Method and device for dynamically adjusting enterprise ai permission based on organization topology awareness

Abstract

The application discloses an enterprise AI permission dynamic adjustment method and device based on organization topology perception, and relates to the technical field of electric digital data processing.The enterprise AI permission dynamic adjustment method based on organization topology perception comprises the following steps: data stream access delay analysis, instruction processing accuracy analysis and instruction platform coverage analysis.The application performs data stream access delay analysis through access delay data to determine whether to perform receiving end structure adjustment optimization, then performs instruction processing accuracy analysis according to a topology network drift rate to determine whether to perform adaptive cross-node optimization, and finally performs instruction platform coverage analysis to determine whether to perform hierarchical data compression optimization, so that the accuracy of permission dynamic adjustment of an intelligent permission management platform when receiving enterprise permission change instructions is improved, and the problem of permission adjustment imbalance of the intelligent permission management platform caused by organization topology perception delay when user permission changes in the prior art is effectively solved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of electric digital data processing, and in particular to an enterprise AI permission dynamic adjustment method and device based on organization topology perception. BACKGROUND

[0002] In the field of enterprise intelligent permission management, with the rapid iteration of generative AI (Artificial Intelligence) technology, enterprise-level AI applications are undergoing a transformation from single-point tools to systematic productivity. The existing technology first extracts employee attributes, department architecture and business relationship data through HR (Human Resource) systems and directory services, constructs an organization topology graph containing role hierarchy, reporting relationship and collaboration mode by means of graph convolution network, and this process is implemented by using a graph database; then, by collecting user behavior logs and access context, combining a time series model to extract feature vectors such as operation frequency and resource access path; when the permission change instruction is sent to the receiving end, the receiving end processes the received multi-data source data, reads the current organization network topology structure, analyzes the instruction and implements the organization topology change in the intelligent management platform, and generates the corresponding implementation strategy for the business request of the target account according to the normal operation parameters and permission, data parameters of the management server using an AI deep learning model.

[0003] For example, the Chinese invention patent with publication number CN116166751B discloses a server comprehensive management cloud platform system and management method based on AI, which includes: a collection module for uploading normal operation parameters and permission parameters, data parameters of each management server to the cloud; a classification module for classifying the business information and function information of each management server; a training module for training the AI model using AI deep learning and classification results; and a management module for inputting the business request of the receiving account into the AI model for analysis and implementation.

[0004] For example, the Chinese patent application with publication number CN119397503A discloses an enterprise permission management method, system, device and medium based on machine learning, which includes: inputting the preprocessed historical operation of the target into the risk assessment model for training, transmitting the real-time operation data of the target user to the trained risk assessment model to obtain an evaluation result, and then performing permission allocation strategy according to the evaluation result.

[0005] The above-mentioned technology at least has the following technical problems:

[0006] In the prior art, the existing organization relationship data extraction mechanism mainly relies on the HR system batch data synchronization mode with a natural month as a period. This full-quantity data migration mechanism directly leads to a minimum perception response interval of post adjustment information in the system layer reaching a monthly level, that is, the organization network topology structure update period is too long, and the existing topology structure is changed by relying on input instructions. The personnel and post relationship update realized by an offline batch processing mode is difficult to meet the real-time topology discovery demand in a digital business process. The data routing topology reconstruction is performed by relying on an organization topology engine. The multi-dimensional correlation influence (including a permission inheritance chain, a business approval flow and an organization topology graph) caused by a key post change event causes a dynamic imbalance between the topology real-time performance of a permission management platform. A dynamic hierarchical cache mechanism is used to optimize the data synchronization coverage efficiency. There is a problem of imbalance in the adjustment of the intelligent permission management platform caused by the delay in the organization topology perception when the user permission changes. SUMMARY

[0007] The embodiments of the present application provide an enterprise AI permission dynamic adjustment method and device based on organization topology perception, solve the problem of imbalance in the adjustment of the intelligent permission management platform caused by the delay in the organization topology perception when the user permission changes in the prior art, and realize the improvement of the permission dynamic adjustment balance of the intelligent permission management platform when receiving the enterprise permission change instruction.

[0008] The embodiments of the present application provide an enterprise AI permission dynamic adjustment method based on organization topology perception, which includes the following steps: in a permission change instruction receiving period, performing data stream access delay analysis on the data stream access process of the organization topology structure according to the obtained access delay data, and simultaneously determining whether to perform receiving end structure adjustment optimization, the permission change instruction receiving period representing the time length corresponding to the receiving end of the intelligent permission management platform receiving the permission change instruction sent by the sending end, and the receiving end structure adjustment optimization representing improving the data stream access rate of the receiving end by adjusting the flow control thread parameter and the traversal path; in a permission change instruction processing period, performing instruction processing accuracy analysis on the data stream processing process of the intelligent permission management platform according to the obtained topology network drift rate, and simultaneously determining whether to perform adaptive cross-node optimization, the permission change instruction processing period representing the time length corresponding to the intelligent permission management platform receiving and processing the permission change instruction, and the adaptive cross-node optimization representing reducing the error rate of the permission change instruction processing by adaptive incremental optimization and cross-node optimization; in a permission change instruction coverage period, performing instruction platform coverage degree analysis on the permission synchronization process of the intelligent permission management platform according to the obtained permission synchronization influence data, and simultaneously determining whether to perform hierarchical data compression optimization, the permission change instruction coverage period representing the time length corresponding to the implementation of the intelligent permission management platform on the permission change instruction, and the hierarchical data compression optimization representing improving the coverage degree of the permission synchronization process by hierarchical recovery optimization and data compression optimization.

[0009] The embodiment of the application provides an enterprise AI permission dynamic adjustment device based on organization topology perception, which comprises a built-in monitoring timer, an embedded transaction log collector, a cache controller and an interface state monitoring router; the built-in monitoring timer is used for monitoring the strategy engine calculation time length and the topology graph traversal time length in real time; the embedded transaction log collector is used for detecting the inventory synchronization error rate in real time; the cache controller is used for measuring the cache recovery interval in real time; and the interface state monitoring router is used for monitoring the topology network drift rate and the network topology node distance in real time.

