Full-process intrusion risk detection method and system for enterprise service platform

By mining risk detection points on the enterprise service platform, performing fuzzy logic conversion and multi-protection mode conversion, and combining the system protection characteristics to train the intrusion detection module, the problems of incomplete protection and poor real-time detection in the existing technology are solved, and the full Efficient intrusion risk detection and protection of processes.

CN118487879BActive Publication Date: 2024-11-05BEIJING LIANXUN XINGYE TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410947267.5
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-07-16
Publication Date
2024-11-05
Estimated Expiration
2044-07-16

AI Technical Summary

Technical Problem

Intrusion risk detection on enterprise service platforms has problems such as incomplete protection, poor real-time and accuracy, and traditional methods are difficult to identify emerging unknown threats and ignore internal threats.

Method used

By exploring the overall architecture and operating mode of the enterprise service platform, determine the risk detection point and perform fuzzy logic conversion and multi-variable protection fraud mode conversion, combine the system protection characteristics to train the intrusion detection module, and adopt the platform protection system as the first line of defense, and the intrusion detection module is The second line of defense is to conduct full-process intrusion risk detection.

Benefits of technology

It has achieved comprehensive, real-time and accurate intrusion detection and protection capabilities to ensure the safe and stable operation of the enterprise service platform.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118487879B_ABST
    Figure CN118487879B_ABST
Patent Text Reader

Abstract

The present invention discloses a full-process intrusion risk detection method and system for an enterprise service platform, which relates to data processing related fields. The method includes: determining the platform system based on the overall architecture and operation mode of the enterprise service platform; mining risk detection points based on the intrusion event chain, performing fuzzy logic conversion and multi-dimensional protection deception mode conversion, and determining the detection database; performing active and passive information warfare protection vulnerability analysis and mining on the platform protection system to determine the system protection characteristics; supervising the training of the intrusion detection module; taking the platform protection system as the first line of defense and the intrusion detection module as the second line of defense, performing full-process intrusion risk detection, and generating an intrusion detection log. The method solves the technical problems of incomplete protection and poor real-time and accuracy of intrusion detection in the existing enterprise service platform intrusion risk detection, and achieves the technical effect of providing comprehensive, real-time and accurate intrusion detection and protection capabilities to ensure the safe and stable operation of the enterprise service platform.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to data processing related fields, and in particular to a full-process intrusion risk detection method and system for an enterprise service platform. Background Art

[0002] With the rapid development of information technology, enterprise service platforms are facing more and more cybersecurity threats. These platforms, especially low-code platforms, are becoming key tools for enterprise digital transformation due to their flexibility and ease of use. However, this flexibility also brings new security risks. Therefore, it is crucial to ensure the full process security of enterprise service platforms. Traditional intrusion risk detection methods often only focus on external threats, ignoring internal threats and threats that have already penetrated the system, resulting in incomplete protection. Signature-based detection systems cannot effectively identify emerging and unknown threats. At the same time, regular security audits and vulnerability assessments cannot provide real-time security protection.

[0003] Among the current related technologies, the intrusion risk detection of enterprise service platforms has technical problems such as incomplete protection and poor real-time and accuracy of intrusion detection. Summary of the invention

[0004] This application provides a full-process intrusion risk detection method and system for an enterprise service platform. It mines the overall architecture and operation mode of the enterprise service platform, determines risk detection points, performs fuzzy logic conversion and multi-dimensional protection deception mode conversion on the risk detection points, combines system protection characteristics and detection databases, trains intrusion detection modules, uses the platform protection system as the first line of defense, and the intrusion detection module as the second line of defense, and performs full-process intrusion risk detection and other technical means, thereby achieving the technical effect of providing comprehensive, real-time and accurate intrusion detection and protection capabilities to ensure the safe and stable operation of the enterprise service platform.

[0005] This application provides a full-process intrusion risk detection method for an enterprise service platform, including:

[0006] Based on the overall architecture and operation mode of the enterprise service platform, the platform system is determined, wherein the platform system includes workflow topology and data flow topology, and the enterprise service platform is a low-code platform; based on the platform system, risk detection points based on the intrusion event chain are mined, and the risk detection points are converted into fuzzy logic and multi-dimensional protection deception modes to determine the detection database; active and passive information warfare protection vulnerability analysis and mining are performed on the platform protection system to determine the system protection characteristics; based on the detection database and the system protection characteristics, an intrusion detection module is supervised and trained, and the intrusion detection module is the lower risk defense line of the platform protection system; with the platform protection system as the first defense line and the intrusion detection module as the second defense line, the enterprise service platform is subjected to full-process intrusion risk detection, and an intrusion detection log is generated.

[0007] In a possible implementation, the risk detection point is converted into a fuzzy logic and a multi-dimensional protection deception mode, the detection database is determined, and the following processing is performed:

[0008] Traverse the platform system, retrieve and call the intrusion event chain, wherein the retrieval scope is the platform's historical information warfare and the same-origin platform information warfare; based on the intrusion event chain, with dynamic and static as the guidance, locate the link node of the intrusion detection, and determine the risk detection point, which includes the container's own risk and business execution risk; determine the multiple protection deception mode, perform multiple mode conversion on the risk detection point, and integrate and determine the detection database.

