the deleted candidate items to obtain a new candidate set; then, the support of each item set in the new candidate set is calculated, and the item sets with support less than the minimum support threshold are deleted to obtain the third frequent item set; after repeating a finite number of iterations, if the obtained frequent item set is an empty set, then the previous frequent item set is the required maximum frequent item set.

[0074] S4, such as Figure 3 As shown, the transaction item set is mapped into a Boolean matrix using the improved Apriori algorithm, and the Boolean matrix is ​​divided into columns to obtain multiple sub-matrices.

[0075] In this embodiment, the transaction item set is mapped into a Boolean matrix to facilitate computer processing and storage, while reducing the storage burden of the database.

[0076] Preferably, the initial data is filtered to minimize meaningless initial data and improve algorithm efficiency.

[0077] S5. Calculate the frequent item sets of each sub-matrix respectively, and concatenate the frequent item sets of each sub-matrix into the maximum frequent item set.

[0078] Preferably, a block calculation method is used to divide the Boolean matrix into blocks by column items.

[0079] It should be noted that when the factor item set is large, the binary tree model is used for calculation and the hierarchical structure is reasonably optimized to achieve the fastest convergence speed.

[0080] S6, such as Figure 4 As shown, based on the maximum frequent itemset, a dual-objective optimization risk control model is constructed, which includes two objective functions: one is to minimize the data security risk level, and the other is to minimize the cost consumption of data security risk.

[0081] In other words, based on the Apriori algorithm, a risk function is constructed for the data structure of data security risk assessment. While ensuring that the network data security risk level is effectively controlled, the risk management cost is included as one of the control objectives, and the cost differences required for each risk source are analyzed. Based on this, a dual-objective optimization risk control model is constructed to minimize the risk level and minimize cost consumption.

[0082] Specifically, the dual-objective optimization risk control model includes:

[0083] Objective function for lowest data security risk level:

[0084]

[0085] P k =f(x i )

[0086] The objective function for minimizing the cost of data security risks is:

[0087]

[0088] Constraints:

[0089]

[0090] Where: P represents the overall risk status value of network data; k is the total number of risk types; P k represents the k risk state value; n represents the total number of risk sources corresponding to k risk; C k represents the cost of controlling k risk; represents the cost of managing all risk sources of risk k; c i Represents the cost of controlling the ith risk source relative to other risk sources; α(p i) represents the status value of the i-th risk source before control; m represents the total number of transactions; x i Represents the risk status value of the i-th risk source after control.

[0091] S7, such as Figure 6 As shown in the figure, the NSGA-II algorithm is used to solve the dual-objective optimization risk control model. After obtaining the Pareto optimal solution set, the maximum satisfaction method is used to solve the optimal compromise solution.

[0092] In one embodiment, the NSGA-II algorithm is used to solve the dual-objective optimization risk control model as follows:

[0093] First, the individual crowding distance of the same layer is initialized, and the calculation formula is:

[0094] H[i] d =0

[0095] Next, the individuals in the same layer are arranged in ascending order according to the mth objective function value, and the calculation formula is:

[0096] H=sort(H,m)

[0097] Next, given a large number M to avoid individuals on the edge from being abandoned, the calculation formula is:

[0098] H[0] d =H[l] d =M

[0099] Next, the crowding distance of the individual in the middle is calculated using the following formula:

[0100] H[i] d =H[l] d +(H[i+1] m -H[i-1] m )

[0101] Next, the crowding distance of the bi-objective optimization problem is calculated using the following formula:

[0102]

[0103] Next, for the optimization problem with q objectives, the calculation formula for the crowding distance of individual i is:

[0104]

[0105] Where: i d It is represented as a minimum rectangle that can only contain individual i and no other individuals; l represents the marginal individual; H[i] m represents the mth objective function value of the ith individual; f represents the objective function; fmax With f min They represent the maximum and minimum values ​​of the objective function respectively.