[0010] The one or more technical solutions provided in the embodiment of the application have at least the following technical effects or advantages:

[0011] 1. By analyzing access delay data, topology network drift rate and permission synchronization influence data in real time, triggering receiving end structure adjustment optimization, adaptive cross-node optimization and hierarchical data compression optimization respectively, real-time perception of organization topology and rapid response of platform permission during user permission change are realized, and the problem of permission adjustment imbalance caused by topology perception delay is solved; the access delay data is compared and corrected in the database to generate a precise coefficient, the real-time influence of the quantitative instruction is received through coupling processing, the delay evaluation accuracy is significantly improved, and the accuracy of the receiving end optimization decision is ensured; cross-node optimization is dynamically implemented by using topology drift rate analysis, which effectively overcomes the instruction transmission deviation caused by network structure changes; a permission synchronization influence index is constructed, and the coverage is accurately located, so that the hierarchical data compression can reduce the synchronization data volume in a targeted manner, and the propagation speed, execution accuracy and system coverage of permission change are improved.

[0012] 2. By comparing the real-time access delay data with the preset inventory data and correcting the access delay coefficient generated by the correction value, the real-time influence degree of the data stream access delay index formed by coupling processing is quantified. This processing improves the accuracy of the delay index acquisition, eliminates the inherent deviation of the system through the correction value calibration link, and makes the coefficient closer to the real delay level; coupling processing integrates multi-dimensional delay characteristics to construct a scientific and reliable quantitative index; the index itself reflects the timeliness loss degree of the permission platform receiving instruction, and provides a decision basis for receiving end structure optimization; the overall evaluation process improves the accuracy due to the standardized correction and comprehensive coupling, and ensures the accuracy of the actual influence judgment of the network access delay.

[0013] 3. The permission synchronization influence index is generated by harmonically averaging the cache recovery interval and the network topology node distance ratio result, accurately quantifying the instruction implementation coverage influence degree, improving the accuracy of the index acquisition; the harmonically averaging method balances the interaction of cache recovery efficiency and network transmission distance, eliminates single extreme value interference, and makes the index reflect the reality of permission synchronization; the generated index accurately represents the limited degree of instruction coverage of the intelligent platform, and the index itself accurately captures the comprehensive influence of cache strategy and network distance on permission synchronization, so that the reliability of the coverage evaluation result is improved; through quantitative analysis of the actual role relationship of instruction propagation range, accurate basis is provided for optimizing the data synchronization mechanism. BRIEF DESCRIPTION OF DRAWINGS

[0014] Figure 1 A flowchart of the enterprise AI permission dynamic adjustment method based on organization topology perception provided by the embodiment of the present application is shown in the figure.

[0015] Figure 2 A flowchart of the enterprise AI permission dynamic adjustment method based on organization topology perception provided by the embodiment of the present application is shown in the figure.

[0016] Figure 3 A flowchart of the enterprise AI permission dynamic adjustment method based on organization topology perception provided by the embodiment of the present application is shown in the figure.

[0017] Figure 4 A flowchart of the enterprise AI permission dynamic adjustment method based on organization topology perception provided by the embodiment of the present application is shown in the figure. DETAILED DESCRIPTION

[0018] The embodiment of the present application provides an enterprise AI permission dynamic adjustment method and device based on organization topology perception, which solves the problem of imbalance of intelligent permission management platform permission adjustment caused by organization topology perception delay when user permission changes in the prior art. Through the permission change instruction receiving period, the data stream access process of the organization topology structure is analyzed according to the obtained access delay data, and it is determined whether to perform receiving end structure adjustment optimization. Then, in the permission change instruction processing period, the data stream processing process of the intelligent permission management platform is analyzed according to the obtained topology network drift rate, and it is determined whether to perform adaptive cross-node optimization. Finally, in the permission change instruction coverage period, the permission synchronization process of the intelligent permission management platform is analyzed according to the obtained permission synchronization influence data, and it is determined whether to perform hierarchical data compression optimization. The embodiment of the present application realizes the improvement of the dynamic adjustment balance of the intelligent permission management platform when receiving enterprise permission change instructions.

[0019] The technical solutions in the embodiments of the present application are to solve the problem of unbalanced adjustment of the intelligent permission management platform caused by the delay of organization topology awareness when the user permission changes, and the general idea is as follows:

[0020] According to the access delay data analysis result, the necessity of receiving end structure adjustment optimization is determined, the adaptive cross-node optimization demand is judged according to the topology network drift rate evaluation instruction processing accuracy, and the hierarchical data compression optimization implementation condition is determined through the permission synchronization influence data analysis instruction platform coverage, so that the effect of improving the dynamic adjustment balance of the intelligent permission management platform when receiving the enterprise permission change instruction is achieved.

[0021] In order to better understand the above technical solutions, the above technical solutions will be described in detail in combination with the drawings of the specification and specific embodiments.

[0022] As shown in Figure 1 The flowchart of the enterprise AI permission dynamic adjustment method based on organization topology awareness provided by the embodiments of the present application, the enterprise AI permission dynamic adjustment method based on organization topology awareness provided by the embodiments of the present application, comprising the following steps: during the permission change instruction receiving period, the data stream access delay of the organization topology structure data stream access process is analyzed according to the obtained access delay data, and it is determined whether to perform receiving end structure adjustment optimization, the permission change instruction receiving period represents the time length corresponding to the receiving end of the intelligent permission management platform receiving the permission change instruction sent by the sending end, and the receiving end structure adjustment optimization represents that the data stream access rate of the receiving end is improved by adjusting the flow control thread parameter and the traversal path; during the permission change instruction processing period, the instruction processing accuracy of the data stream processing process of the intelligent permission management platform is analyzed according to the obtained topology network drift rate, and it is determined whether to perform adaptive cross-node optimization, the permission change instruction processing period represents the time length corresponding to the processing of the permission change instruction received by the intelligent permission management platform, and the adaptive cross-node optimization represents that the error rate of the permission change instruction processing is reduced by adaptive incremental optimization and cross-node optimization; during the permission change instruction coverage period, the instruction platform coverage of the permission synchronization process of the intelligent permission management platform is analyzed according to the obtained permission synchronization influence data, and it is determined whether to perform hierarchical data compression optimization, the permission change instruction coverage period represents the time length corresponding to the implementation coverage of the permission change instruction by the intelligent permission management platform, and the hierarchical data compression optimization represents that the coverage of the permission synchronization process is improved by hierarchical recovery optimization and data compression optimization.