[0009] In a possible implementation, before performing the multivariate mode conversion on the risk detection point, the following processing is performed:

[0010] Each intrusion event chain includes at least one risk detection point; clustering the intrusion event chains to determine a first clustering result, clustering the risk detection points to determine a second clustering result, integrating the first clustering result with the second clustering result to determine multiple clustering clusters; traversing the multiple clustering clusters, performing logical fuzzy conversion, and determining a fuzzy logic library based on the risk detection points.

[0011] In a possible implementation, the fuzzy conversion of logic is performed by executing the following processing:

[0012] Taking into account both universality and detection response rate, the fuzzy relaxation is determined, wherein the fuzzy relaxation corresponds to the multiple clusters one by one; a bidirectional fuzzy conversion unit is established, wherein the bidirectional fuzzy conversion unit includes a fuzzy conversion branch and a defuzzification branch, and the fuzzy conversion method includes at least information sharpening, information blunting and homologous conversion; the fuzzy relaxation is mapped to the multiple clusters, and based on the bidirectional fuzzy conversion unit, logical fuzzy conversion is performed to integrate and determine the fuzzy logic library.

[0013] In a possible implementation, the determining of the system protection characteristic performs the following processing:

[0014] Active information warfare is the spontaneous protection behavior of the platform protection system, and passive information warfare is the passive protection behavior of external intervention; based on risk characteristics and protection characteristics, the first characteristic set and the second characteristic set based on the active information warfare and the passive information warfare are mined; the first characteristic set and the second characteristic set are integrated, and the protection probability is marked to determine the system protection characteristics.

[0015] In a possible implementation, the supervised training intrusion detection module performs the following processing:

[0016] Determine an initial learning rate distribution, wherein the initial learning rate distribution is a balanced distribution; traverse the detection database, determine an overlapping portion based on the first feature set, perform a learning rate decreasing process, determine an overlapping portion based on the second feature set, perform a learning rate increasing process, and determine a learning rate adjustment distribution; based on the learning rate adjustment distribution, calibrate the initial learning rate, and perform supervised learning to obtain the intrusion detection module based on the detection database and the system protection characteristics.

[0017] In a possible implementation, after the intrusion detection log is generated, the following processing is performed:

[0018] Read the intrusion detection log and intrusion record of the predetermined time zone to identify undetected intrusion events; if the undetected intrusion events are not an empty set, locate the risk detection point; based on the risk detection point, learn and update the intrusion detection module.

[0019] This application also provides a full-process intrusion risk detection system for the enterprise service platform, including:

[0020] A platform system determination module, the platform system determination module is used to determine the platform system based on the overall architecture and operation mode of the enterprise service platform, wherein the platform system includes a workflow topology and a data flow topology, and the enterprise service platform belongs to a low-code platform; a detection database determination module, the detection database determination module is used to mine risk detection points based on the intrusion event chain based on the platform system, perform fuzzy logic conversion and multi-dimensional protection deception mode conversion on the risk detection points, and determine the detection database; a system protection characteristic determination module, the system protection characteristic determination module is used to analyze and mine the protection vulnerabilities of the platform protection system for active and passive information warfare, and determine the system protection characteristics; an intrusion detection training module, the intrusion detection training module is used to supervise the training of the intrusion detection module based on the detection database and the system protection characteristics, and the intrusion detection module is the lower risk defense line of the platform protection system; a full-process intrusion risk detection module, the full-process intrusion risk detection module is used to use the platform protection system as the first line of defense and the intrusion detection module as the second line of defense to perform full-process intrusion risk detection on the enterprise service platform and generate an intrusion detection log.

[0021] The full-process intrusion risk detection method and system for the enterprise service platform proposed in this application first determines the platform system based on the overall architecture and operation mode of the enterprise service platform, wherein the platform system includes workflow topology and data flow topology, and the enterprise service platform is a low-code platform. Then, based on the platform system, risk detection points based on the intrusion event chain are mined, and fuzzy logic conversion and multi-dimensional protection deception mode conversion are performed on the risk detection points to determine the detection database. Then, active and passive information warfare protection vulnerability analysis and mining are performed on the platform protection system to determine the system protection characteristics. Then, based on the detection database and system protection characteristics, the intrusion detection module is supervised and trained. The intrusion detection module is the lower risk defense line of the platform protection system. Finally, with the platform protection system as the first line of defense and the intrusion detection module as the second line of defense, the enterprise service platform is subjected to full-process intrusion risk detection and an intrusion detection log is generated, thereby achieving the technical effect of providing comprehensive, real-time and accurate intrusion detection and protection capabilities to ensure the safe and stable operation of the enterprise service platform. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the technical solution of the embodiment of the present invention, the accompanying drawings of the embodiment of the present invention will be briefly introduced below. A flow chart is used in the present application to illustrate the operations performed by the system according to the embodiment of the present application. It should be understood that the preceding or following operations are not necessarily performed accurately in order. On the contrary, various steps can be processed in reverse order or simultaneously as needed. At the same time, other operations can also be added to these processes, or one or more operations can be removed from these processes.

[0023] Figure 1 A flowchart of a full-process intrusion risk detection method for an enterprise service platform provided in an embodiment of the present application.

[0024] Figure 2 A schematic diagram of the structure of a full-process intrusion risk detection system for an enterprise service platform provided in an embodiment of the present application.

[0025] Explanation of the accompanying drawings: platform system determination module 10, detection database determination module 20, system protection characteristic determination module 30, intrusion detection training module 40, full-process intrusion risk detection module 50. DETAILED DESCRIPTION

[0026] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below.

[0027] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application will be further described in detail below in conjunction with the accompanying drawings. The described embodiments should not be regarded as limiting the present application. All other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of this application.

