Fine-grained access control method and system based on risk identification

Abstract

The invention discloses a fine-grained access control method and system based on risk identification, and belongs to the technical field of information security. According to the method, user subject attributes, behavior attributes and system environment attribute information are collected in real time, a standardized decision matrix is constructed, an interval type-2 fuzzy set (IT2FS) is used for conducting fuzzy modeling on the attributes, and an upper membership matrix and a lower membership matrix are generated. And calculating the dynamic weight of the attribute index in combination with a CRITIC method, introducing a time decay factor to dynamically correct a risk score through an improved TOPSIS method, calculating the Euclidean distance between an access request and a positive / negative ideal solution, and generating a normalized risk closeness degree. And based on the risk score and a preset threshold value, dynamically matching a hierarchical permission strategy, and adopting a static rule and a priority coverage mechanism to eliminate permission conflicts. According to the method, multi-dimensional risk assessment and dynamic weight adjustment are fused, the problems of insufficient real-time performance, subjective weight dependence and weak uncertainty processing capability in a traditional method are solved, the accuracy and security of access control are remarkably improved, and the method is suitable for scenes with high security requirements such as cloud computing and finance.

Description

Technical Field

[0001] The present invention relates to the technical field of system access control, and in particular to a fine-grained access control method and system based on risk identification. Background Art

[0002] Access control is one of the key technologies in the field of information security and is widely used to protect sensitive data and computing services. With the development of information technology, especially the popularization of big data, the protection of information assets has become particularly important. Therefore, access control technology has assumed the important responsibility of protecting information and resource security. At present, traditional access control methods (such as RBAC and ABAC) have certain limitations in dealing with dynamic and complex access scenarios. Although risk-based access control methods have been gradually proposed, most risk-based access control methods often rely on subjective weight calculations (such as AHP) or ignore the correlation between different risk indicators (such as entropy weight method), and perform poorly in the face of uncertainty and real-time decision-making, and ignore dynamic factors such as user behavior and environmental changes. Dynamic access control cannot respond to abnormal changes in the system in a timely manner. Therefore, the present invention proposes an access control method based on risk assessment, which performs real-time assessment according to multiple factors such as user behavior characteristics, subject attributes, and environmental status, and dynamically adjusts access rights based on the assessment results, which becomes an effective means to improve system security.

[0003] The present invention realizes fine-grained risk assessment of access control by combining interval type-2 fuzzy sets, CRITIC weight calculation and TOPSIS evaluation method. Through comprehensive analysis of user subject, behavior and environmental attributes, the present invention can more accurately assess risks and adjust access rights according to real-time changes, significantly improving the real-time performance and accuracy of access control.

[0004] The present invention provides a fine-grained access control method based on risk assessment, which is applied to a fine-grained access control system based on risk assessment. The method comprises: Step 1: When a user initiates an access request, collect user subject attributes, behavior attributes, and system environment attribute information in real time and define their types (benefit type / cost type), including: Attribute Name Type Belonging Dimension Selection Basis Role Permission Level Benefit Type Subject Dimension The higher the role permission level, the larger the allowed access range. It is a positive indicator. Range normalization eliminates dimensional differences and retains the directionality Risk Historical Decay Value Cost Type Subject Dimension The greater the historical risk value, the higher the current risk. Reverse normalization is required. By taking the reciprocal transformation, cost-type attributes are unified for positive comparison Time Period - Operation Deviation Degree Cost Type Behavior Dimension The greater the deviation degree, the more abnormal the operation. The cost-type formula is used to map high deviation values to low scores, reflecting the risk growth trend Geographical Movement Abnormality Index Cost Type Behavior Dimension The greater the geographical location jump, the higher the risk. After normalization, the index directionality is retained to ensure that high abnormal values correspond to low normalization results API Call Compliance Benefit Type Behavior Dimension The higher the compliance score, the lower the risk. Direct range normalization is used to retain the numerical characteristics of positive indicators Abnormal Request Ratio Cost Type Behavior Dimension The higher the abnormal request ratio, the greater the risk. By taking the reciprocal transformation, high ratio values are mapped to low scores, which conforms to the definition of cost-type attributes Network Fluctuation Entropy Value Cost Type Environment Dimension The greater the entropy value, the more unstable the network environment. After reverse normalization, high entropy values correspond to low scores to ensure consistency with other cost-type attribute logics The collected data is standardized: the subject attributes are converted into numerical values through mapping functions or normalization methods; the behavior attributes are quantified through statistical calculations; the system environment attributes are calculated based on historical data and real-time monitoring; through data preprocessing, the user attribute data is converted into a computable numerical form, and the risk attribute set is collected , quantify various risk indicators.