[0023] In the embodiment, the method realizes dynamic closed-loop optimization of the whole process of permission change, forms a synergistic effect of data flow access acceleration, instruction processing precision improvement, and permission coverage range expansion, improves the dynamic adaptability of permission management to complex organization topology, guarantees the timeliness and accuracy of permission change in a multi-node collaborative scenario, and improves the coverage of permission change instructions on the whole platform, thereby realizing the improvement of the balance of dynamic adjustment of the intelligent permission management platform when receiving enterprise permission change instructions.

[0024] As shown in Figure 2 The flowchart of the data flow access delay analysis of the enterprise AI permission dynamic adjustment method based on organization topology awareness provided by the embodiment of the application is shown in FIG. 1. The specific design logic is as follows: judging whether the data flow access delay index exceeds the standard, if not, marking it as qualified and performing instruction processing accuracy analysis; if it exceeds the standard, generating a first data adjustment value to adjust the flow control thread parameter, and judging whether the reduction meets the preset range after re-measuring the index; if it meets the requirement, performing instruction processing accuracy analysis; if it does not meet the requirement, generating a second data adjustment value to adjust the traversal path, and after re-measuring the index: if it does not exceed the standard, entering the next link; if it still exceeds the standard, checking the optimization times, if it does not exceed the limit, re-generating the data adjustment value for circulation, if it exceeds the limit, sending a preliminary warning instruction.

[0025] Further, the data flow access delay analysis of the data flow access process of the organization topology structure is performed according to the obtained access delay data. The specific steps include: comparing the obtained access delay data with the preset access delay data in the database in terms of difference, and simultaneously performing correction processing by combining the access delay data correction value to obtain the access delay data coefficient, and performing coupling processing to obtain the data flow access delay index.

[0026] The access delay data includes the policy engine calculation time length, the inventory synchronization error rate, and the topology graph traversal time length. The policy engine calculation time length represents the time required by the intelligent permission management platform to analyze the permission change instruction. The inventory synchronization error rate is used to reflect the degree of inconsistency between the data domain and the local cache data of the intelligent permission management platform. The topology graph traversal time length represents the time required by the intelligent permission management platform to traverse the real-time organization topology structure associated with the permission. The preset access delay data includes the preset policy engine calculation time length, the inventory synchronization error rate, and the topology graph traversal time length. The access delay data correction value includes the policy engine calculation time length correction value, the inventory synchronization error rate correction value, and the topology graph traversal time length correction value. The access delay data coefficient includes the engine calculation time length coefficient, the inventory synchronization error rate coefficient, and the topology graph traversal time length coefficient. The data flow access delay index represents the influence degree quantization data of the access delay data on the real-time instruction receiving of the intelligent permission management platform.

[0027] Specifically, the specific expression of the policy engine calculation time length coefficient Q1 is as follows: , wherein Q1 represents a policy engine calculation duration coefficient corresponding to the end of the data stream access delay analysis period, C1 represents a policy engine calculation duration correction value, Q PE represents a policy engine calculation duration corresponding to the end of the data stream access delay analysis period, PE0 represents a preset policy engine calculation duration.

[0028] The specific expression of the inventory synchronization error rate coefficient Q2 is: , wherein Q2 represents an inventory synchronization error rate coefficient corresponding to the end of the data stream access delay analysis period, C2 represents an inventory synchronization error rate correction value, Q ER represents an inventory synchronization error rate corresponding to the end of the data stream access delay analysis period, ER0 represents a preset inventory synchronization error rate.

[0029] The specific expression of the topology graph traversal duration coefficient Q3 is: , wherein Q3 represents a topology graph traversal duration coefficient corresponding to the end of the data stream access delay analysis period, C3 represents a topology graph traversal duration correction value, Q TM represents a topology graph traversal duration corresponding to the end of the data stream access delay analysis period, TM0 represents a preset topology graph traversal duration.

[0030] The specific expression of the data stream access delay index I M is: , , wherein I M represents a data stream access delay index corresponding to the end of the data stream access delay analysis period.

[0031] In this embodiment, the preset access delay data is represented by the result of summing and averaging the historical access delay data corresponding to the end of the historical data stream access delay analysis period, the access delay data is obtained by acquiring the access delay data corresponding to the end of the data stream access delay analysis period, and the policy engine calculation duration correction value, the inventory synchronization error rate correction value, and the topology graph traversal duration correction value are numerical values preset in the database for measuring the influence degree of the policy engine calculation duration, the inventory synchronization error rate, and the topology graph traversal duration on the data stream access delay index. The database stores corresponding correction values for each parameter, and there is a preset mapping relationship between them, which can be one-to-one or many-to-one. In actual application, by inputting real-time policy engine calculation duration, inventory synchronization error rate, and topology graph traversal duration, the corresponding correction values can be accurately obtained, providing a quantitative basis for evaluating the real-time influence of the data stream access delay index on the permission change instruction, helping to accurately calculate the data stream access delay index, and the value range of the three correction values is 0 to 1, and the sum of the three is equal to 1.