[0028] In the following description, reference is made to "some embodiments", which describe a subset of all possible embodiments, but it is understood that "some embodiments" may be the same subset or different subsets of all possible embodiments, and may be combined with each other without conflict, and the terms "first\second" involved are merely to distinguish similar objects and do not represent a specific ordering of objects. The terms "including" and "having" and any variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or modules that are not clearly listed or inherent to these processes, methods, products, or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those generally understood by technicians in the technical field of this application. The terms used herein are for the purpose of describing the embodiments of the present application only.

[0029] The present application embodiment provides a full-process intrusion risk detection method for an enterprise service platform, such as Figure 1 As shown, the method includes:

[0030] Step S100, based on the overall architecture and operation mode of the enterprise service platform, determine the platform system, wherein the platform system includes workflow topology and data flow topology, and the enterprise service platform belongs to a low-code platform. Specifically, the enterprise service platform is a comprehensive platform that provides various services to enterprises, including application development, process management, data analysis and other functions, and belongs to a low-code platform, wherein the low-code platform refers to an application development platform that enables developers to quickly build applications with less code by providing a graphical interface and pre-built modules, which is used to speed up application development and improve development efficiency. The overall architecture of the enterprise service platform refers to the overall structure and composition of the enterprise service platform at the physical and software levels, wherein the physical structure includes servers, network devices, storage devices, etc., and the software architecture includes the operating system, middleware, database, etc. used. The operation mode of the enterprise service platform refers to the methods and methods followed by the enterprise service platform during operation, including user interaction (such as front-end interface, API interface, etc.), data processing (such as data input, processing, output, etc.), etc. The platform system is determined based on the detailed information of the functional modules and components of the enterprise service platform and the analysis results of key business scenarios and processes. The platform system refers to the structure and functional system of the enterprise service platform as a whole, including workflow topology and data flow topology. The workflow topology is the topological structure that describes the business processing flow within the enterprise service platform, including the dependencies between various business links, data flow direction, etc.; the data flow topology is the topological structure that describes the transmission and processing of data within the enterprise service platform, including the source, destination, processing nodes, etc. of the data.

[0031] Step S200, based on the platform system, the risk detection points based on the intrusion event chain are mined, the risk detection points are converted into fuzzy logic and multi-dimensional protection deception mode, and the detection database is determined. Specifically, based on the platform system, the intrusion event chain is identified and analyzed. The intrusion event chain is a series of intrusion events arranged in chronological order. These intrusion events form an information chain based on valid data information, describing the complete process from detection to successful attack by the attacker. In the intrusion event chain, key risk detection points that pose a threat to platform security are identified, that is, points that may expose platform security vulnerabilities, such as specific system vulnerabilities, weak passwords, unauthorized access, etc. Since intrusion behaviors are often ambiguous and uncertain, a series of fuzzy logic rules are defined based on evaluation criteria such as probability, frequency or behavior patterns, and the risk detection points are converted using the defined fuzzy logic rules to more accurately describe and detect potential intrusion behaviors. The fuzzy logic conversion may include quantitative scoring, weight allocation or probability calculation of risk detection points. Multi-dimensional protection deception is a security strategy that confuses attackers by creating multiple systems or services that appear to be real but are actually deceptive, thereby protecting the real system from attack. According to the characteristics and potential risks of the enterprise service platform, a series of deception patterns are designed by modifying system responses, simulating user behaviors, or generating false data, including false login pages, fake database services, or misleading log files. The designed deception patterns are applied to risk detection points, making it difficult for attackers to distinguish between real systems and deceptive systems. By monitoring and analyzing the behavior of attackers in deceptive environments, intrusion behaviors disguised as normal data flows are identified. Once intrusion behaviors are identified, the emergency response mechanism is immediately activated, and the attackers are traced and analyzed to obtain information about the attackers. The risk detection point information and intrusion behavior information identified after fuzzy logic conversion and multi-dimensional protection deception mode conversion are collected and stored in the detection database, that is, the detection database is a database used to store and manage processed and converted risk detection information.

[0032] In a possible implementation, the risk detection point is converted into fuzzy logic and multiple protection deception mode, and the detection database is determined. Step S200 further includes step S210, traversing the platform system, retrieving and calling the intrusion event chain, wherein the retrieval scope is the platform historical information warfare and the same source platform information warfare. Specifically, the retrieval scope is determined, including the platform's historical information warfare records (all information related to security events that have occurred in the enterprise service platform in the past, including attack events, security vulnerabilities, security logs, etc.) and the same source platform (other platforms with similar business or technical architecture as the enterprise service platform) Information warfare records, according to the hierarchical structure of the platform system, gradually traverse from the bottom to the upper layer, and within the determined retrieval scope, use keyword search, pattern matching and other technologies to retrieve all information related to the intrusion event and organize it into a complete intrusion event chain. Step S220, based on the intrusion event chain, with dynamic and static as the orientation, the link node of the intrusion detection is located to determine the risk detection point, and the risk detection point includes the container's own risk and the business execution risk. Specifically, guided by dynamic (such as behavioral analysis during real-time runtime) and static (such as code review, configuration check), the intrusion event chain is analyzed in detail. In the intrusion event chain, the key link nodes are located. These link nodes are the entrances or key control points for attackers to enter the system. According to the results of the link node analysis, the risk detection points are determined, including container inherent risks and business execution risks. Among them, the container inherent risks refer to the security risks existing in the container technology itself or the container configuration, such as container escape, improper permission configuration, etc.; business execution risks refer to the security risks existing in the business execution process, such as logical vulnerabilities in business processes, sensitive data leakage, etc. Step S230, determine the multi-layer protection deception mode, perform multi-layer mode conversion on the risk detection points, and integrate and determine the detection database. Specifically, the characteristics and patterns of intrusion behaviors disguised as normal data flows are analyzed, including data flow characteristics, behavior patterns, etc. Based on the analysis results of disguised intrusion behaviors, a series of multi-dimensional protection deception patterns are designed. These multi-dimensional protection deception patterns can simulate normal system behaviors, but at the same time contain traps that can identify and trigger intrusion detection mechanisms. The designed traps are deployed and configured at risk detection points, and a monitoring mechanism is established to monitor the triggering of traps and captured intrusion behaviors in real time. Once intrusion behaviors are discovered, the emergency response process is immediately initiated to trace the attackers, isolate the attack source, repair vulnerabilities, etc., and integrate the risk detection point information identified after multi-dimensional pattern conversion into the detection database. This implementation method comprehensively integrates various security risks faced by enterprise service platforms through a wide range of retrieval and precise risk positioning, achieving the technical effect of improving the comprehensiveness of risk analysis.