[0106] Finally, repeat the above operations for different objective functions.

[0107] Crowding refers to the density of individuals around a point in a given population, expressed as id. id can be expressed as a minimum rectangle where only individual i can exist but no other individuals can exist, which effectively avoids the error caused by the decision maker specifying the sharing radius in the NSGA algorithm. Figure 5 As shown in the figure, the crowding comparison operator is introduced to ensure that the population maintains diversity and uniform distribution in a wide range, thereby ensuring that the NSGA-II algorithm has good convergence performance on the Pareto surface. In the process of individual selection, first, for two individuals with different i values, the one with a smaller ranking number is selected. Secondly, for two individuals with the same i value, the one with a less crowded surrounding environment is selected. When the above conditions are met, the algorithm is executed and the iterative calculation results are output.

[0108] Preferably, by calculating the satisfaction of each objective function corresponding to each Pareto solution, the solution with the highest comprehensive satisfaction is selected as the optimal compromise solution.

[0109] The calculation formula of the standardized satisfaction of each non-dominated solution is:

[0110]

[0111] The calculation formula of the small fuzzy satisfaction function is:

[0112]

[0113] According to the membership function, the satisfaction of each objective function corresponding to each Pareto solution is calculated. a represents the standardized satisfaction of the ath non-dominated solution, A represents the number of non-inferior solutions, B is the number of objective functions, and f b represents the function value of solving the b-th objective function, f bmax With f bmin They represent the maximum and minimum values ​​of the bth objective function in the solution set respectively.

[0114] The present invention can comprehensively and effectively evaluate and control data security risks by combining the improved Apriori algorithm and the NSGA-II algorithm. At the same time, by constructing a dual-objective optimization risk control model, it can ensure the lowest data security risk level while also taking into account the cost consumption issue, thus achieving the best balance between risk control and cost.

[0115] It should be understood that the frequent itemsets involved in this application refer to the set of items (or item sets) that frequently appear in a data set. Support is the frequency or probability of an item set appearing in a data set. Crowding is an indicator used in the NSGA-II algorithm to measure the density of individuals in a population, which helps to maintain the diversity and uniform distribution of the population. The Pareto optimal solution set is a multi-objective optimization problem in which a solution cannot be found that is superior to other solutions in all objectives. These solutions are called Pareto optimal solutions, and their set is called the Pareto optimal solution set. The maximum satisfaction method is a method for selecting the best solution from multiple possible solutions, in which the satisfaction of different objectives is taken into account.

[0116] In detail, in the present invention, firstly, the improved Apriori algorithm is optimized on the basis of the traditional Apriori algorithm to cope with the challenges encountered in processing large-scale data sets. When processing a large amount of data, the original Apriori algorithm will generate a large number of intermediate item sets, which not only consumes a large amount of memory, but also leads to low algorithm execution efficiency. The present invention effectively reduces the number of intermediate item sets by introducing a series of optimization strategies, such as pruning, candidate item set screening, etc., thereby significantly improving the execution efficiency of the algorithm. At the same time, the improved Apriori algorithm can also maintain stable performance when the number of elements increases, avoiding the performance degradation problem that may occur in traditional algorithms. Secondly, the NSGA-II algorithm is used to achieve multi-objective optimization in network data security risk assessment. In the field of network data security, risk assessment often involves multiple objectives, such as risk level, cost consumption, response time, etc. The NSGA-II algorithm can find the best balance point between multiple objectives through non-dominated sorting and elite retention strategies, and generate a series of Pareto optimal solutions. These solutions provide decision makers with a variety of risk control strategy options, which help to achieve the security and stability of network data in practical applications. In terms of specific implementation, the present invention first uses network data as input, uses the improved Apriori algorithm to mine association rules, and identifies potential security risks. Then, based on the mined association rules, the NSGA-II algorithm is used to perform multi-objective optimization to generate a series of risk control strategies. These strategies can be adjusted and selected according to actual needs to meet different risk control requirements. In short, the technical solution of the present invention achieves efficient assessment and accurate control of network data security risks by combining the improved Apriori algorithm and the NSGA-II algorithm. This solution not only improves the accuracy and efficiency of risk assessment, but also provides a scientific and reasonable strategy selection for risk control, which is of great significance for ensuring the security and stability of network data.