[0032] It should be noted that the data stream access delay index increases with the increase of the policy engine calculation time length, the inventory synchronization error rate and the topology graph traversal time length, wherein the increment of the policy engine calculation time length triggers more frequent inventory data verification requests, which increases the inventory synchronization load and thus amplifies the synchronization error rate; the accumulation of inventory synchronization errors increases the frequency of topology graph traversal, performs data consistency repair, and thus prolongs the topology structure traversal time length; the increase of the topology graph traversal time length causes the real-time deviation of the context information update during the policy engine calculation process, which increases the policy engine calculation time length.

[0033] By considering the above mutual influence mechanism, the relationship between the data stream access delay index and each variable can be more comprehensively understood, which is crucial for real-time evaluation during the data stream access delay index acquisition process. By optimizing the policy engine calculation time length, the inventory synchronization error rate and the topology graph traversal time length, the intelligent permission management platform improves the balance of dynamic permission adjustment when receiving enterprise permission change instructions, effectively solving the imbalance problem of intelligent permission management platform permission adjustment caused by organization topology perception delay when user permission changes.

[0034] Further, whether to perform receiving end structure adjustment optimization is determined, and the specific steps are: comparing the obtained data stream access delay index with the preset data stream access delay index in the database: if the obtained data stream access delay index is greater than the preset data stream access delay index in the database, it is recorded as unqualified data stream access and receiving end structure adjustment optimization is performed; if the obtained data stream access delay index is not greater than the preset data stream access delay index in the database, it is recorded as qualified data stream access and instruction processing accuracy analysis is performed.

[0035] Among them, the specific steps of receiving end structure adjustment optimization are: the obtained data stream access delay index deviation and serialization delay deviation are respectively processed in proportion to obtain data stream access delay index deviation score and serialization delay deviation score, and the obtained data stream access delay index deviation score and serialization delay deviation score are simultaneously processed by sum average to obtain a first data adjustment value, the first data adjustment value is the result of the sum average processing of the data stream access delay index deviation score and the serialization delay deviation score after the mapping in the database, which is used to prompt the adaptive flow control manager to adjust the flow control thread parameters based on the obtained first data adjustment value to reduce the delay caused by congestion.

[0036] After the adjustment and optimization of the receiving end structure, it is judged whether the reduction amplitude of the obtained data stream access delay index deviation is within the preset reduction amplitude range in the database. If yes, the adjustment and optimization of the receiving end structure is completed and the instruction processing accuracy analysis is performed. Otherwise, the second data adjustment value is obtained by performing proportional processing on the reacquired data stream access delay index deviation and serialization delay deviation respectively and then performing summation average processing on the results, which is used to prompt the path management controller to reduce the system response delay caused by redundant traversal.

[0037] The preset reduction amplitude range represents a closed interval corresponding to the maximum and minimum values of the historical data stream access delay index deviation reduction amplitude after the historical receiving end structure adjustment and optimization in the database. If the reacquired data stream access delay index is not greater than the preset data stream access delay index in the database within a specified number of receiving end structure adjustment and optimization, the structure adjustment and optimization is completed and the instruction processing accuracy analysis is performed. Otherwise, the receiving end structure adjustment and optimization instruction is sent again. If it is not within the specified number of receiving end structure adjustment and optimization, the preliminary processing warning is directly performed.

[0038] In the embodiment, the current flow control thread parameters are adjusted in real time based on the obtained current flow control thread parameters by the cascading permission adjustment algorithm to improve the multi-thread parallel efficiency. Meanwhile, the historical first data adjustment value is input into the neural network model in the reinforcement learning model as sample data, the data adjustment-flow control path adjustment neural network model is obtained by training based on the cascading permission adjustment algorithm, the first and second data adjustment values obtained currently are input into the data adjustment-flow control path adjustment neural network model, and the corresponding flow control thread parameter adjustment value and traversal path adjustment value are output. The data stream access delay index deviation represents the difference between the obtained data stream access delay index and the preset data stream access delay index in the database, the serialization delay deviation represents the difference between the obtained serialization delay and the preset serialization delay in the database, and the data stream access delay index deviation reduction amplitude represents the absolute value of the difference between the data stream access delay indexes obtained before and after the receiving end structure adjustment and optimization.

[0039] The example generates flow control parameter adjustment instructions to eliminate the queuing delay caused by thread congestion. The shortest access path sequence marking mechanism of the path management controller reduces the length of the redundant traversal link, synchronously compresses the variable interval of the system response delay, and the multi-layer feedback mechanism improves the probability of the system reaching a stable state within multiple iterations, reduces the data loss rate, and optimizes the overall process to form a low-delay control architecture, achieving dynamic optimal point balance of the three indicators of transmission efficiency, computing resource occupancy rate and instruction processing accuracy.

[0040] As Figure 3As shown, the flow chart of the corresponding instruction processing accuracy analysis of the enterprise AI permission dynamic adjustment method based on organization topology awareness provided by the embodiment of the application, the specific design logic is: judge whether the topology network drift rate is over standard, if not over standard, then perform instruction platform coverage analysis; if over standard, generate a first incremental adjustment value to adjust the routing filtering coefficient, and then judge whether the reduction meets the preset range after re-measuring the drift rate; if it meets, then perform instruction platform coverage analysis; if it does not meet, then generate a second incremental adjustment value to adjust the dynamic window size, and then re-measure the drift rate: if it is not over standard, then perform instruction platform coverage analysis; if it is still over standard, then check the optimization times, if it is not over limit, then re-generate the incremental adjustment value and cycle; if it is over limit, then adjust the traction step after re-measuring the drift rate, if it is over standard, then trigger the phase-locked loop warning, otherwise, perform instruction platform coverage analysis.

[0041] Further, according to the acquired topology network drift rate, the data flow processing process of the intelligent permission management platform is analyzed for instruction processing accuracy, and the specific steps are: comparing the acquired topology network drift rate with the preset topology network drift rate in the database: if the acquired topology network drift rate is greater than the preset topology network drift rate in the database, then it is recorded as unqualified instruction processing and adaptive cross-node optimization is performed; if the acquired topology network drift rate is not greater than the preset topology network drift rate in the database, then it is recorded as qualified instruction processing and instruction platform coverage analysis is performed.