[0005] Step 2: Membership degree calculation based on interval type-2 fuzzy sets. Based on the standardized decision matrix, use the interval type-2 fuzzy membership function to perform fuzzy modeling on the attribute indicators, obtain the upper and lower membership degrees of each attribute, and form a fuzzy decision matrix; 1. Selection of membership function: Gaussian type is used to describe continuous attributes including time period - operation deviation degree and network fluctuation entropy value; trapezoidal type is applicable to indicators with obvious intervals including role permission level, geographical movement anomaly index, and abnormal request ratio; triangular type is suitable for symmetrically distributed fuzzy information including risk history decay value and API call compliance.

[0006] 2. Calculate the upper and lower membership degrees of the indicators. Upper membership degree: used to describe the loosest risk estimate, considering uncertainties such as measurement errors and system fluctuations; lower membership degree: used to describe the most conservative risk estimate, set based on the statistical analysis of historical data.

[0007] 3. Calculate the interval mean 。

[0008] 4. Evaluate each attribute using the interval type-2 model to generate a fuzzy decision matrix \(\mathcal{D}=\left [ {{\mathcal{X}}_{ij}} \right ]_{m\times 7} , {\chi}_{ij=\left [ {{\mu}^{l}_{ij},{\mu}^{u}_{ij}} \right ]} 。

[0009] Step 3: Calculate the index weights using the CRITIC method. Based on the fuzzy decision matrix, use the CRITIC method to calculate the standard deviation and correlation of each attribute, determine the weights of each attribute indicator, and perform normalization processing on the weights; 1. Calculate the standard deviation of each indicator to measure the variability of the indicator: Benefit-type attributes ( , ) are calculated according to the formula ; Cost-type attributes ( , , , , ) are calculated according to the formula , obtaining the standardized matrix \(R=\left [ {{r}_{ij}} \right ]_{m\times 7} , Use the standard deviation formula: , where , calculate the standard deviation of each metric to measure its variability.

[0010] 2. Calculate the correlation between metrics: Calculate the correlation coefficient between metrics : Use the conflict formula: , and then calculate the information volume .

[0011] 3. Calculate the weights based on the information volume. The weight calculation formula is .

[0012] Step 4: Dynamic TOPSIS risk assessment: Combine the membership values and weights of each attribute, and use the improved TOPSIS method to calculate the risk closeness of access requests. Further, combine the time decay factor to dynamically correct the risk score, and then calculate the risk progress, generate the risk score and map it to the [0, 1] interval; Introduce the time decay factor and model it with an exponential decay function, , is the time decay coefficient, t is the time interval, and the least squares method is used for optimization in this method .

[0013] Correct the weighted value of each metric .

[0014] Construct the weighted fuzzy decision matrix \(\mathcal{V}=[{w}_{j·{\mathcal{X}}_{ij}}]_{m×7}\) ; Dynamically determine the positive and negative ideal solutions: The positive ideal solution \(\mathcal{V}^{+}=([{max}_{i}{\mu}^{u}_{i1},{max}_{i}{\mu}^{l}_{i1}],...,[{max}_{i}{\mu}^{u}_{i7},{max}_{i}{\mu}^{l}_{i7}])\) , the negative ideal solution \(\mathcal{V}^{-}=([{min}_{i}{\mu}^{l}_{i1},{min}_{i}{\mu}^{u}_{i1}],...,[{min}_{i}{\mu}^{l}_{i7},{min}_{i}{\mu}^{u}_{i7}])\) .

[0015] Calculate the interval distances of each request to the positive and negative ideal solutions and where: , : Request i The lower and upper bounds of the membership degree of the attribute j ; , : Attribute j The lower and upper bounds of the positive ideal solution of the attribute (taking the maximum value of all requests); : Attribute j The lower and upper bounds of the negative ideal solution of the attribute (taking the minimum value of all requests); and then calculate the risk closeness degree as the risk score.