[0117] The present invention is further described below in conjunction with specific implementation cases.

[0118] refer to Figure 7 To verify the effectiveness of the improved Apriori algorithm, the present invention performs comparative analysis in Matlab, setting each item in the intermediate item set to occupy 1 storage space. At the same time, the number of transactions T, the number of elements Co, and the minimum support minSup are changed by the control variable method to compare the space consumption of the Apriori algorithm before and after the improvement. The comparison results of the temporary storage space required by the Apriori algorithm before and after the improvement are shown in Figure 2. Figure 7 As shown. Figure 7 As shown in (a), as the number of transactions increases, the maximum number of intermediate item sets of the Apriori algorithm increases accordingly. When T = 110, the maximum value is as high as 5000. The number of intermediate item sets of the improved Apriori algorithm is stable around 100. Figure 7 From (b) in the figure, we can see that as the number of elements increases, the maximum number of intermediate item sets of the Apriori algorithm increases, while the number of intermediate item sets of the improved Apriori algorithm is stable around 0. Figure 7 From (c), we can see that when the minimum support is 30, the maximum number of intermediate item sets of the Apriori algorithm decreases. When the minimum support of the improved Apriori algorithm is 24, the maximum number of intermediate item sets decreases to 0. This shows that the improved Apriori algorithm consumes less space and is more efficient.

[0119] refer to Figure 8 In this paper, the example database Northwind is selected as the data source under the same environment, and the C# programming language is used to conduct empirical analysis and comparative tests on the Apriori algorithm and the improved Apriori algorithm. The comparison results of the running time of the two algorithms are shown in Figure 8 As shown. Figure 8 It can be seen that the running time of each algorithm increases with the increase in the number of data records. At the same time, it can be seen that the running time of the improved Apriori algorithm is lower than that of the Apriori algorithm, and its running time growth rate is lower than that of the Apriori algorithm. Among them, when the number of data records is 500, the running time of the improved Apriori algorithm is about 6×104ms, while the running time of the Apriori algorithm is as high as 8×104ms. This shows that the computational efficiency of the improved Apriori algorithm is higher.

[0120] refer to Fig. 9, the present invention further uses 1800 sets of security assessment data of a certain railway unit in my country in 2021-2022 for risk analysis. After data preprocessing, a basic data set is constructed. The study sets the minimum support to 20% of the total data, that is, min Sup = 360, and sets the minimum confidence to min Con = 50%. The top 10 risk factors and confidence levels of network security of a railway unit are as follows Fig. 9 As shown. Fig. 9 It can be seen that the probability of occurrence of the main risk factors faced by the unit's network security is more than 0.7, and the confidence level of the association between each factor and medium and high risks is around 75%. The calculation results are basically consistent with the "main network security risks" mentioned in the unit's 2021 network security work summary, thus verifying the accuracy and reliability of the calculation results.

[0121] refer to Fig.10 To verify the effectiveness of the NSGA-II algorithm, the present invention first performs a performance test on it. The experiment applies it to the DTLZ1 and DTLZ2 test functions, and the Pareto frontier results obtained are as follows: Fig.10 As shown. Fig.10 From (a) in , we can see that the NSGA-II algorithm has achieved a relatively uniform and well-converged Pareto frontier in the DTLZ1 test function. Fig.10 From (b) in the figure, we can see that the Pareto front distribution obtained by the NSGA-II algorithm in the DTLZ2 test function is also relatively uniform, indicating that it has good stability and convergence performance.