[0033] In a possible implementation, before the risk detection point is converted into a multivariate mode, step S200 further includes step S240, where each intrusion event chain includes at least one risk detection point. Specifically, for each intrusion event chain, the various steps and stages contained therein are analyzed, and in each step or stage of the intrusion event chain, the position that may become the breakthrough point of the attacker, i.e., the risk detection point, is identified. The risk detection point is the vulnerability, loophole or improper configuration of the system or service, and the identified risk detection point is classified and marked. Step S250, clustering the intrusion event chain to determine a first clustering result, clustering the risk detection point to determine a second clustering result, and fusing the first clustering result with the second clustering result to determine multiple cluster clusters. Specifically, clustering algorithms (such as K-means, hierarchical clustering, etc.) are used to cluster the intrusion event chains, and grouping is performed according to the similarity of the intrusion event chains (such as attack targets, means, impact, etc.). The clustering algorithm is also used to cluster the risk detection points, and the risk detection points are grouped according to factors such as the type, location, and importance of the risk detection points. The clustering results of the intrusion event chain and the clustering results of the risk detection points are merged to form multiple cluster clusters, each cluster cluster represents a group of similar intrusion event chains and risk detection points. Step S260, traverse the multiple cluster clusters, perform logical fuzzy conversion, and determine the fuzzy logic library based on the risk detection points. Specifically, traverse the multiple cluster clusters one by one, and in each cluster cluster, according to the principle of fuzzy logic (a mathematical tool for dealing with uncertainty and ambiguity, which describes and processes imprecise information by defining fuzzy sets and fuzzy operations), the risk detection points are fuzzified, including defining fuzzy sets, fuzzy membership functions, etc., so as to make reasoning and decisions in an uncertain and ambiguous environment, and organize the results after fuzzy conversion into a fuzzy logic library to support the decision-making and actions of the security detection and response system. This implementation method groups similar intrusion event chains and risk detection points through cluster analysis, reduces the processing of redundant information, and improves detection efficiency. By fuzzifying risk detection points, it reduces false positives and missed positives caused by inaccurate or ambiguous information, achieving the technical effect of improving the efficiency and accuracy of security detection and response systems.

[0034] In a possible implementation, the fuzzy conversion of the logic is performed, and step S260 further includes step S261, taking into account the universality and the detection response rate, and determining the fuzzy relaxation, wherein the fuzzy relaxation corresponds to the multiple clusters one by one. Specifically, the intrusion event chain and risk detection point in each cluster are analyzed to determine its characteristics, importance and urgency. Based on the analysis results, the indicators of fuzzy relaxation are set, such as response time, false alarm rate, missed alarm rate, etc., wherein fuzzy relaxation refers to the tolerance or limit on the degree of information fuzziness during fuzzification processing. The higher the fuzzy relaxation, the more fuzzy the information after fuzzification processing; conversely, the more accurate it is. Comprehensively analyze the importance of the clusters and the performance of the detection system, and determine a suitable fuzzy relaxation value for each cluster. Step S262, establish a bidirectional fuzzy conversion unit, the bidirectional fuzzy conversion unit includes a fuzzy conversion branch and a defuzzification branch, and the fuzzy conversion method includes at least information sharpening, information passivation and homologous conversion. Specifically, according to the needs of the safety detection system, different fuzzy conversion methods are designed, such as information sharpening (increasing the certainty of information by reducing its fuzziness), information blunting (reducing its certainty by increasing its fuzziness) and homology conversion (converting information from one fuzzy expression form to another fuzzy expression form). Design a defuzzification algorithm to convert fuzzy information into an accurate and understandable form when needed. Integrate the fuzzy conversion branch and the defuzzification branch into a bidirectional fuzzy conversion unit for flexible fuzzification and defuzzification operations when needed. Step S263, map the fuzzy relaxation with the multiple clusters, perform logical fuzzy conversion based on the bidirectional fuzzy conversion unit, and integrate and determine the fuzzy logic library. Specifically, map each cluster with the corresponding fuzzy relaxation so that the correct fuzzy relaxation can be applied during the logical fuzzy conversion. According to the mapping relationship, use the bidirectional fuzzy conversion unit to perform logical fuzzy conversion on the risk detection points in each cluster, including using the fuzzy conversion branch for fuzzification and using the defuzzification branch for defuzzification, and integrate the results after logical fuzzy conversion into the fuzzy logic library. This implementation method can flexibly adjust the fineness of the fuzzy conversion according to the importance and urgency of different clusters by introducing fuzzy relaxation. Through the bidirectional fuzzy conversion unit, it ensures that fuzzification and defuzzification operations can be performed quickly when needed, achieving the technical effect of optimizing detection efficiency and reducing the load and response time of the detection system.

