Abnormal access behavior analysis method, electronic equipment, medium and program product

By constructing a baseline model and analyzing access behavior data, identifying abnormal access behaviors and generating access relationship maps, the problem of high false alarm rates and missed alarm rates in baseline detection technology is solved, and the security and protection capabilities of the network system are improved.

CN119066574AActive Publication Date: 2024-12-03BEIJING TOPSEC NETWORK SECURITY TECH +2

Patent Information

Application Number
CN202411120363.9
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-15
Publication Date
2024-12-03
Estimated Expiration
2044-08-15

AI Technical Summary

Technical Problem

Baseline-based attack behavior detection technology has problems with high false alarm rates and low false alarm rates when identifying abnormal access behaviors, making it difficult to adapt to complex cyber threat scenarios.

Method used

By constructing a baseline model, the operator model in the operator library analyzes access behavior data, identifys abnormal access behaviors, and generates an access relationship map, combining individual and group behavior comparison, cluster analysis and rule learning to improve identification accuracy.

Benefits of technology

It effectively reduces the false alarm rate and missed alarm rate of abnormal access behavior, improves the security and protection capabilities of the network system, and provides an important reference for formulating and optimizing security strategies.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119066574A_ABST
    Figure CN119066574A_ABST
Patent Text Reader

Abstract

The invention provides an abnormal access behavior analysis method, which comprises the following steps that: operator models in an operator library are utilized to construct a baseline model, and the baseline model represents a baseline rule used for describing a normal behavior mode of an access behavior; obtaining access behavior data; analyzing the access behavior data based on the baseline model, and identifying an abnormal access behavior; and generating an access relation graph according to the access behavior data and an identification result of the abnormal access behavior. The disclosure also provides an electronic device, a computer readable medium, and a computer program product. According to the invention, the special analysis of the abnormal access behavior is realized, and the security and protection capability of the network system are improved through the access atlas analysis.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present disclosure relates to the technical field of data analysis, and in particular to an analysis method for abnormal access behavior, an electronic device, a computer-readable medium, and a computer program product. Background Art

[0002] Traditional security strategies usually rely on rule engines and firewalls to prevent malicious attacks. However, this approach has some limitations, such as difficulty in adapting to new threat patterns and easy to be bypassed. Therefore, baseline-based attack behavior detection technology has emerged. Baseline-based attack behavior detection technology determines normal user behavior based on the analysis of user historical behavior data, which serves as the basis for judging whether user behavior is attack behavior, and can effectively improve security protection capabilities.

[0003] However, baseline-based attack behavior detection technology still has defects in certain application scenarios. Summary of the invention

[0004] The embodiments of the present disclosure provide a method for analyzing abnormal access behavior, an electronic device, a computer-readable medium, and a computer program product.

[0005] In a first aspect, an embodiment of the present disclosure provides a method for analyzing abnormal access behavior, comprising: constructing a baseline model using an operator model in an operator library, wherein the baseline model represents a baseline rule for describing a normal behavior pattern of access behavior; obtaining access behavior data; analyzing the access behavior data based on the baseline model to identify abnormal access behavior; and generating an access relationship map based on the access behavior data and the identification results of the abnormal access behavior.

[0006] In some embodiments, the access behavior data is analyzed based on the baseline model to identify abnormal access behavior, including: analyzing the degree of deviation of the access behavior data from the baseline rule according to the baseline model, identifying abnormal access data that characterizes the abnormal access behavior, and obtaining an initial identification result; generalizing process features based on the access behavior data and the abnormal access data to obtain an identification result of the abnormal access behavior.

[0007] In some embodiments, process feature generalization is performed based on the access behavior data and the abnormal access data to obtain an identification result of the abnormal access behavior, including: analyzing the difference between the individual access behavior and the group access behavior represented by the initial identification result based on the initial identification result, and outputting a first identification result of the abnormal access behavior, wherein the first identification result conforms to a normal individual and group behavior baseline; and / or clustering the access behavior represented by the access behavior data based on the access behavior data and the initial identification result, aggregating and outputting a second identification result of the abnormal access behavior; and / or evaluating the initial identification result according to a preset rule, and outputting a third identification result of the abnormal access behavior, wherein the third identification result satisfies the preset rule.

[0008] In some embodiments, an access relationship map is generated based on the access behavior data and the identification results of abnormal access behaviors, including: performing rule mapping and data association on the access behavior data to form a data set; analyzing the data set through multiple dimensions to form analysis results of multiple dimensions based on the first identification result, and / or the second identification result, and / or the third identification result; performing association analysis on the analysis results of the multiple dimensions to generate the access relationship map, wherein the access relationship map includes an asset portrait map and a personnel portrait map; performing graph analysis on the asset portrait map and the personnel portrait map to identify abnormal access behaviors and threat events.