[0016] Step Five: Compare the access evaluation result with the preset threshold, dynamically adjust the access control decision, and perform operations of allowing access, restricting access, or denying access: After the user initiates an access request to a specific object resource and its risk is evaluated, start the subsequent access control process: Dynamic permission adjustment, and the access permission adjustment module authorizes dynamically according to the evaluation result: Hierarchical access control decision: According to the risk closeness degree Sort, set the threshold , : If the risk closeness degree is 1, prohibit access; If the risk closeness degree is higher than or equal to , dynamically downgrade the permission to read-only low-sensitivity data permission; If the risk closeness degree is lower than , allow access and verify the basic permission; If the risk closeness degree is between the two, trigger two-factor verification and adjust the permission according to the resource sensitivity: downgrade the access permission of the current session in real time, and cross-verify the identity of the high-risk subject through a combination of dynamic token verification and password verification.

[0017] The present invention provides a fine-grained access control system based on risk assessment, including: an attribute collection and preprocessing module, an uncertainty modeling module, a comprehensive risk assessment module, and a dynamic access control decision module, wherein: The attribute collection and preprocessing module is used to collect access subject attributes, behavior attributes, and system environment attributes, including user role permission levels, risk history attenuation values, time period-operation deviation degrees, geographical movement anomaly indexes, API call compliance, abnormal request ratios, and network fluctuation entropy values. The collected data is processed through standardization to unify the dimension and numerical range, ensuring the consistency and comparability of data input.

[0018] An uncertainty modeling module, which is used to model based on the collected standardized attribute information by using the interval type-2 fuzzy set theory, calculate the upper and lower membership degrees of each attribute index to express the uncertainty in the attribute values, form a fuzzy decision matrix, and provide a fuzzy decision basis for comprehensive risk assessment.

[0019] The comprehensive risk assessment module integrates the fuzzy decision matrix of the user's multi-dimensional attributes and the dynamic weight coefficients. Based on the fuzzy decision matrix and the normalized weight set, it uses the improved TOPSIS method to comprehensively evaluate the user's risk, and introduces a time decay factor to dynamically correct the risk value. The dynamic weight is adjusted in real time according to the actual access environment and the change of the user's behavior pattern. The evaluation process includes membership degree aggregation, risk membership degree calculation and risk score normalization, and finally outputs the standardized risk closeness to guide the access control decision.

[0020] The dynamic access control decision module compares and analyzes the real-time risk score generated by the comprehensive risk assessment module with the preset risk threshold in the system, and dynamically adjusts the user's access rights. If the risk assessment result is lower than the security threshold, normal access is allowed; if the risk assessment result is in the intermediate sensitive interval, secondary verification measures (SMS verification code, dynamic identity verification) are triggered; if the risk assessment result is higher than the blocking threshold, the access request is directly blocked. This module supports intelligent resolution of permission conflicts and conflict detection, ensuring the flexibility and security of access control, and realizing accurate fine-grained access control based on risk adaptation.

[0021] The system records the processing process of access requests, including information such as the start time, end time, risk assessment result, and access control decision of the access request. All audit data will be used for subsequent security monitoring and review for later analysis and improvement. Brief Description of the Drawings

[0022] The specification drawings forming a part of this application are used to provide a further understanding of this application. The schematic embodiments of this application and their descriptions are used to explain this application and do not constitute an improper limitation of this application.

[0023] Figure 1 : Flowchart of the fine-grained access control method based on risk identification: It shows the core process of the present invention, from multi-dimensional attribute collection to dynamic permission decision-making, covering the weight calculation, dynamic score generation and threshold mapping links in risk assessment, reflecting the closed-loop logic of risk assessment and access control.

[0024] Figure 2 : System module structure diagram: It is the system architecture of the present invention, clarifying the functions of each module and the data interaction relationship, and supporting the real-time performance and reliability of fine-grained access control. Detailed Embodiments

[0025] The present invention will be further described below in conjunction with the accompanying drawings and embodiments. The following detailed descriptions are all illustrative and are intended to provide further explanations for the present application. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by those of ordinary skill in the technical field to which the present application belongs.