[0122] refer to Fig.11 , the present invention continues to use the Multi-Objective Evolutionary Algorithm based on Decomposition (MOEA / D) algorithm for comparison. In order to better compare the Pareto frontier of dual-objective risk control, the application system and data security of a railway unit are taken as an example to normalize the risk level and cost consumption function value. The Pareto frontier results of dual-objective risk control solved by the two algorithms are shown in the figure. Fig.11 As shown. Fig.11 From (a) in the figure, we can see that the Pareto optimal solution set obtained by the MOEA / D algorithm is unevenly distributed. At the same time, we can see that the values ​​of the dual objective functions obtained by the algorithm are higher than those of the NSGA-II algorithm. Fig.11From (b) in the figure, we can see that the Pareto optimal solution set obtained by the NSGA-II algorithm has good diversity and uniform distribution. This shows that the algorithm has a strong global search capability when processing the risk control model constructed by the research institute, and can search in a wide solution space and obtain a non-dominated solution set with good diversity. In addition, from Fig.11 It can also be seen that all non-dominated solutions form a curve that is approximately an inverse function, that is, the two objective function values ​​are approximately inversely proportional. This shows that the optimization of one objective function will inevitably lead to the degradation of another objective function.

[0123] refer to Fig.12 , the present invention further traverses the network data security risk assessment results, and finds the risk analysis indicators corresponding to the application system and data security of a certain railway unit, including E1 application system security configuration, E2 monitoring and defense, E3 application system identity authentication, and E4 important data security assurance. The experiment runs MATLAB and obtains a set of Pareto optimal solutions. Then, four Pareto frontiers with risk values ​​close to 0.3, 0.4, 0.5 and 0.6 are selected from the figure, and four application system and data security risk control solutions are obtained. The results of the Pareto optimal solution and the four control solutions are shown in the figure. Fig.12 As shown. Fig.12 From (a) in the figure, we can see that the relationship between risk level and cost consumption is an inverse correlation. Fig.12 As can be seen from (b), as the risk level decreases, the cost consumption increases. Taking into account the risk status value and control cost, the study adopts the third solution to control the application system and data security risks. In this solution, the risk level is only 0.386, and the cost consumption is only 240,480 yuan. This shows that the data security risk assessment and control strategy proposed in the present invention is highly feasible, which effectively improves the use value of data and brings practical value and benefits to enterprises or organizations.

[0124] In one embodiment, a data security risk assessment and control device is also provided, including:

[0125] The extraction module is used to extract transaction item sets from the network data to be evaluated and generate all possible single item sets based on the transaction item sets.

[0126] The filtering module is used to calculate the support of each single item set, delete the item sets whose support is less than the preset minimum support threshold, and obtain frequent item sets.

[0127] The iteration module is used to iteratively combine the items in the frequent itemset in various ways, generate new candidate sets and calculate the support, and obtain the maximum frequent itemset by continuously deleting the itemsets that do not meet the minimum support threshold.

[0128] The mapping module is used to map the transaction item set into a Boolean matrix using the improved Apriori algorithm, and to divide the Boolean matrix into columns to obtain multiple sub-matrices.

[0129] The concatenation module is used to respectively calculate the frequent item sets of each sub-matrix, and concatenate the frequent item sets of each sub-matrix into the maximum frequent item set.

[0130] A construction module is used to construct a dual-objective optimization risk control model based on the maximum frequent itemset.

[0131] The solution module is used to solve the dual-objective optimization risk control model using the NSGA-II algorithm. After obtaining the Pareto optimal solution set, the maximum satisfaction method is used to solve the optimal compromise solution.

[0132] Among them, regarding the specific implementation methods of the above-mentioned modules, reference can be made to the corresponding contents disclosed in the aforementioned embodiments, which will not be repeated here.