[0009] In some embodiments, the asset portrait map and the personnel portrait map are subjected to a graph analysis to identify abnormal access behaviors and threat events, including: obtaining and displaying the user's portrait information from the access relationship map based on the user's keywords obtained from the search engine; calculating the degree of deviation of the user's access behavior from the baseline rule with a preset time length as the statistical granularity, identifying and displaying abnormal access behaviors and threat events with a degree of deviation exceeding the upper and lower limits.

[0010] In some embodiments, a baseline model is constructed using operator models in an operator library, including: obtaining multiple operator models from the operator library to construct an initial baseline model; training the initial baseline model using training samples; and correcting the concatenation of operator models in the initial baseline model according to training results and business scenarios to obtain the baseline model.

[0011] In some embodiments, constructing a baseline model using an operator model in an operator library includes: constructing the baseline model through a model construction function; wherein the model construction function includes an operator management unit, a program management unit, a model management unit, a task management unit, a service management unit, and a shortcut management unit; the model construction function also includes a visualization interface, and constructing the baseline model through the model construction function includes: in response to an operation instruction obtained through the visualization interface, performing the following operations: executing, through the operator management unit, an operation of managing the operator model, such as acquiring the operator model, customizing the operator model, etc.; executing, through the program management unit, an operation of managing the program for training the initial baseline model; executing, through the model management unit, an operation of managing the baseline model, such as operator model concatenation; executing, through the task management unit, a task management operation; executing, through the service management unit, a service management operation; and executing a shortcut management operation through the shortcut management unit.

[0012] In some embodiments, obtaining access behavior data includes: obtaining the access behavior data through multiple receiving servers deployed in a distributed manner, wherein different receiving servers correspond to different log sources; converting the access behavior data collected from multiple receiving servers into a standard data format; monitoring the data collection status of each log source to determine the health status of the log source; and monitoring each receiving server to determine the health status of the receiving server.

[0013] In a second aspect, an embodiment of the present disclosure provides an electronic device, comprising: one or more processors; a memory on which one or more programs are stored, and when the one or more programs are executed by the one or more processors, the one or more processors implement the method for analyzing abnormal access behavior described in the first aspect of the embodiment of the present disclosure.

[0014] In a third aspect, an embodiment of the present disclosure provides a computer-readable medium having a computer program stored thereon, and when the computer program is executed by a processor, the method for analyzing abnormal access behavior described in the first aspect of the embodiment of the present disclosure is implemented.

[0015] In a fourth aspect, an embodiment of the present disclosure provides a computer program product, including a computer program or instructions, which, when executed by a processor, implements the method for analyzing abnormal access behavior described in the first aspect of the embodiment of the present disclosure.

[0016] The disclosed embodiment builds a baseline model to establish baseline rules that describe normal behavior patterns of access behaviors, and then uses the baseline model to analyze access behavior data. On this basis, an access relationship map is generated, which can be used specifically for identifying abnormal access behaviors, thereby achieving specialized analysis and detection of abnormal access behaviors based on the baseline, reducing the false positive and false negative rates of abnormal access behaviors, and providing an important reference for formulating and optimizing security policies, thereby improving the security and protection capabilities of network systems. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 is a flow chart of a method for analyzing abnormal access behavior in an embodiment of the present disclosure;

[0018] Figure 2 is a schematic diagram of the architecture of a model building function in an embodiment of the present disclosure;

[0019] Figure 3 is a schematic diagram of the architecture of a data collection and management function in an embodiment of the present disclosure;

[0020] Figure 4 is a schematic diagram of the composition of an electronic device in an embodiment of the present disclosure;

[0021] Figure 5 It is a schematic diagram of an architecture for analyzing abnormal access behavior;

[0022] Figure 6 This is another architectural diagram for analyzing abnormal access behavior;

[0023] Figure 7 It is a schematic diagram of the architecture of graph analysis. DETAILED DESCRIPTION

[0024] In order to enable those skilled in the art to better understand the technical solution of the present disclosure, the technical solution of the present disclosure is described in detail below with reference to the accompanying drawings.

[0025] Example embodiments will be described more fully below with reference to the accompanying drawings, but the example embodiments may be embodied in different forms and should not be construed as limited to the embodiments set forth herein. On the contrary, the purpose of providing these embodiments is to make the present disclosure thorough and complete and to enable those skilled in the art to fully understand the scope of the present disclosure.

[0026] In the absence of conflict, the various embodiments of the present disclosure and the various features therein may be combined with each other.

[0027] As used herein, the term "and / or" includes any and all combinations of one or more of the associated listed items.