[0035] Step S300, analyze and mine the active and passive information warfare protection vulnerabilities of the platform protection system to determine the system protection characteristics. Specifically, the platform protection system is the security infrastructure of the enterprise service platform, which is used to monitor and protect against external and internal threats in real time, including components such as firewalls, intrusion prevention systems (IPS), and security event management systems (SIEM). In information warfare, active information warfare refers to the behavior of actively launching attacks or taking defensive measures, and passive information warfare refers to the behavior of formulating strategies or responding to attacks by collecting and analyzing information. Analyze the active and passive information warfare scenarios that the platform protection system may face, and determine the potential risk points and strategies that attackers may take in different scenarios. Use security scanning tools, penetration testing and other means to conduct a comprehensive inspection of each component and module of the platform protection system to identify potential protection vulnerabilities, that is, weaknesses or defects that may be exploited by intruders. Perform a detailed analysis of each identified protection vulnerability, evaluate the security risks and potential losses it may bring, and determine the severity of the vulnerability, the scope of impact, and the priority of repair. Based on the analysis results, determine the overall protection characteristics of the platform protection system, that is, the security protection capabilities and characteristics of the platform protection system, including protection strength, protection range, protection strategy, etc. The system protection characteristics determine the system's resistance and response speed when facing different security threats.

[0036] In a possible implementation, the system protection characteristics are determined, and step S300 further includes step S310, wherein active information warfare is the spontaneous protection behavior of the platform protection system, and passive information warfare is the passive protection behavior with external intervention. Specifically, active information warfare is the spontaneous protection behavior of the platform protection system, and is the system's own protection capability that can actively identify, prevent, and respond to threats, such as active defense systems, automatic updates of security patches, etc. Passive information warfare is a passive protection behavior with external intervention, that is, a security protection behavior that relies on external operations or external system support, and is a response after a threat has occurred or been detected. Step S320, based on risk characteristics and protection characteristics, a first feature set and a second feature set based on the active information warfare and the passive information warfare are mined. Specifically, the risks faced by the enterprise service platform are analyzed, including known and potential threats, attack methods, attack paths, etc., and protection characteristics corresponding to risk characteristics are mined according to the definitions of active information warfare and passive information warfare, wherein the first characteristic set (active information warfare) includes security mechanisms, strategies, tools, etc. within the system, such as firewall rules, intrusion detection system (IDS) rules, data encryption strategies, etc.; the second characteristic set (passive information warfare) includes manual security operation processes, emergency response plans, external security service support, etc. Step S330, integrating the first characteristic set with the second characteristic set, and marking the protection probability to determine the system protection characteristics. Specifically, the first characteristic set and the second characteristic set are integrated to form a comprehensive system protection characteristic set, and each characteristic in the system protection characteristic set is marked with a protection probability according to historical data, expert evaluation or statistical analysis. Based on the integrated characteristic set and protection probability marking, the overall framework and key elements of the system protection characteristics are determined. This implementation method forms a complete and refined protection system by combining active information warfare and passive information warfare, mining and integrating different protection characteristics, and achieving the technical effect of improving the comprehensiveness of system protection characteristic mining.

[0037] Step S400, based on the detection database and the system protection characteristics, supervise the training of the intrusion detection module, the intrusion detection module is the risk defense line of the platform protection system. Specifically, extract feature data related to the intrusion event from the detection database, including network traffic, system logs, user behavior, etc., assign labels to the extracted feature data, and obtain a labeled data set. According to the application scenario and data characteristics, select an intrusion detection algorithm, such as rule-based detection, statistics-based detection, machine learning-based detection, etc., integrate system protection characteristics such as firewall rules, intrusion prevention system (IPS) configuration, etc. into the intrusion detection module, use the labeled data set to train the intrusion detection module, and continuously adjust the model parameters and optimize the algorithm during the training process to improve the detection accuracy of the model and reduce the false alarm rate. The trained intrusion detection module is deployed to the platform protection system as the risk defense line of the platform protection system, which is used to analyze the real-time data stream according to the trained model and algorithm, and identify potential intrusion behaviors.