[0133] In one embodiment of the present invention, a computer device is provided, the computer device comprising a processor and a memory, the memory being used to store a computer program, the computer program comprising program instructions, and the processor being used to execute the program instructions stored in the computer storage medium. The processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc., which are the computing core and control core of the terminal, which are suitable for implementing one or more instructions, and are specifically suitable for loading and executing one or more instructions to implement the corresponding method flow or corresponding functions; the processor described in the embodiment of the present invention can be used to implement the operation of a data security risk assessment and control method.

[0134] In one embodiment of the present invention, a data security risk assessment and control method can be stored in a computer-readable storage medium if it is implemented in the form of a software functional unit and sold or used as an independent product. Based on this understanding, the present invention implements all or part of the process in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. Computer-readable storage media include permanent and non-permanent, removable and non-removable media, and information storage can be implemented by any method or technology. Information can be computer-readable instructions, data structures, program modules or other data.

[0135] The computer storage medium can be any available medium or data storage device that can be accessed by a computer, including but not limited to magnetic storage (such as floppy disks, hard disks, magnetic tapes, magneto-optical disks (MO), etc.), optical storage (such as CD, DVD, BD, HVD, etc.), and semiconductor storage (such as ROM, EPROM, EEPROM, non-volatile memory (NANDFLASH), solid-state drive (SSD)), etc.

[0136] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that include computer-usable program code.

[0137] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products of the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0138] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0139] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process in the computer or other programmable device. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0140] In the description of the present invention, the terms "one embodiment", "some embodiments", "examples", "specific examples", or "some examples" etc. mean that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, without contradiction.

[0141] In the description of the present invention, it is necessary to understand that the terms "upper", "lower", "bottom", "inside", "outside", etc. indicate orientations or positional relationships based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the present invention.

[0142] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features. In the description of the present invention, the meaning of "plurality" is at least two, such as two, three, etc., unless otherwise clearly and specifically defined.

[0143] In the present invention, unless otherwise clearly specified and limited, the terms "connected", "connection" and the like should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection, an electrical connection, or communication with each other; it can be a direct connection, or an indirect connection through an intermediate medium, it can be the internal connection of two elements or the interaction relationship between two elements, unless otherwise clearly defined. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0144] In the present invention, unless otherwise clearly specified and limited, a first feature being "above" or "below" a second feature may mean that the first and second features are in direct contact, or the first and second features are in indirect contact through an intermediate medium. Moreover, a first feature being "above", "above" or "above" a second feature may mean that the first feature is directly above or obliquely above the second feature, or simply means that the first feature is higher in level than the second feature. A first feature being "below", "below" or "below" a second feature may mean that the first feature is directly below or obliquely below the second feature, or simply means that the first feature is lower in level than the second feature.

[0145] In the present invention, the terms "one embodiment", "some embodiments", "examples", "specific examples", or "some examples" etc. mean that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the described specific features, structures, materials or characteristics may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, without contradiction.

[0146] Finally, it should be noted that the above-described embodiments are only specific implementations of the present invention, which are used to illustrate the technical solutions of the present invention, rather than to limit them. The protection scope of the present invention is not limited thereto. Although the present invention is described in detail with reference to the above-described embodiments, ordinary technicians in the field should understand that any technician familiar with the technical field can still modify the technical solutions recorded in the above-described embodiments within the technical scope disclosed by the present invention, or can easily think of changes, or make equivalent replacements for some of the technical features therein; and these modifications, changes or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention shall be based on the protection scope of the claims.

Claims

1. A data security risk assessment and control method, characterized in that: include: Extract transaction item sets from the network data to be evaluated, and generate all possible single item sets based on the transaction item sets; Calculate the support of each single item set, delete the item sets whose support is less than the preset minimum support threshold, and obtain frequent item sets; Iteratively combine the items in the frequent itemsets to generate new candidate sets and calculate the support. By continuously deleting the itemsets that do not meet the minimum support threshold, the maximum frequent itemset is obtained. The improved Apriori algorithm is used to map the transaction item set into a Boolean matrix, and the Boolean matrix is ​​divided into columns to obtain multiple sub-matrices. Calculate the frequent item sets of each submatrix respectively, and concatenate the frequent item sets of each submatrix into the maximum frequent item set; Based on the maximum frequent itemset, a dual-objective optimization risk control model is constructed; The NSGA-II algorithm is used to solve the dual-objective optimization risk control model. After obtaining the Pareto optimal solution set, the maximum satisfaction method is used to solve the optimal compromise solution.