[0028] The terms used herein are only used to describe specific embodiments and are not intended to limit the present disclosure. As used herein, the singular forms "a", "an" and "the" are also intended to include the plural forms, unless the context clearly indicates otherwise. It will also be understood that when the terms "comprising" and / or "made of" are used in this specification, the presence of the features, wholes, steps, operations, elements and / or components is specified, but the presence or addition of one or more other features, wholes, steps, operations, elements, components and / or groups thereof is not excluded.

[0029] Unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by those of ordinary skill in the art. It will also be understood that terms such as those defined in commonly used dictionaries should be interpreted as having a meaning consistent with their meaning in the context of the relevant art and the present disclosure, and will not be interpreted as having an idealized or overly formal meaning unless explicitly defined as such herein.

[0030] Baseline-based attack behavior detection technology collects a large amount of historical data to determine the normal behavior of users and converts it into a series of baseline rules. When user behavior deviates from these baseline rules, it indicates that there may be an abnormal situation that requires further investigation and processing.

[0031] In daily network security management activities, there are often abnormal access behaviors such as unauthorized access, abnormal login behavior, abuse of authority, abnormal data access, etc. The disclosed embodiment aims to detect the above abnormal access behaviors based on the baseline to achieve special detection and analysis of abnormal access behaviors.

[0032] First, as Figure 1 As shown, the embodiment of the present disclosure provides a method for analyzing abnormal access behavior, including:

[0033] S1. constructing a baseline model using an operator model in an operator library, wherein the baseline model represents a baseline rule for describing a normal behavior pattern of an access behavior;

[0034] S2. Obtain access behavior data;

[0035] S3, analyzing the access behavior data based on the baseline model to identify abnormal access behavior;

[0036] S4. Generate an access relationship graph based on the access behavior data and the identification results of abnormal access behaviors.

[0037] In the disclosed embodiment, an operator library is pre-configured, and a large number of operator models are built in the operator library. An operator model is an operator that is pre-defined and trained based on a machine learning algorithm, and can be used to perform specific computing tasks such as data processing, visualization, and machine learning; an operator model can be a single operator, or a process or network composed of multiple operators; an operator model supports optimizing its performance through training.

[0038] The present disclosure does not specifically limit the access behavior data. In some embodiments, the access behavior data includes access behavior data generated by various business systems, user authentication, terminals, threat detection, and traffic auditing devices.

[0039] like Figure 5 As shown, the embodiments of the present disclosure do not specifically limit abnormal access behaviors. In some embodiments, abnormal access behaviors include but are not limited to:

[0040] Unusual login attempts, including frequent login failures, logins from different geographic locations or unusual times;

[0041] Unauthorized file or directory access, where a user attempts to access a file or directory that they do not have permission to access;

[0042] unusual data transfers, large amounts of data to or from unusual geographic locations, external systems, or at unusual times;

[0043] Abnormal privilege escalation, where a user attempts to elevate their privileges or access restricted resources;

[0044] Abnormal network activities, including abnormal network traffic, communications with malicious IP addresses, unauthorized remote access, etc.

[0045] Data theft, including attempts by users to access sensitive data or files, or to copy or download files without authorization;

[0046] Malware activity is the discovery of malware or malicious code on a user's system or network.

[0047] In the embodiment of the present disclosure, the access relationship map represents the relationship map between users and resources. By analyzing the access relationship map, the behavior patterns and threat events of abnormal access behaviors can be discovered.

[0048] The disclosed embodiment builds a baseline model to establish baseline rules that describe normal behavior patterns of access behaviors, and then uses the baseline model to analyze access behavior data. On this basis, an access relationship map is generated, which can be used specifically for identifying abnormal access behaviors, thereby achieving specialized analysis and detection of abnormal access behaviors based on the baseline, reducing the false positive and false negative rates of abnormal access behaviors, and providing an important reference for formulating and optimizing security policies, thereby improving the security and protection capabilities of network systems.

[0049] In the embodiment of the present disclosure, in the process of analyzing access behavior data based on the baseline model to identify abnormal access behavior, the abnormal access behavior can be identified only through the baseline model; or it can be identified first through the baseline model and then the process features are generalized to improve the accuracy of the identification.

[0050] Accordingly, in some embodiments, analyzing the access behavior data based on the baseline model to identify abnormal access behavior includes:

[0051] Analyzing the degree of deviation between the access behavior data and the baseline rule according to the baseline model, identifying abnormal access data representing abnormal access behavior, and obtaining an initial identification result;

[0052] Process feature generalization is performed according to the access behavior data and the abnormal access data to obtain an identification result of the abnormal access behavior.

[0053] In the embodiment of the present disclosure, in the baseline model, if the degree of deviation between the access behavior data and the baseline rules is between the preset upper and lower limits, it means that the access behavior data is data generated by normal access behavior; if the degree of deviation between the access behavior data and the baseline rules exceeds the upper or lower limit, it means that the access behavior data is data generated by abnormal access behavior.