[0042] The specific steps of determining whether to perform adaptive cross-node optimization are: obtaining a first incremental adjustment value by mapping the obtained topological network drift rate deviation and cache hit rate deviation in the database, the first incremental adjustment value being a result obtained by mapping the topological network drift rate deviation and the cache hit rate deviation in the database, and being used to prompt the routing-aware controller to suppress network drift caused by core node processing resources based on the obtained first incremental adjustment value; after one adaptive incremental optimization, determining whether the obtained topological network drift rate reduction range is within a preset reduction range in the database, if yes, completing the adaptive incremental optimization, otherwise, obtaining a second incremental adjustment value by mapping the re-obtained topological network drift rate deviation and cache hit rate deviation in the database, the second incremental adjustment value being a result obtained by mapping the re-obtained topological network drift rate deviation and cache hit rate deviation in the database, and being used to prompt the cache controller of the dynamic hierarchical cache area to suppress drift caused by changes in the topological network structure characteristics due to permission changes, and the preset reduction range being a closed interval corresponding to the maximum and minimum values of the historical topological network drift rate reduction range after historical adaptive incremental optimization in the database; if the re-obtained topological network drift rate is not greater than the preset topological network drift rate in the database within a specified number of adaptive incremental optimization times, completing the adaptive incremental optimization and performing instruction platform coverage analysis, otherwise, sending an adaptive incremental optimization instruction again; if not within the specified number of adaptive incremental optimization times, directly performing cross-node optimization.

[0043] The specific steps of cross-node optimization are: reducing the steady-state error caused by phase jitter based on the phase-locked loop traction step adjustment value obtained after adaptive incremental optimization, the traction step adjustment value indicating a result obtained by mapping the re-obtained topological network drift rate in the database after adaptive incremental optimization; after cross-node optimization, determining whether the re-obtained topological network drift rate is greater than the preset topological network drift rate in the database, if yes, performing phase-locked loop warning, otherwise, completing adaptive cross-node optimization and performing instruction platform coverage analysis.

[0044] In the embodiment, the topology network drift rate is determined by an HR system (human resource management system integrating human resource related processes, data and strategies through digital means), a network relationship radar is combined to obtain a node input of network architecture change, a time decay weighted algorithm is input to output a topology network drift rate result, a cache hit rate is obtained through an operating system performance counter, a preset topology network drift rate is represented by a result obtained by summing and averaging a topology network drift rate corresponding to an end of a historical instruction processing accuracy analysis period, a topology network drift rate reduction amplitude represents a difference between topology network drift rates obtained before and after adaptive incremental optimization, a topology network drift rate deviation represents a difference between an obtained topology network drift rate and a preset topology network drift rate in a database, a cache hit rate deviation represents a difference between an obtained cache hit rate and a preset cache hit rate in the database, a routing filter coefficient is adjusted in real time based on an obtained current routing filter coefficient through a departure incremental synchronization algorithm to suppress network drift caused by core node processing resources, and a historical first incremental adjustment value is input to a multivariate linear regression model as sample data, a routing filter-cache window linear regression model is obtained based on the departure incremental synchronization algorithm, and a first and second incremental adjustment value is input to the routing filter-cache window linear regression model, and corresponding routing filter parameter adjustment value and dynamic window size adjustment value are output.

[0045] The example reduces the structural disturbance decay rate triggered by permission change through dynamic window size elastic adaptation, compresses the phase jitter peak-to-peak value through a nonlinear correction strategy of traction step, reduces the probability of network drift rate over-limit event occurrence through a closed-loop control architecture, improves the data access path optimization rate through the introduction of a dynamic hierarchical cache area intelligent preloading mechanism, and shortens the period under a high-frequency topology change scenario through cross-node collaborative optimization under multiple constraint conditions, and balances the relationship between load and efficiency.

[0046] As shown in Figure 4 The flowchart of the corresponding instruction platform coverage degree analysis of the enterprise AI permission dynamic adjustment method based on organization topology perception provided by the embodiment of the application is shown in FIG. 1. The specific design logic is as follows: it is judged whether the permission synchronization influence index is over-standard. If not, the instruction platform coverage degree analysis is completed. If yes, a first synchronization adjustment value is generated to adjust the verification period, and the index is re-measured to judge whether the reduction amplitude meets the preset range. If yes, the instruction platform coverage degree analysis is completed. If no, a second synchronization adjustment value is generated to adjust the relay node forwarding threshold, and the index is re-measured. If not over-standard, the instruction platform coverage degree analysis is completed. If still over-standard, the optimization times are checked. If not over-limit, the synchronization adjustment value is re-generated for circulation. If over-limit, the index is re-measured after adjusting the fragmentation granularity. If over-standard, a fragmentation warning is sent. Otherwise, the instruction platform coverage degree analysis is completed.

[0047] Further, the permission synchronization process of the intelligent permission management platform is instructed to analyze the platform coverage degree according to the acquired permission synchronization influence data. The specific steps are as follows: the acquired cache recovery interval and network topology node distance are respectively processed in proportion to obtain cache recovery interval scores and network topology node distance scores; the acquired cache recovery interval scores and network topology node distance scores are processed by harmonic mean to obtain a permission synchronization influence index, which represents the influence degree quantization data of the platform coverage degree of the permission synchronization influence data.

[0048] In the embodiment, the edge node lagging in the policy update process is identified by the dynamic calibration technology of the cache recovery interval score, a preloading compensation mechanism is constructed synchronously to shorten the startup period, the hot spot area of cross-node communication is predicted to avoid, the permission attenuation effect caused by multi-level transfer is reduced, the integrity and real-time requirement of permission change is ensured to achieve a balance state in different network density scenarios, and the compatibility of the permission synchronization mechanism to heterogeneous nodes and the global stability of maintaining the policy synchronization radiation range in the dynamic node expansion scenario are improved.