[0038] In a possible implementation, the supervised training intrusion detection module, step S400 further includes step S410, determining the initial learning rate distribution, wherein the initial learning rate distribution is a balanced distribution. Specifically, an initial learning rate is set for the supervised training process. The learning rate is a hyperparameter that determines the step size of the weight update of the model during the training process. The initial learning rate distribution is a balanced distribution, indicating that all features or parameters have the same learning rate in the initial training stage, that is, the model pays the same attention to all features or parameters at the beginning of training. Step S420, traverse the detection database, determine the overlapping part based on the first feature set, perform a learning rate decreasing process, determine the overlapping part based on the second feature set, perform a learning rate increasing process, and determine the learning rate adjustment distribution. Specifically, the detection database is traversed, and the detection database contains various attack pattern samples. Based on the overlapping part of the first feature set (the feature set of active information warfare), the attack pattern that the system can better protect is identified; based on the overlapping part of the second feature set (the feature set of passive information warfare), the attack pattern that the system protection is relatively weak and needs to rely on passive information warfare (such as manual intervention or external services) to accurately detect is identified. For the overlapping part of the first feature set, that is, the attack pattern that the system can already protect well, the learning rate is decreased, because these patterns are relatively easy to identify and do not require an excessively large learning rate for parameter update; for the overlapping part of the second feature set, that is, the attack pattern that the system protection is relatively weak, the learning rate is increased to help the model learn and adapt to these difficult-to-identify attack patterns faster. Based on the above adjustment, a new learning rate distribution is formed, that is, the learning rate adjustment distribution. Step S430, based on the learning rate adjustment distribution, the initial learning rate is calibrated, and the intrusion detection module is obtained by supervised learning based on the detection database and the system protection characteristics. Specifically, the initial learning rate is calibrated using the learning rate adjustment distribution. Based on the calibrated learning rate and the detection database, the intrusion detection module is trained using the supervised learning method. During the training process, the system's protection characteristics are combined to match the model's learning process with the system's actual protection capabilities. This implementation method ensures that the model focuses more resources and attention on attack modes that pose a greater threat to the system during the training process by dynamically adjusting the learning rate, achieving the technical effect of improving the accuracy and efficiency of the intrusion detection module.

[0039] Step S500, using the platform protection system as the first line of defense and the intrusion detection module as the second line of defense, performs full-process intrusion risk detection on the enterprise service platform and generates an intrusion detection log. Specifically, the full-process intrusion risk detection refers to a comprehensive security detection of the entire business logic and data processing flow of the enterprise service platform, including each link and component from user input to data output, from the front end to the back end. Specifically, the built-in functions and security mechanisms of the platform protection system are utilized to monitor and protect the enterprise service platform in real time. The platform protection system intercepts and blocks suspicious network traffic, malicious files or user behaviors according to predefined rules and policies, and the platform protection system records detected security events and abnormal behaviors in the log. When the platform protection system cannot completely prevent the intrusion behavior or there is a missed report, the intrusion detection module is used as the second line of defense for further detection. The intrusion detection module receives the real-time data stream or log information from the platform protection system, and applies the pre-trained machine learning model or rule library for data analysis. According to the analysis results, it is determined whether there is a potential intrusion behavior, and the detection results are associated and compared with the log information of the platform protection system. If the intrusion detection module detects an intrusion behavior, the corresponding alarm mechanism is triggered and a detailed intrusion detection log is generated. Among them, the platform protection system and the intrusion detection module will generate corresponding log information to record the detected security events, abnormal behaviors, intrusion behaviors, etc. for subsequent analysis. The embodiment of the present application adopts the overall architecture and operation mode of the enterprise service platform to mine, determine the risk detection point, perform fuzzy logic conversion and multivariate protection deception mode conversion on the risk detection point, combine the system protection characteristics and the detection database, train the intrusion detection module, use the platform protection system as the first line of defense, and use the intrusion detection module as the second line of defense to perform full-process intrusion risk detection and other technical means, so as to provide comprehensive, real-time and accurate intrusion detection and protection capabilities, and ensure the technical effect of safe and stable operation of the enterprise service platform.

[0040] In a possible implementation, after the intrusion detection log is generated, step S500 further includes step S510, reading the intrusion detection log and intrusion record in a predetermined time zone to identify undetected intrusion events. Specifically, a time period is determined as a predetermined time zone, such as the last 24 hours, a week, etc., for regularly evaluating the performance of the intrusion detection module, and all intrusion detection logs and known intrusion records in the predetermined time zone are read from the storage system, wherein the intrusion detection log records all detection events, including detected intrusion events and undetected normal or abnormal events; the intrusion record is a known intrusion event record from the security incident response team or the security operation center. The detection events in the intrusion detection log are compared with the known intrusion records to find the intrusion events not recorded in the intrusion detection log, that is, the undetected intrusion events. Step S520, if the undetected intrusion events are not an empty set, locate the risk detection point. Specifically, check whether the set of undetected intrusion events identified in step S510 is an empty set. If the set of undetected intrusion events is not empty, analyze these intrusion events to determine why the intrusion detection module failed to detect these intrusion events, and find the potential reasons or configuration points that caused the intrusion detection module to fail to detect known intrusion events, that is, the risk detection points. Step S530, based on the risk detection points, learn and update the intrusion detection module. Specifically, according to the determined risk detection points, the intrusion detection module is retrained using a new data set, and the model is adjusted and optimized to improve the detection capability of the intrusion detection module. This implementation method achieves the technical effect of improving the detection accuracy and adaptability of the intrusion detection module by regularly checking and identifying undetected intrusion events and adjusting and optimizing the intrusion detection module.

[0041] In the above, refer to Figure 1 The whole process intrusion risk detection method of the enterprise service platform according to the embodiment of the present invention is described in detail. Figure 2 A full-process intrusion risk detection system for an enterprise service platform according to an embodiment of the present invention is described.

[0042] The full-process intrusion risk detection system of the enterprise service platform according to the embodiment of the present invention is used to solve the technical problems of incomplete protection, poor real-time and accuracy of intrusion detection in existing enterprise service platform intrusion risk detection, and achieve the technical effect of providing comprehensive, real-time and accurate intrusion detection and protection capabilities to ensure the safe and stable operation of the enterprise service platform. The full-process intrusion risk detection system of the enterprise service platform includes: a platform system determination module 10, a detection database determination module 20, a system protection feature determination module 30, an intrusion detection training module 40, and a full-process intrusion risk detection module 50.