[0054] In the disclosed embodiment, generalizing process features refers to generalizing the results obtained from the baseline model analysis to a larger range or more common situations, thereby improving the accuracy of identifying abnormal access behaviors.

[0055] In some embodiments, process feature generalization is performed using at least one of individual-group comparison, cluster analysis, and rule learning.

[0056] Individual-group comparison refers to performing an individual-group comparison algorithm analysis on the analysis results obtained by the baseline model based on the normal individual-group behavior baseline; if the analysis results obtained by the baseline model fall within the normal individual-group behavior baseline range, the recognition result is output; if it falls within the abnormal individual-group behavior baseline range, the baseline model is optimized, and the analysis results of the optimized baseline model are again analyzed by the individual-group comparison algorithm, and the recognition result is output.

[0057] Cluster analysis refers to dividing objects in a data set into different groups (clusters) so that objects in the same group have a higher similarity and objects in different groups have a lower similarity, by aggregating abnormal access behaviors and outputting results. In some embodiments, the K-means clustering algorithm is used in cluster analysis.

[0058] Rule learning refers to the rule evaluation of data through cross-validation, accuracy verification, and recall verification through rule learning algorithms. If the rule requirements are met, the results are output; if not, the rules are optimized, re-learned, and the results are output.

[0059] Accordingly, in some embodiments, generalizing process features according to the access behavior data and the abnormal access data to obtain an identification result of the abnormal access behavior includes:

[0060] Analyzing the difference between the individual access behavior and the group access behavior represented by the initial recognition result according to the initial recognition result, and outputting a first recognition result of the abnormal access behavior, wherein the first recognition result conforms to a normal individual and group behavior baseline; and / or

[0061] Clustering the access behaviors represented by the access behavior data according to the access behavior data and the initial identification result, aggregating and outputting a second identification result of the abnormal access behavior; and / or

[0062] The initial recognition result is evaluated according to a preset rule, and a third recognition result of the abnormal access behavior is output, wherein the third recognition result satisfies the preset rule.

[0063] In some embodiments, generating an access relationship graph based on the access behavior data and the identification result of abnormal access behavior includes:

[0064] Performing rule mapping and data association on the access behavior data to form a data set;

[0065] Analyze the data set in multiple dimensions according to the first recognition result, and / or the second recognition result, and / or the third recognition result to form analysis results in multiple dimensions;

[0066] Performing correlation analysis on the analysis results of the multiple dimensions to generate the access relationship map, wherein the access relationship map includes an asset portrait map and a personnel portrait map;

[0067] Perform graph analysis on the asset portrait graph and the personnel portrait graph to identify abnormal access behaviors and threat events.

[0068] In some embodiments, the asset portrait map and the personnel portrait map are analyzed to identify abnormal access behaviors and threat events, including:

[0069] According to the user's keywords obtained from the search engine, the user's portrait information is obtained from the access relationship graph and displayed;

[0070] The deviation degree of the user's access behavior from the baseline rule is counted with the preset time length as the statistical granularity, and abnormal access behavior and threat events with the deviation degree exceeding the upper and lower limits are identified and displayed.

[0071] In some embodiments, access graph analysis discovers abnormal patterns and potential threats of access violations and abnormal behaviors through two dimensions: personnel and assets. The disclosed embodiment also provides a search engine, which can query and display user portrait information by entering keywords of users (personnel or assets), for example, including:

[0072] Basic information of personnel / assets, including avatar, work number, department, email, tag, asset, IP, account, active time, etc.; operations of adding or removing personnel from the observation list; operations of adjusting the scoring mode;

[0073] Risk score, including the risk score for the day and the risk score for the historical period;

[0074] Risk overview: Use charts to display the distribution of threat events in the recent historical period, display the risk tags involved in threat events, display the threat scenarios involved in threat events and their proportions, display the top N threat event information, display detailed information related to threat events, and display the association relationship (topology) between various types of entities and threats;

[0075] Personnel / asset portraits use charts to display the top N personnel tags involved in risk scores or threats, display risk information detected in the recent historical period, display information on various threat events associated with the recent historical period, group information, daily activity in the recent historical period, commonly used assets in the recent historical period, frequently accessed ports in the recent historical period, frequently accessed domain names in the recent historical period, and the top N threats in the recent historical period.

[0076] In some embodiments, when identifying abnormal access behaviors and threat events, the behavior baseline and outlier abnormal information of personnel or assets are displayed with a statistical granularity of 1 hour based on the Airflow offline model and scheduling training data. When the true value is not between the upper and lower limits, a threat event is generated.

[0077] The embodiments of the present disclosure do not specifically limit how to construct a baseline model using operator models in an operator library.

[0078] In some embodiments, constructing a baseline model using an operator model in an operator library includes:

[0079] Acquire multiple operator models from the operator library to build an initial baseline model;

[0080] Training the initial baseline model using training samples;

[0081] According to the training results and business scenarios, the concatenation of operator models in the initial baseline model is corrected to obtain the baseline model.