[0026] Compared with the prior art, the present invention has the following advantages: improving real-time performance: using CRITIC to calculate weights, avoiding the subjectivity of AHP calculation, and improving the degree of automation; low computational complexity, supporting real-time access control decisions; enhancing the ability to handle uncertainty: interval type-2 fuzzy sets can effectively model fuzziness, improving the accuracy and robustness of evaluation; improving the accuracy of risk assessment: combining CRITIC and TOPSIS methods to improve the rationality of risk ranking; applicable to a variety of complex access control scenarios, such as cloud computing and multi-level privilege management.

[0027] Risk assessment of user login requests Data collection: and normalize the attribute user role permission level, risk history decay value, time period-operation deviation degree, geographical movement anomaly index, API call compliance, abnormal request ratio, network fluctuation entropy value according to the method specified in the present invention: User A: [0.8 0.3 0.5 0.2 0.9 0.1 0.4] User B: [0.6 0.5 0.6 0.3 0.7 0.2 0.5] User C: [0.9 0.2 0.4 0.1 0.95 0.05 0.3].

[0028] . Standard deviation of each attribute: role permission level 0.1, risk history decay value 0.15, time period-operation deviation degree 0.12, geographical movement anomaly index 0.08, API call compliance 0.05, abnormal request ratio 0.1, network fluctuation entropy value 0.07.

[0029] In this embodiment, select the corresponding membership function according to the requirements of the invention, calculate the upper and lower membership degrees of each attribute, use the normalized attribute values of users A, B, and C for membership degree calculation, calculate the upper and lower membership degrees of each attribute, and construct a weighted fuzzy decision matrix according to the attribute values of users A, B, and C: According to the calculated membership degrees, the following fuzzy decision matrix is obtained: 0.333 0.4 1 0.333 1 1 0.606 0.5 1 0.606 0.667 0.2 0.5 1 1 0.6 0.606 0 1 0 0.606

[0030] Calculate the correlation matrix between attributes 1 0.2 0.1 0.3 0.4 0.3 0.2 0.2 1 0.5 0.1 0.3 0.2 0.1 0.1 0.5 1 0.4 0.5 0.4 0.3 0.3 0.1 0.4 1 0.6 0.5 0.4 0.4 0.3 0.5 0.6 1 0.7 0.5 0.3 0.2 0.4 0.5 0.7 1 0.6 0.2 0.1 0.3 0.4 0.5 0.6 1 Calculate the correlation coefficient between each indicator according to the calculation formula specified in this method, and calculate the information amount of each attribute, obtaining: Role permission level 0.15, Risk history decay value 0.12, Period-operation deviation degree 0.13, Geographic movement anomaly index 0.11, API call compliance 0.14, Abnormal request ratio 0.12, Network fluctuation entropy value 0.10.

[0031] Perform normalization to obtain the final weights: Example calculated weights: Role permission level 0.20, Risk history decay value 0.15, Period-operation deviation degree 0.18, Geographic movement anomaly index 0.14, API call compliance 0.16, Abnormal request ratio 0.12, Network fluctuation entropy value 0.05.

[0032] Set the time decay coefficient λ = 0.1 and the time interval t = 1, and correct the weighted value of each attribute. The weights after decay correction are: Role permission level 0.19, Risk history decay value 0.14, Period-operation deviation degree 0.17, Geographic movement anomaly index 0.13, API call compliance 0.15, Abnormal request ratio 0.11, Network fluctuation entropy value 0.05.

[0033] Construct a weighted fuzzy decision matrix: Use the corrected weights to weight the standardized matrix to construct a weighted fuzzy decision matrix: 0.152 0.114 0.171 0.042 0.071 0.028 0.085 0.102 0.068 0.026 0.042 0.013 0.135 0.105 0.143 0.011 0.022 0.006 0.005 0.022 0.015 .

[0034] Calculate the distance between each user and the positive and negative ideal solutions, and calculate the risk closeness through the distance: User A 0.75, User B 0.85, User C 0.65.

[0035] Trigger an authorization decision and make an access control decision based on the risk closeness value of each user. The thresholds set here are: = 0.8, = 0.5; The final decision is as follows: The risk closeness of User A is 0.75, which is between 0.5 and 0.8. Trigger a risk negotiation decision: The system enters the risk negotiation stage and adjusts the access permissions according to the resource sensitivity; Resource sensitivity assessment: The resource that User A requests to access is a financial data resource, which belongs to a high-sensitivity resource; Real-time downgrade of permissions: Due to the relatively high risk closeness, the system downgrades User A's access permission from full access permission to read-only mode until further verification is completed; Trigger secondary verification: The system requires User A to authenticate through dynamic token verification and password verification. If the verification is successful, User A's full access permission will be restored; if the verification fails, the access permission will continue to be restricted.