[0049] Further, the specific steps of determining whether to perform hierarchical data compression optimization are as follows: the acquired permission synchronization influence index is compared with the preset permission synchronization influence index in the database; if the acquired permission synchronization influence index is greater than the preset permission synchronization influence index in the database, it is recorded as unqualified instruction platform coverage degree and hierarchical recovery optimization is performed; if the acquired permission synchronization influence index is not greater than the preset permission synchronization influence index in the database, it is recorded as qualified instruction platform coverage degree and enterprise AI permission dynamic adjustment is completed.

[0050] The specific steps of hierarchical recovery optimization are as follows: the acquired permission synchronization influence index deviation and serialization efficiency deviation are respectively processed in proportion to obtain permission synchronization influence index deviation scores and serialization efficiency deviation scores; the acquired permission synchronization influence index deviation scores and serialization efficiency deviation scores are processed by sum average to obtain a first synchronization adjustment value, which is used to prompt the configuration manager to increase the synchronization efficiency and coverage rate based on the acquired first synchronization adjustment value; after one hierarchical recovery optimization, it is judged whether the acquired permission synchronization influence index deviation reduction amplitude is within the preset reduction amplitude range in the database; if yes, the hierarchical recovery optimization is completed; otherwise, the acquired permission synchronization influence index deviation scores and serialization efficiency deviation scores are processed by sum average to obtain a second synchronization adjustment value, which is the result of mapping the permission synchronization influence index deviation scores and serialization efficiency scores in the database, and is used to prompt the policy control layer to reduce the synchronization policy delay caused by too many communication hops based on the acquired second synchronization adjustment value. The preset reduction amplitude range is a closed interval corresponding to the maximum and minimum values of the historical permission synchronization influence index deviation reduction amplitude after historical hierarchical recovery optimization in the database.

[0051] If it is judged that the re-acquired permission synchronization impact indicator is not greater than the preset permission synchronization impact indicator in the database within the specified number of hierarchical recovery optimization times, hierarchical recovery optimization is completed and enterprise AI permission dynamic adjustment is completed, otherwise, a hierarchical recovery re-optimization instruction is sent; if it is not within the specified number of hierarchical recovery optimization times, data compression optimization is directly performed; the specific steps of data compression optimization are: based on the data shard granularity adjustment value obtained after hierarchical recovery optimization to improve the integrity and parallel efficiency of data transmission, the shard granularity adjustment value represents the result obtained by mapping the re-acquired permission synchronization impact indicator after hierarchical recovery optimization in the database; after data compression optimization, it is judged whether the re-acquired permission synchronization impact indicator is greater than the preset permission synchronization impact indicator in the database, if yes, a shard warning is performed, otherwise, hierarchical data compression optimization is completed and enterprise AI permission dynamic adjustment is completed.

[0052] In the embodiment, the preset permission synchronization impact indicator is represented by the result obtained by averaging the sum of the corresponding permission synchronization impact indicators at the end of the historical instruction platform coverage analysis period, the permission synchronization impact indicator deviation reduction amplitude represents the difference between the permission synchronization impact indicators obtained before and after hierarchical recovery optimization, the permission synchronization impact indicator deviation represents the difference between the obtained permission synchronization impact indicator and the preset permission synchronization impact indicator in the database, and the serialization efficiency deviation represents the difference between the obtained serialization efficiency and the preset serialization efficiency in the database; based on the obtained current verification period, the verification period is adjusted in real time by a time series prediction algorithm to improve synchronization efficiency and coverage, and at the same time, the historical first synchronization adjustment value is input into the feedback neural network model in the neural network model as sample data, and the central threshold-forwarding period neural network model is obtained by training based on the time series prediction algorithm, and the current obtained first and second synchronization adjustment values are input into the central threshold-forwarding period neural network model to output the corresponding verification period adjustment value and forwarding threshold adjustment value.

[0053] The verification period dynamic calibration captures the real-time state of the edge node data retention, forms a synchronization that takes into account timeliness and integrity, and reduces the verification resource competition caused by batch strategy updates; the transmission forwarding threshold adjustment adjusts the logical topology of the inter-node communication path, reduces the strategy effective delay caused by multi-level relay, and synchronously enhances the queue scheduling flexibility of the relay node to sudden permission changes; adaptive adjustment of the data shard granularity ensures the linear constraint between data integrity and transmission efficiency, maintains the semantic integrity of the data in the compression process, and achieves dynamic balance of transmission utilization; the hierarchical progressive warning-optimization closed-loop system establishes a multi-dimensional steady-state control capability of permission synchronization, which can not only inhibit the coverage collapse caused by shard abnormalities in high-frequency change scenarios, but also ensure that the minimum synchronization resource load is maintained during steady-state operation.

[0054] The enterprise AI permission dynamic adjustment device based on organization topology perception provided by the embodiment of the application comprises: a built-in monitoring timer, an embedded transaction log collector, a cache controller and an interface state monitoring router; the built-in monitoring timer is used for monitoring the strategy engine calculation time length and the topology graph traversal time length in real time; the embedded transaction log collector is used for detecting the inventory synchronization error rate in real time; the cache controller is used for measuring the cache recovery interval in real time; and the interface state monitoring router is used for monitoring the topology network drift rate and the network topology node distance in real time.

[0055] To sum up, the embodiment of the application triggers the receiving end structure adjustment optimization, the adaptive cross-node optimization and the hierarchical data compression optimization respectively by analyzing the access delay data, the topology network drift rate and the permission synchronization influence data in real time, realizes the real-time perception of the organization topology and the rapid response of the platform permission when the user permission changes, solves the permission adjustment imbalance problem caused by topology perception delay; the access delay data is compared and corrected in the database to generate a precise coefficient, the real-time influence of the coupling processing is quantified, the delay evaluation accuracy is significantly improved, and the accuracy of the receiving end optimization decision is ensured; the cross-node optimization is dynamically implemented by using the topology drift rate analysis, the instruction transmission deviation caused by the network structure change is effectively overcome; the permission synchronization influence index is constructed, the coverage is accurately located, the hierarchical data compression can reduce the synchronization data volume in a targeted manner, and the propagation speed, the execution accuracy and the system coverage range of the permission change are improved.