[0082] In some embodiments, constructing a baseline model using an operator model in an operator library includes:

[0083] Building the baseline model through a model building function;

[0084] The model building function includes an operator management unit, a program management unit, a model management unit, a task management unit, a service management unit, and a shortcut management unit;

[0085] The model building function also includes a visualization interface, and the baseline model is built through the model building function, including:

[0086] In response to the operation instruction obtained through the visual interface, the following operations are performed:

[0087] The operator management unit performs operations such as acquiring the operator model, customizing the operator model, etc. to manage the operator model;

[0088] An operation of managing a program for training the initial baseline model is performed by the program management unit;

[0089] The model management unit executes operations such as operator model concatenation to manage the baseline model;

[0090] Execute task management operations through the task management unit;

[0091] Executing service management operations through the service management unit;

[0092] The shortcut management operation is performed through the shortcut management unit.

[0093] In the embodiment of the present disclosure, the user is supported to generate operation instructions by performing operations such as dragging and dropping in the visual interface.

[0094] Figure 2 FIG. 1 is a schematic diagram of an architecture of a model building function in an embodiment of the present disclosure. Figure 2As shown, the model building functional unit includes operator management unit, program management unit, model management unit, task management unit, service management unit, and shortcut management unit. The operator management unit includes management of input and output operators, machine learning operators, and visualization operators; the program management unit includes management of supervised learning modeling, unsupervised learning modeling, and deep learning modeling; model management includes variable management, fragment management, and group management; task management includes management of immediate tasks, scheduled tasks, and model editing; service management includes management of retrieval engines, message queues, and JDBC; shortcut management includes management of adding data, selecting models, and adding tasks.

[0095] In some embodiments, the model building function has good scalability, supports custom operators and applies them to the system; the operators can directly operate the database and can analyze data from any source.

[0096] The disclosed embodiment does not specifically limit how to obtain access behavior data.

[0097] In some embodiments, obtaining access behavior data includes:

[0098] Acquiring the access behavior data through a plurality of receiving servers deployed in a distributed manner, wherein different receiving servers correspond to different log sources;

[0099] Converting the access behavior data collected from the plurality of receiving servers into a standard data format;

[0100] Monitor the data collection status of each log source to determine the health status of the log source;

[0101] Each of the receiving servers is monitored to determine the health status of the receiving server.

[0102] In the embodiment of the present disclosure, a data collection function is provided. Figure 3 As shown, it includes the management of receivers, parsing templates, log sources and other links, can monitor the data collection of specific log sources and understand the health status of the log sources; it can also realize distributed data collection through distributed deployment of multiple receiving servers, and can monitor the receiving servers to understand the health status of the receiving servers.

[0103] Second, refer to Figure 4 , an embodiment of the present disclosure provides an electronic device, comprising:

[0104] One or more processors 101;

[0105] A memory 102, on which one or more programs are stored. When the one or more programs are executed by one or more processors, the one or more processors implement the abnormal access behavior analysis method described in the first aspect of the embodiment of the present disclosure;

[0106] One or more I / O interfaces 103 are connected between the processor and the memory and are configured to implement information exchange between the processor and the memory.

[0107] Among them, the processor 101 is a device with data processing capabilities, including but not limited to a central processing unit (CPU) and the like; the memory 102 is a device with data storage capabilities, including but not limited to random access memory (RAM, more specifically SDRAM, DDR, etc.), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory (FLASH); the I / O interface (read-write interface) 103 is connected between the processor 101 and the memory 102, and can realize information interaction between the processor 101 and the memory 102, including but not limited to a data bus (Bus) and the like.

[0108] In some embodiments, the processor 101 , the memory 102 , and the I / O interface 103 are connected to each other via a bus 104 , and further connected to other components of the computing device.

[0109] In a third aspect, an embodiment of the present disclosure provides a computer-readable medium having a computer program stored thereon, and when the program is executed by a processor, the method for analyzing abnormal access behavior described in the first aspect of the embodiment of the present disclosure is implemented.

[0110] In a fourth aspect, an embodiment of the present disclosure provides a computer program product, including a computer program or instructions, which, when executed by a processor, implements the method for analyzing abnormal access behavior described in the first aspect of the embodiment of the present disclosure.

[0111] In a fifth aspect, an embodiment of the present disclosure provides an analysis device for abnormal access behavior, including:

[0112] A model building function, used to build a baseline model using an operator model in an operator library, wherein the baseline model represents a baseline rule for describing a normal behavior pattern of an access behavior;

[0113] Data collection function, used to obtain access behavior data;

[0114] An analysis function, for analyzing the access behavior data based on the baseline model to identify abnormal access behavior;

[0115] The graph function is used to generate an access relationship graph based on the access behavior data and the identification results of abnormal access behavior.