[0036] The risk proximity of User B is 0.85, which is higher than 0.8 but not equal to 1. Although the risk proximity is relatively high, since the accessed data resource is of low sensitivity, the system still downgrades the permission to read-only mode and requires secondary verification.

[0037] The risk proximity of User C is 0.65, which is between 0.5 and 0.8; triggering a risk negotiation decision: the system enters the risk negotiation stage and adjusts the access permission according to the resource sensitivity; resource sensitivity assessment: the resource that User C requests to access is an ordinary report document, belonging to a low-sensitivity resource; real-time permission downgrade: due to the relatively low risk proximity, the system decides not to downgrade the permission and still allows User C to access the resource; secondary verification trigger: even though the risk proximity is relatively low, the system still requires User C to perform secondary verification; according to the setting, User C must pass the dynamic token verification and can continue to access the resource after passing the verification.

[0038] The present invention provides a fine-grained access control method based on risk assessment. By comprehensively analyzing information such as user attributes, behaviors, and environments, combining interval type-2 fuzzy sets (IT2FS), CRITIC weight calculation, and TOPSIS evaluation method, the risk proximity of users is dynamically evaluated, and the access permission is adjusted in real time according to the evaluation results. By introducing a time decay factor, a weighted fuzzy decision matrix, and the calculation of the distances from positive and negative ideal solutions, the present invention can effectively and precisely control user access behaviors, enhancing the security and flexibility of the system.

[0039] The present invention can be applied to various scenarios requiring access control and authentication, and is particularly suitable for application environments with strict security requirements for resource access in fields such as finance, healthcare, and government. With the continuous development of information security technology, the method of the present invention can be further optimized and adjusted to adapt to new security challenges and requirements.

[0040] Although the present invention has been described in detail through specific embodiments, for those skilled in the art, the specific implementation of the present invention can be modified and changed according to actual needs, and these modifications and changes should not exceed the protection scope of the present invention. It is hoped that the present invention can provide new ideas and technical solutions for achieving more secure and efficient access control.

Claims

1. A fine-grained access control method and system based on risk identification, characterized in that It includes the following steps: Step S1: Real-time collect the information of the main attributes, behavioral attributes, and system environment attributes of the access request user, and perform standardization processing to generate a standardized decision matrix; the attributes include: Main body dimension: Role permission level (benefit type), Risk history decay value (cost type); Behavioral dimension: Time period - operation deviation degree (cost type), Geographic movement anomaly index (cost type), API call compliance (benefit type), Abnormal request ratio (cost type); Environment dimension: Network fluctuation entropy value (cost type); Step S1.1: Construct the original data matrix of the access request, , where represents the observation of the i-th request on the j-th attribute; Step S1.2: Define the set of risk attributes Including: Subject dimension: Role permission level , Risk historical decay value ; Behavior dimension: Time period - Operation deviation degree , Geographic movement anomaly index , API call compliance , Abnormal request ratio ; Environment dimension: Network fluctuation entropy value ; Step S2: Based on the interval type-2 fuzzy set (IT2FS), perform fuzzy modeling on the standardized attributes, calculate the upper and lower membership degrees of each attribute, and generate a fuzzy decision matrix; among them: For the role permission level and the geographic movement anomaly index, use the trapezoidal membership function; for the network fluctuation entropy value and the time period - operation deviation degree, use the Gaussian membership function; for the risk history decay value and the API call compliance, use the triangular membership function; Step S2.1: Perform interval type-2 model evaluation on each attribute to generate a fuzzy matrix , ; Step S3: Use the CRITIC method to calculate the weights of each attribute index, including: Calculate the standard deviation and correlation coefficient of the index, determine the attribute conflict and comprehensive information volume; perform normalization processing on the information volume to generate a weight set; Step S3.1: Construct a weighted fuzzy decision matrix ; Step S4: Calculate the risk closeness of the access request based on the improved TOPSIS method, including: introducing a time decay factor dynamically correct the weighted fuzzy decision matrix; calculate the Euclidean distances from the positive / negative ideal solutions, generate a risk score and map it to the interval [0, 1]; Step S4.1: Dynamically determine the positive and negative ideal solutions; Step S4.2: Calculate the interval distances from each request to the positive and negative ideal solutions and ; Step S4.3: Optimize the risk proximity degree through a time decay factor , where is the time interval fitted from historical data ; Step S5: Hierarchical access, dynamic permission mapping, and match the permission template from the preset fuzzy rule library; Step S6: Permission conflict resolution: Based on the operation dependency relationship and the risk score change history, eliminate or override the conflicting permissions.