[0056] Those skilled in the art will understand that the embodiments of the application can be provided as a method, a system or a computer program product. Therefore, the application can take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Moreover, the application can take the form of a computer program product implemented on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0057] The application is described with reference to flowcharts and / or block diagrams according to the embodiments of the application. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, and the combination of the flows and / or blocks in the flowcharts and / or block diagrams can be implemented by computer program instructions. These computer program instructions can be provided to a general-purpose computer, a special-purpose computer, an embedded processor or other programmable data processing devices to produce a machine, so that the instructions executed by the computer or other programmable data processing devices produce a device that implements the functions specified in the flowcharts and / or block diagrams. Figure 1 The device that implements the functions specified in one flow or multiple flows and / or blocks Figure 1 The device that implements the functions specified in one flow or multiple flows and / or blocks

[0058] These computer program instructions can also be stored in a computer- readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instructions which implement the Figure 1 function specified in the flow or flows and / or blocks Figure 1 of the block or blocks.

[0059] The computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions that are executed on the computer or other programmable apparatus provide steps for implementing the Figure 1 function specified in the flow or flows and / or blocks Figure 1 of the block or blocks.

[0060] While the preferred embodiments of the application have been described, additional variations and modifications can be made to the embodiments by those of skill in the art once they have the benefit of the present disclosure. Therefore, the appended claims are intended to encompass within their scope all possible variations and modifications of the preferred embodiments. 1

[0061] It is apparent that a person skilled in the art can make various changes and modifications to the application without departing from the spirit and scope thereof. Therefore, if these modifications and changes fall within the scope of the claims and their equivalents, it is intended to include them in the application.

Claims

1. A method for dynamically adjusting enterprise AI permissions based on organization topology awareness, characterized in that, The method comprises the following steps: A1, during the permission change instruction receiving period, performing data stream access delay analysis on the data stream access process of the organization topology structure according to the obtained access delay data, and determining whether to perform receiving end structure adjustment optimization, the permission change instruction receiving period represents the time length corresponding to the receiving end of the intelligent permission management platform receiving the permission change instruction sent by the sending end, and the receiving end structure adjustment optimization represents improving the data stream access rate of the receiving end by adjusting the flow control thread parameter and the traversal path; The data stream access delay analysis on the data stream access process of the organization topology structure according to the obtained access delay data comprises the following steps: The obtained access delay data and the preset access delay data in the database are compared in terms of difference, and are modified by combining the access delay data correction value to obtain the access delay data coefficient, and are coupled to obtain the data stream access delay index; The access delay data comprises the policy engine calculation time length, the inventory synchronization error rate and the topology graph traversal time length, and the data stream access delay index represents the influence degree quantization data of the access delay data on the real-time of the instruction receiving of the intelligent permission management platform; A2, during the permission change instruction processing period, performing instruction processing accuracy analysis on the data stream processing process of the intelligent permission management platform according to the obtained topology network drift rate, and determining whether to perform adaptive cross-node optimization, the permission change instruction processing period represents the time length corresponding to the processing of the permission change instruction received by the intelligent permission management platform, and the adaptive cross-node optimization represents reducing the error rate of the processing of the permission change instruction by adaptive incremental optimization and cross-node optimization; A3, during the permission change instruction coverage period, performing instruction platform coverage analysis on the permission synchronization process of the intelligent permission management platform according to the obtained permission synchronization influence data, and determining whether to perform hierarchical data compression optimization, the permission change instruction coverage period represents the time length corresponding to the implementation coverage of the permission change instruction by the intelligent permission management platform, and the hierarchical data compression optimization represents improving the coverage of the permission synchronization process by hierarchical recovery optimization and data compression optimization.

2. The method of claim 1, wherein the method further comprises: The specific steps of determining whether to perform receiving end structure adjustment optimization are as follows: The obtained data stream access delay index is compared with the preset data stream access delay index in the database: If the obtained data stream access delay index is greater than the preset data stream access delay index in the database, it is recorded as unqualified data stream access and receiving end structure adjustment optimization is performed; If the obtained data stream access delay index is not greater than the preset data stream access delay index in the database, it is recorded as qualified data stream access and instruction processing accuracy analysis is performed.

3. The method of claim 2, wherein the method further comprises: The specific steps of receiving end structure adjustment optimization are as follows: The data stream access delay index deviation and the serialization delay deviation are respectively processed in proportion to obtain data stream access delay index deviation scores and serialization delay deviation scores, and the data stream access delay index deviation scores and the serialization delay deviation scores are processed in summation and averaging to obtain a first data adjustment value, which is used to prompt the adaptive flow control manager to reduce the delay caused by congestion based on the obtained first data adjustment value; After one-time reception end structure adjustment optimization, it is judged whether the obtained data stream access delay index deviation reduction range is within the preset reduction range in the database. If yes, the reception end structure adjustment optimization is completed and instruction processing accuracy analysis is performed. Otherwise, the result of the summation and averaging processing after the reacquired data stream access delay index deviation and serialization delay deviation are respectively processed in proportion is taken as a second data adjustment value, which is used to prompt the path management controller to reduce the system response delay caused by redundant traversal; If the reacquired data stream access delay index is not greater than the preset data stream access delay index in the database within a specified number of reception end structure adjustment optimization times, the structure adjustment optimization is completed and instruction processing accuracy analysis is performed. Otherwise, a reception end structure adjustment re-optimization instruction is sent. If it is not within the specified number of reception end structure adjustment optimization times, preliminary processing early warning is directly performed.