[0116] In some embodiments, the analysis function is used to analyze the degree of deviation of the access behavior data from the baseline rule according to the baseline model, and identify abnormal access data that characterizes abnormal access behavior; and generalize process features based on the access behavior data and the abnormal access data to obtain an identification result of the abnormal access behavior.

[0117] In some embodiments, the analysis function analyzes the difference between the individual access behavior and the group access behavior represented by the initial identification result based on the initial identification result, and outputs a first identification result of the abnormal access behavior, wherein the first identification result conforms to the normal individual and group behavior baseline; and / or clusters the access behavior represented by the access behavior data based on the access behavior data and the initial identification result, aggregates and outputs a second identification result of the abnormal access behavior; and / or evaluates the initial identification result according to preset rules, and outputs a third identification result of the abnormal access behavior, wherein the third identification result satisfies the preset rules.

[0118] In some embodiments, a graph function is used to perform rule mapping and data association on the access behavior data to form a data set; based on the first recognition result, and / or the second recognition result, and / or the third recognition result, the data set is analyzed through multiple dimensions to form analysis results of multiple dimensions; the analysis results of the multiple dimensions are associated with each other to generate the access relationship graph, wherein the access relationship graph includes an asset portrait graph and a personnel portrait graph; graph analysis is performed on the asset portrait graph and the personnel portrait graph to identify abnormal access behaviors and threat events.

[0119] In some embodiments, the graph function is used to obtain and display the user's portrait information from the access relationship graph based on the user's keywords obtained from the search engine; to count the degree of deviation of the user's access behavior from the baseline rule with a preset time length as the statistical granularity, to identify and display abnormal access behaviors and threat events whose deviation exceeds the upper and lower limits.

[0120] In some embodiments, the model building function is used to obtain multiple operator models from the operator library to build an initial baseline model; train the initial baseline model using training samples; and correct the concatenation of operator models in the initial baseline model based on the training results and business scenarios to obtain the baseline model.

[0121] In some embodiments, the model building function includes an operator management unit, a program management unit, a model management unit, a task management unit, a service management unit, and a shortcut management unit; the model building function also includes a visualization interface, and building the baseline model through the model building function includes: in response to an operation instruction obtained through the visualization interface, performing the following operations: obtaining the operator model, customizing the operator model, etc. to manage the operator model through the operator management unit; managing the program for training the initial baseline model through the program management unit; managing the baseline model through the model management unit, such as operator model concatenation; performing task management operations through the task management unit; performing service management operations through the service management unit; and performing shortcut management operations through the shortcut management unit.

[0122] In some embodiments, the data collection function is used to obtain the access behavior data through multiple receiving servers deployed in a distributed manner, wherein different receiving servers correspond to different log sources; convert the access behavior data collected from multiple receiving servers into a standard data format; monitor the data collection status of each log source to determine the health status of the log source; monitor each receiving server to determine the health status of the receiving server.

[0123] In order to enable those skilled in the art to more clearly understand the technical solutions provided by the embodiments of the present disclosure, the technical solutions provided by the embodiments of the present disclosure are described in detail below through specific embodiments:

[0124] Example

[0125] This embodiment aggregates the data generated by business systems, user authentication, terminals, threat detection, and traffic auditing equipment, and uses data collection, data cleaning, data conversion and other data preparation methods to generalize process features using individual group comparison, cluster analysis, and regularity learning to build a baseline model based on the baseline. Finally, based on knowledge graphs, big data technology, and portrait labeling technology, the data of each business system is effectively integrated, and asset and personnel portraits are displayed to improve network security and data protection. The architecture diagram of this embodiment for analyzing abnormal access behavior is shown in the figure below. Figure 5 , Figure 6 shown.

[0126] (1) Data Collection

[0127] like Figure 3As shown, this embodiment provides open data collection management, including management of receiver management, analysis management, log source management and other links. This embodiment can monitor the data collection of a specific log source through log source monitoring to understand the health status of the log source. Distributed data collection can be achieved through the distributed deployment of multiple receiving servers, and the receiving server can be monitored through receiving server monitoring to understand the health status of the receiving server.

[0128] Receiver management is based on keyword feature matching or regular feature matching to manage User Datagram Protocol (UDP), Transmission Control Protocol (TCP), AVRO, message queue, NETFLOW, SFTP, files, REST, etc.

[0129] Log source management is based on task scheduling or resource management, and manages SYSLOG, file sharing, database tables, message stacks, APIs, etc.

[0130] In this embodiment, Figure 3 , Figure 5 , Figure 6 As shown, log sources involve various types of equipment such as business systems, user authentication, terminals, threat detection, and traffic auditing, including but not limited to business log agents, OS log agents, 4A, bastion hosts, domain control servers, EDR, zombie worms, honeypots, behavioral auditing, and database auditing.