[0043] The platform system determination module 10 is used to determine the platform system based on the overall architecture and operation mode of the enterprise service platform, wherein the platform system includes a workflow topology and a data flow topology, and the enterprise service platform belongs to a low-code platform; the detection database determination module 20 is used to mine risk detection points based on the intrusion event chain based on the platform system, perform fuzzy logic conversion and multi-dimensional protection deception mode conversion on the risk detection points, and determine the detection database; the system protection characteristic determination module 30 is used to analyze and mine the protection vulnerabilities of the platform protection system for active and passive information warfare, and determine the system protection characteristics; the intrusion detection training module 40 is used to supervise the training of the intrusion detection module based on the detection database and the system protection characteristics, and the intrusion detection module is the lower risk defense line of the platform protection system; the full-process intrusion risk detection module 50 is used to use the platform protection system as the first line of defense and the intrusion detection module as the second line of defense to perform full-process intrusion risk detection on the enterprise service platform and generate an intrusion detection log.

[0044] The specific configuration of the detection database determination module 20 will be described in detail below. As described above, the risk detection point is subjected to fuzzy logic conversion and multi-dimensional protection deception mode conversion to determine the detection database. The detection database determination module 20 may further include: an intrusion event chain calling unit for traversing the platform system, retrieving and calling the intrusion event chain, wherein the retrieval scope is the platform historical information war and the same source platform information war; a risk detection point determination unit for locating the link node of the intrusion detection based on the intrusion event chain, guided by dynamic and static, to determine the risk detection point, wherein the risk detection point includes the container's own risk and the business execution risk; a multi-dimensional mode conversion unit for determining the multi-dimensional protection deception mode, performing multi-dimensional mode conversion on the risk detection point, and integrating and determining the detection database.

[0045] Among them, before the risk detection point is converted into a multivariate mode, the detection database determination module 20 may further include: an intrusion event chain determination unit for each intrusion event chain including at least one risk detection point; a clustering unit for clustering the intrusion event chain to determine a first clustering result, clustering the risk detection point to determine a second clustering result, fusing the first clustering result with the second clustering result to determine a plurality of clustering clusters; a fuzzy logic library determination unit for traversing the plurality of clustering clusters, performing logical fuzzy conversion, and determining a fuzzy logic library based on the risk detection point.

[0046] Among them, the fuzzy logic library determination unit for performing logical fuzzy conversion may further include: a fuzzy relaxation determination subunit for determining the fuzzy relaxation taking into account both universality and detection response rate, wherein the fuzzy relaxation corresponds one-to-one to the multiple clusters; a bidirectional fuzzy conversion establishment subunit for establishing a bidirectional fuzzy conversion unit, the bidirectional fuzzy conversion unit includes a fuzzy conversion branch and a defuzzification branch, and the fuzzy conversion method includes at least information sharpening, information blunting and homologous conversion; the fuzzy conversion subunit is used to map the fuzzy relaxation and the multiple clusters, and based on the bidirectional fuzzy conversion unit, perform logical fuzzy conversion to integrate and determine the fuzzy logic library.

[0047] The specific configuration of the system protection characteristic determination module 30 will be described in detail below. As described above, to determine the system protection characteristics, the system protection characteristic determination module 30 may further include: an information warfare definition unit for active information warfare as the spontaneous protection behavior of the platform protection system, and passive information warfare as the passive protection behavior of external intervention; a characteristic mining unit for mining the first characteristic set and the second characteristic set based on the active information warfare and the passive information warfare based on the risk characteristics and the protection characteristics; a characteristic integration unit for integrating the first characteristic set and the second characteristic set, and performing protection probability marking to determine the system protection characteristics.

[0048] The specific configuration of the intrusion detection training module 40 will be described in detail below. As described above, the supervised training intrusion detection module, the intrusion detection training module 40 may further include: an initial learning rate distribution determination unit for determining an initial learning rate distribution, wherein the initial learning rate distribution is a balanced distribution; a learning rate adjustment distribution determination unit for traversing the detection database, determining an overlapping portion based on the first feature set, performing a learning rate decreasing process, determining an overlapping portion based on the second feature set, performing a learning rate increasing process, and determining a learning rate adjustment distribution; a supervised learning unit for calibrating the initial learning rate based on the learning rate adjustment distribution, and performing supervised learning to obtain the intrusion detection module based on the detection database and the system protection characteristics.

[0049] The specific configuration of the full-process intrusion risk detection module 50 will be described in detail below. As described above, after generating the intrusion detection log, the full-process intrusion risk detection module 50 may further include: an undetected intrusion event identification unit for reading the intrusion detection log and the intrusion record of the predetermined time zone to identify the undetected intrusion event; a risk detection point positioning unit for locating the risk detection point if the undetected intrusion event is not an empty set; and a learning update unit for learning and updating the intrusion detection module based on the risk detection point.

[0050] The full-process intrusion risk detection system for the enterprise service platform provided in the embodiment of the present invention can execute the full-process intrusion risk detection method for the enterprise service platform provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0051] Although the present application makes various references to certain modules in the system according to the embodiments of the present application, any number of different modules may be used and run on the user terminal and / or server, and the various units and modules included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of the functional units are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of the present invention.