4. The method of claim 1, wherein the method further comprises: The specific steps of the instruction processing accuracy analysis of the data stream processing process of the intelligent permission management platform according to the obtained topology network drift rate are as follows: The obtained topology network drift rate is compared with the preset topology network drift rate in the database: If the obtained topology network drift rate is greater than the preset topology network drift rate in the database, it is recorded as unqualified instruction processing and adaptive incremental optimization is performed; If the obtained topology network drift rate is not greater than the preset topology network drift rate in the database, it is recorded as qualified instruction processing and instruction platform coverage analysis is performed.

5. The method of claim 4, wherein the method further comprises: The specific steps of the adaptive incremental optimization are as follows: A first incremental adjustment value is obtained by mapping the obtained topology network drift rate deviation and cache hit rate deviation in the database, which is used to prompt the routing awareness controller to suppress the network drift caused by core node processing resources based on the obtained first incremental adjustment value; After one-time adaptive incremental optimization, it is judged whether the obtained topology network drift rate reduction range is within the preset reduction range in the database. If yes, the adaptive incremental optimization is completed. Otherwise, a second incremental adjustment value is obtained by mapping the reacquired topology network drift rate deviation and cache hit rate deviation in the database, which is used to prompt the cache controller of the dynamic hierarchical cache area to suppress the drift caused by topology network structure feature changes based on the obtained second incremental adjustment value; If the reacquired topology network drift rate is not greater than the preset topology network drift rate in the database within a specified number of adaptive incremental optimization times, the adaptive incremental optimization is completed and instruction platform coverage analysis is performed. Otherwise, an adaptive incremental re-optimization instruction is sent. If not within the specified adaptive incremental optimization times, directly cross-node optimization is performed; The specific steps of the cross-node optimization are: based on the phase-locked loop traction step adjustment value obtained after adaptive incremental optimization to reduce the steady-state error caused by phase jitter, the traction step adjustment value represents the result of the re-acquired topology network drift rate in the database mapping; After cross-node optimization, it is judged whether the re-acquired topology network drift rate is greater than the preset topology network drift rate in the database. If yes, a phase-locked loop warning is performed, otherwise the adaptive cross-node optimization is completed and the instruction platform coverage analysis is performed.

6. The method of claim 1, wherein the method further comprises: The specific steps of the instruction platform coverage analysis according to the obtained permission synchronization influence data are: The obtained cache recovery interval and network topology node distance are respectively processed in proportion to obtain cache recovery interval score and network topology node distance score. The obtained cache recovery interval score and network topology node distance score are processed in harmonic average to obtain a permission synchronization influence index, which represents the influence degree quantization data of the platform coverage of the permission synchronization influence data.

7. The method of claim 6, wherein the method further comprises: The specific steps of determining whether to perform hierarchical data compression optimization are: The obtained permission synchronization influence index is compared with the preset permission synchronization influence index in the database: If the obtained permission synchronization influence index is greater than the preset permission synchronization influence index in the database, it is recorded as unqualified instruction platform coverage and hierarchical recovery optimization is performed; If the obtained permission synchronization influence index is not greater than the preset permission synchronization influence index in the database, it is recorded as qualified instruction platform coverage and enterprise AI permission dynamic adjustment is completed.

8. The method of claim 7, wherein the method further comprises: The specific steps of the hierarchical recovery optimization are: The obtained permission synchronization influence index deviation and serialization efficiency deviation are respectively processed in proportion to obtain permission synchronization influence index deviation score and serialization efficiency deviation score. The obtained permission synchronization influence index deviation score and serialization efficiency deviation score are processed in sum average to obtain a first synchronization adjustment value, which is used to prompt the configuration manager to increase synchronization efficiency and coverage based on the obtained first synchronization adjustment value; After one hierarchical recovery optimization, it is judged whether the obtained permission synchronization influence index deviation reduction amplitude is within the preset reduction amplitude range in the database. If yes, the hierarchical recovery optimization is completed, otherwise the obtained permission synchronization influence index deviation score and serialization efficiency deviation score are processed in sum average to obtain a second synchronization adjustment value, which is the result of the permission synchronization influence index deviation score and the serialization efficiency score in the database. The second synchronization adjustment value is used to prompt the policy control layer to reduce the synchronization strategy delay caused by too many communication hops based on the obtained second synchronization adjustment value; If within the specified hierarchical recovery optimization times, it is judged whether the re-acquired permission synchronization influence index is not greater than the preset permission synchronization influence index in the database. If yes, the hierarchical recovery optimization is completed and the enterprise AI permission dynamic adjustment is completed, otherwise the hierarchical recovery optimization instruction is sent again. If not within the specified hierarchical recovery optimization times, directly proceed with data compression optimization; The specific steps of the data compression optimization are: based on the data fragment granularity adjustment value obtained after hierarchical recovery optimization to improve the integrity and parallel efficiency of data transmission, the fragment granularity adjustment value represents the result obtained by mapping the reacquired permission synchronization impact indicator in the database after hierarchical recovery optimization; After data compression optimization, determine whether the reacquired permission synchronization impact indicator is greater than the preset permission synchronization impact indicator in the database, if yes, proceed with fragment warning, otherwise complete hierarchical data compression optimization and complete enterprise AI permission dynamic adjustment.

9. A device for dynamically adjusting AI permissions of an enterprise based on organization topology awareness, applying the method for dynamically adjusting AI permissions of an enterprise based on organization topology awareness according to any one of claims 1-8, characterized in that, Comprise: Built-in monitoring timer, embedded transaction log collector, cache controller and interface state monitoring router; The built-in monitoring timer is used to monitor the strategy engine calculation time and topology graph traversal time in real time; The embedded transaction log collector is used to detect inventory synchronization error rate in real time; The cache controller is used to measure the cache recovery interval in real time; The interface state monitoring router is used to monitor the topology network drift rate and network topology node distance in real time.

Images