[0131] In this embodiment, data preparation includes data import processing, data storage, data cleaning, data conversion, etc.

[0132] (2) Model construction

[0133] The construction of the baseline model is based on the machine learning operator algorithm, which allows for a customized drag-and-drop method to train abnormal behavior samples, and then through operator concatenation, the original algorithm analysis results can be corrected. The final algorithm can be more in line with the enterprise business scenarios and improve the algorithm accuracy.

[0134] The data processing engine provided by this function can complete the conversion of any data into system standard data; it has built-in rich data processing, visualization, and machine learning operators to meet any data processing requirements; it also has good scalability, supports custom operators and applies them to the system; operators can directly operate the database and analyze data from any source.

[0135] Functional architecture such as Figure 2 , Figure 5 , Figure 6As shown. The model building functional unit includes operator management unit, program management unit, model management unit, task management unit, service management unit, and shortcut management unit. The operator management unit includes management of input and output operators, machine learning operators, and visualization operators; the program management unit includes management of supervised learning modeling, unsupervised learning modeling, and deep learning modeling; model management includes variable management, fragment management, and group management; task management includes management of immediate tasks, scheduled tasks, and model editing; service management includes management of retrieval engines, message queues, and JDBC; shortcut management includes management of adding data, selecting models, and adding tasks.

[0136] (3) Model analysis

[0137] like Figure 5 , Figure 6 As shown, the third step of the construction and analysis technology of the baseline-based access violation anomaly analysis model is model analysis. Model analysis builds a large number of operator models in the operator library, combines them with feature data to form a feature algorithm, and then uses individual group comparison, cluster analysis, and rule learning to generalize process features.

[0138] Individual-group comparison: The individual-group comparison algorithm is based on the data behavior baseline. The prepared data is analyzed by the individual-group comparison algorithm. If it belongs to the normal baseline range, the result is output. If it belongs to the abnormal baseline range, the baseline is optimized, and the prepared data is analyzed again and the result content is output.

[0139] Cluster analysis: The cluster analysis algorithm is used to divide the objects (samples) in the data set into different groups (clusters), so that the objects in the same group have a higher similarity and the objects in different groups have a lower similarity. The K-means clustering algorithm is used to aggregate access violation abnormal behaviors and output the result content.

[0140] Rule learning: The rule learning algorithm uses the Apriori algorithm to perform cross-validation, accuracy verification, and recall verification on the prepared data. If the rule requirements are met, the results are output. If not, the rules are optimized, re-learned, and the result content is output.

[0141] like Figure 5 As shown, in this embodiment, abnormal access behaviors include but are not limited to:

[0142] Unusual login attempts, including frequent login failures, logins from different geographic locations or unusual times;

[0143] Unauthorized file or directory access, where a user attempts to access a file or directory that they do not have permission to access;

[0144] unusual data transfers, large amounts of data to or from unusual geographic locations, external systems, or at unusual times;

[0145] Abnormal privilege escalation, where a user attempts to elevate their privileges or access restricted resources;

[0146] Abnormal network activities, including abnormal network traffic, communications with malicious IP addresses, unauthorized remote access, etc.

[0147] Data theft, including attempts by users to access sensitive data or files, or to copy or download files without authorization;

[0148] Malware activity is the discovery of malware or malicious code on a user's system or network.

[0149] (4) Spectral analysis

[0150] like Figure 6 , Figure 7 As shown, the steps to implement access graph analysis technology are as follows:

[0151] 1. Obtain the data sources used for access graph analysis from the personnel library, rule library, and asset library.

[0152] 2. Perform data preprocessing work such as rule mapping and data association on data sources to form a standard, consistent and ordered data set.

[0153] 3. Analyze the data set through multiple dimensional analysis methods such as alarm analysis, vulnerability analysis, attack analysis, and behavior analysis, and form analysis results in multiple dimensions.

[0154] 4. Perform correlation analysis on the analysis results of the generated data sets in multiple dimensions, and finally generate asset portrait maps and personnel portrait maps and their corresponding analysis portrait results.

[0155] In this embodiment, the following effects can be achieved through access graph analysis:

[0156] 1. Comprehensive understanding of access behavior: Access graph analysis can draw out the association network between users or entities, helping system administrators to fully understand users' access behavior and access patterns. This helps to discover abnormal behavior and potential threats.

[0157] 2. Quickly discover abnormal behavior: By real-time monitoring and analysis of access graphs, abnormal patterns that are inconsistent with normal behavior can be quickly discovered, such as abnormal access paths, abnormal access frequencies, etc., thereby providing early warning of possible security risks.