[0052] The above specific implementation manner does not constitute a limitation to the protection scope of the present application. It should be understood by those skilled in the art that various modifications, combinations and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present application should be included in the protection scope of the present application. In some cases, the actions or steps recorded in the present application can be performed in an order different from that in the embodiment and can still achieve the desired results. In addition, the process depicted in the accompanying drawings does not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

Claims

1. A full-process intrusion risk detection method for an enterprise service platform, characterized in that: The method comprises: Based on the overall architecture and operation mode of the enterprise service platform, determine the platform system, wherein the platform system includes a workflow topology and a data flow topology, and the enterprise service platform is a low-code platform; Based on the platform system, risk detection points based on the intrusion event chain are mined, fuzzy logic conversion and multi-dimensional protection deception mode conversion are performed on the risk detection points, and a detection database is determined; Conduct active and passive information warfare vulnerability analysis on the platform protection system to determine the system protection characteristics; Based on the detection database and the system protection characteristics, supervise and train an intrusion detection module, which is a risk defense line under the platform protection system; Using the platform protection system as the first line of defense and the intrusion detection module as the second line of defense, the enterprise service platform is subjected to full-process intrusion risk detection, and an intrusion detection log is generated; The risk detection point is subjected to fuzzy logic conversion and multi-dimensional protection deception mode conversion to determine the detection database, including: Traverse the platform system, retrieve and call the intrusion event chain, wherein the retrieval scope is the platform's historical information warfare and the same-source platform information warfare; Based on the intrusion event chain, the intrusion detection link node is located with dynamic and static orientation to determine the risk detection point, which includes the container's own risk and the business execution risk; Determine multiple protection deception modes, perform multiple mode conversion on the risk detection points, and integrate and determine the detection database; Before the risk detection point is converted into a multivariate mode, the following steps are included: Each intrusion event chain includes at least one risk detection point; Clustering the intrusion event chain to determine a first clustering result, clustering the risk detection points to determine a second clustering result, fusing the first clustering result and the second clustering result to determine a plurality of cluster clusters; The plurality of clusters are traversed, logical fuzzy conversion is performed, and a fuzzy logic library based on the risk detection point is determined.

2. The full-process intrusion risk detection method for an enterprise service platform as claimed in claim 1, characterized in that: The fuzzy conversion of logic includes: Taking into account the universality and the detection response rate, determining the fuzzy relaxation, wherein the fuzzy relaxation corresponds to the multiple clusters one by one; Establishing a bidirectional fuzzy conversion unit, the bidirectional fuzzy conversion unit includes a fuzzy conversion branch and a defuzzification branch, and the fuzzy conversion mode includes at least information sharpening, information passivation and homology conversion; The fuzzy relaxation is mapped to the plurality of clusters, and based on the bidirectional fuzzy conversion unit, a logical fuzzy conversion is performed to integrate and determine the fuzzy logic library.

3. The full-process intrusion risk detection method for an enterprise service platform as claimed in claim 1, characterized in that: The determining of the system protection characteristics comprises: Active information warfare is the spontaneous protection behavior of the platform protection system, and passive information warfare is the passive protection behavior of external intervention; Based on the risk characteristics and the protection characteristics, a first characteristic set and a second characteristic set based on the active information warfare and the passive information warfare are mined; The first characteristic set and the second characteristic set are integrated, and protection probability marking is performed to determine the system protection characteristic.

4. The full-process intrusion risk detection method for an enterprise service platform as claimed in claim 3, characterized in that: The supervised training intrusion detection module comprises: Determining an initial learning rate distribution, wherein the initial learning rate distribution is a balanced distribution; Traversing the detection database, determining the overlapping part based on the first characteristic set, performing a learning rate decreasing process, determining the overlapping part based on the second characteristic set, performing a learning rate increasing process, and determining a learning rate adjustment distribution; Based on the learning rate adjustment distribution, the initial learning rate is calibrated, and supervised learning is performed to obtain the intrusion detection module based on the detection database and the system protection characteristics.

5. The full-process intrusion risk detection method for an enterprise service platform as claimed in claim 1, characterized in that: After the intrusion detection log is generated, the following steps are included: Read intrusion detection logs and intrusion records in a predetermined time zone to identify undetected intrusion events; If the undetected intrusion events are not an empty set, locating the risk detection point; Based on the risk detection point, the intrusion detection module is learned and updated.

6. The full-process intrusion risk detection system of the enterprise service platform is characterized by: The system is used to implement the full-process intrusion risk detection method of the enterprise service platform according to any one of claims 1 to 5, and the system includes: A platform system determination module, the platform system determination module is used to determine the platform system based on the overall architecture and operation mode of the enterprise service platform, wherein the platform system includes a workflow topology and a data flow topology, and the enterprise service platform is a low-code platform; A detection database determination module, the detection database determination module is used to mine risk detection points based on the intrusion event chain based on the platform system, perform fuzzy logic conversion and multi-dimensional protection deception mode conversion on the risk detection points, and determine the detection database; A system protection characteristic determination module, which is used to analyze and mine the protection vulnerabilities of the platform protection system for active and passive information warfare, and determine the system protection characteristics; An intrusion detection training module, the intrusion detection training module is used to supervise and train an intrusion detection module based on the detection database and the system protection characteristics, the intrusion detection module is a risk defense line under the platform protection system; A full-process intrusion risk detection module is used to perform full-process intrusion risk detection on the enterprise service platform with the platform protection system as the first line of defense and the intrusion detection module as the second line of defense, and generate an intrusion detection log.