2. The fine-grained access control method based on risk identification according to claim 1, wherein The standardization processing in the above step 1 includes: Standardize the fuzzy matrix according to the attribute type: for benefit attributes ( , ), convert according to the formula ; for cost attributes ( , , , , ), convert according to the formula . Perform range normalization on the original data matrix to obtain a normalized matrix .

3. A fine-grained access control method based on risk identification according to claim 1, characterized in that The specific CRITIC weight calculation in the above step 3 is: Calculate the standard deviation of each attribute: ; Among calculating the correlation coefficients between the metrics : ; Conflict with correlation coefficient: ; Determine weights based on information volume .

4. A fine-grained access control method based on risk identification according to claim 1, characterized in that, The improved TOPSIS method includes that the time decay factor in the step 4 is , including: Calculate the positive ideal solution as the maximum membership degree interval of each attribute , and the negative ideal solution as the minimum membership degree interval of each attribute ; Calculate the interval distances of each request to the positive and negative ideal solutions ; ; is the time decay coefficient, and t is the time interval. The least squares method is used for optimization in this method ; Revised weight: ; Optimize risk proximity: .

5. The fine-grained access control method based on risk identification according to claim 1, wherein The hierarchical access control decision in the above step five includes the following operations: According to the risk proximity Sort and set a threshold , : If the proximity is 1, access is prohibited; If the risk proximity is higher than or equal to , the dynamic permission is downgraded to read-only mode; If the risk proximity is lower than , access is allowed and the basic permissions are verified; If the risk proximity is between the two, trigger secondary verification and adjust the permissions according to the resource sensitivity: Real-time downgrade the access permissions of the current session, and cross-verify the identity of high-risk subjects through a combination of dynamic token verification and password verification.

6. According to the fine-grained access control method based on risk identification described in claim 1, the conflict resolution in the above step six includes: Pre-define a static conflict rule library to prohibit the simultaneous granting of mutually exclusive permissions (read and delete); Dynamic conflict handling: When the risk value changes and causes a conflict between the new permission and the granted permission, overwrite it according to the priority: Security policy permission > Functional permission; Basic operation permission > Advanced operation permission.

7. A system adopting the fine-grained access control method based on risk identification described in any one of claims 1-6, characterized in that: It includes an attribute collection and preprocessing module, an uncertainty modeling module, a comprehensive risk assessment module, and a dynamic access control decision module: The attribute collection and preprocessing module real-time collects and standardizes the multi-dimensional attribute data of the user's main body, behavior, and environment, and performs data preprocessing to provide a structured input basis for subsequent risk assessment; The uncertainty modeling module performs fuzzy processing on the standardized attributes through the interval type-2 fuzzy set (IT2FS) based on 7 user attributes, generates a fuzzy decision matrix including upper and lower membership degrees, quantifies the data uncertainty, and provides a computable fuzzy decision basis for risk assessment; The comprehensive risk assessment module integrates multi-dimensional attribute data and dynamic weights, dynamically establishes an evaluation chain for the user's risk level, and realizes the dynamic generation of standardized risk scores, providing a quantitative basis for permission decisions. The dynamic access control decision module dynamically adjusts the permission policy according to the real-time risk score and the preset threshold, and realizes accurate access control with risk adaptability by intelligently resolving permission conflicts and performing hierarchical responses (allow / secondary verification / block).

8. A computer device, comprising a memory and a processor, the memory storing a computer program, characterized in that, When the processor executes the computer program, the steps of the fine-grained access control method and system based on risk identification according to any one of claims 1 to 7 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that When the computer program is executed by the processor, the steps of the fine-grained access control method and system based on risk identification according to any one of claims 1 to 7 are implemented.

Images