2. A data security risk assessment and control method according to claim 1, characterized in that: The dual-objective optimization risk control model includes: Objective function for lowest data security risk level: P k =f(x i ) The objective function for minimizing the cost of data security risks is: Constraints: Where: P represents the overall risk status value of network data; k is the total number of risk types; P k represents the k risk state value; n represents the total number of risk sources corresponding to k risk; C k represents the cost of controlling k risk; represents the cost of managing all risk sources of risk k; c i Represents the cost of controlling the ith risk source relative to other risk sources; α(p i ) represents the status value of the i-th risk source before control; m represents the total number of transactions; x i Represents the risk status value of the i-th risk source after control.

3. A data security risk assessment and control method according to claim 1, characterized in that: The NSGA-II algorithm is used to solve the dual-objective optimization risk control model, specifically including: Initialize the individual crowding distance of the same layer, and the calculation formula is: H[i] d =0 The individuals in the same layer are arranged in ascending order according to the mth objective function value. The calculation formula is: H=sort(H,m) Given a large number M to avoid individuals on the edge from being abandoned, the calculation formula is: H[0] d =H[l] d =M The crowding distance of the individual in the middle is calculated using the following formula: H[i] d =H[l] d +(H[i+1] m -H[i-1] m ) Calculate the crowding distance of the bi-objective optimization problem, and the calculation formula is: For an optimization problem with q objectives, the calculation formula for the crowding distance of individual i is: Where: i d It is represented as a minimum rectangle that can only contain individual i and no other individuals; l represents the marginal individual; H[i] m represents the mth objective function value of the ith individual; f represents the objective function; f max With f min They represent the maximum and minimum values ​​of the objective function respectively.

4. A data security risk assessment and control method according to claim 1, characterized in that: The column-by-column partitioning of the Boolean matrix is ​​specifically as follows: The block calculation method is used to divide the Boolean matrix into blocks.

5. A data security risk assessment and control method according to claim 1, characterized in that: After obtaining the Pareto optimal solution set, the maximum satisfaction method is used to solve the optimal compromise solution, specifically: By calculating the satisfaction of each objective function corresponding to each Pareto solution, the solution with the highest comprehensive satisfaction is selected as the optimal compromise solution.

6. A data security risk assessment and control device, characterized in that: include: An extraction module is used to extract transaction item sets from the network data to be evaluated, and generate all possible single item sets based on the transaction item sets; The filtering module is used to calculate the support of each single item set, delete the item sets whose support is less than the preset minimum support threshold, and obtain frequent item sets; Iteration module, used to iteratively combine the items in the frequent itemset, generate new candidate sets and calculate the support, and obtain the maximum frequent itemset by continuously deleting the itemsets that do not meet the minimum support threshold; A mapping module is used to map the transaction item set into a Boolean matrix using the improved Apriori algorithm, and to divide the Boolean matrix into columns to obtain multiple sub-matrices; A concatenation module, used to calculate the frequent item sets of each submatrix respectively, and concatenate the frequent item sets of each submatrix into the maximum frequent item set; A construction module, used for constructing a dual-objective optimization risk control model based on the maximum frequent itemset; The solution module is used to solve the dual-objective optimization risk control model using the NSGA-II algorithm. After obtaining the Pareto optimal solution set, the maximum satisfaction method is used to solve the optimal compromise solution.

7. A device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of a data security risk assessment and control method as described in any one of claims 1 to 5 are implemented.

8. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of a data security risk assessment and control method as described in any one of claims 1 to 5 are implemented.