[0158] 3. Improve security response efficiency: The access graph early warning system can send timely alerts to the security team, enabling them to take quick action to respond to potential threats, shorten the response time of security incidents, and reduce losses.

[0159] 4. Strengthen security management and supervision: Access graph analysis can provide more comprehensive and accurate data support for security management and supervision, help strengthen the control and supervision of system access, and ensure the safe and stable operation of information systems.

[0160] 5. Optimize security strategies: By analyzing the access graph, you can gain an in-depth understanding of the security risks and vulnerabilities in the system, provide important reference for formulating and optimizing security strategies, and further improve the security and protection capabilities of the system.

[0161] Access graph analysis technology extracts information from massive amounts of traffic, logs, and behavioral data to extract a comprehensive and detailed picture of multi-dimensional information about assets and personnel. After modeling analysis, correlation analysis, and data visualization, it effectively integrates data from various business systems based on knowledge graphs, big data technology, and portrait labeling technology to display a comprehensive picture of assets and personnel, solving the problems of poor correlation between data in various dimensions of assets and personnel and shallow granularity in data mining and analysis.

[0162] (5) Output data

[0163] Based on knowledge graphs, big data technology and portrait labeling technology, the data of various business systems are effectively integrated to output access graph analysis content of personnel portraits and asset portraits.

[0164] It will be appreciated by those skilled in the art that all or some of the steps, systems, and functional modules / units in the methods disclosed above may be implemented as software, firmware, hardware, and appropriate combinations thereof. In hardware implementations, the division between the functional modules / units mentioned in the above description does not necessarily correspond to the division of physical components; for example, a physical component may have multiple functions, or a function or step may be performed by several physical components in cooperation. Some or all physical components may be implemented as software executed by a processor, such as a central processing unit, a digital signal processor, or a microprocessor, or implemented as hardware, or implemented as an integrated circuit, such as an application-specific integrated circuit. Such software may be distributed on a computer-readable medium, which may include a computer storage medium (or non-transitory medium) and a communication medium (or temporary medium). As known to those skilled in the art, the term computer storage medium includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information (such as computer-readable instructions, data structures, program modules, or other data). Computer storage media include, but are not limited to, RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, digital versatile disks (DVD) or other optical disk storage, magnetic cassettes, magnetic tapes, magnetic disk storage or other magnetic storage devices, or any other medium that can be used to store the desired information and can be accessed by a computer. In addition, it is well known to those skilled in the art that communication media typically contain computer-readable instructions, data structures, program modules, or other data in a modulated data signal such as a carrier wave or other transport mechanism, and may include any information delivery media.

[0165] Example embodiments have been disclosed herein, and although specific terms are employed, they are used and should be interpreted only in a general illustrative sense and not for limiting purposes. In some instances, it will be apparent to those skilled in the art that, unless otherwise expressly noted, features, characteristics, and / or elements described in conjunction with a particular embodiment may be used alone or in combination with features, characteristics, and / or elements described in conjunction with other embodiments. Therefore, those skilled in the art will appreciate that various changes in form and detail may be made without departing from the scope of the present disclosure as set forth in the appended claims.

Claims

1. A method for analyzing abnormal access behavior, include: Building a baseline model using an operator model in an operator library, wherein the baseline model represents a baseline rule for describing a normal behavior pattern of an access behavior; Obtain access behavior data; Analyzing the access behavior data based on the baseline model to identify abnormal access behavior; An access relationship graph is generated based on the access behavior data and the identification results of abnormal access behaviors.

2. The analysis method according to claim 1, in, Analyzing the access behavior data based on the baseline model to identify abnormal access behavior includes: Analyzing the degree of deviation between the access behavior data and the baseline rule according to the baseline model, identifying abnormal access data representing abnormal access behavior, and obtaining an initial identification result; Process feature generalization is performed according to the access behavior data and the abnormal access data to obtain an identification result of the abnormal access behavior.

3. The analysis method according to claim 2, in, Process feature generalization is performed according to the access behavior data and the abnormal access data to obtain an identification result of the abnormal access behavior, including: Analyzing the difference between the individual access behavior and the group access behavior represented by the initial recognition result according to the initial recognition result, and outputting a first recognition result of the abnormal access behavior, wherein the first recognition result conforms to a normal individual and group behavior baseline; and / or Clustering the access behaviors represented by the access behavior data according to the access behavior data and the initial identification result, aggregating and outputting a second identification result of the abnormal access behavior; and / or The initial recognition result is evaluated according to a preset rule, and a third recognition result of the abnormal access behavior is output, wherein the third recognition result satisfies the preset rule.

4. The analysis method according to claim 3, in, Generate an access relationship graph based on the access behavior data and the identification result of abnormal access behavior, including: Performing rule mapping and data association on the access behavior data to form a data set; Analyze the data set in multiple dimensions according to the first recognition result, and / or the second recognition result, and / or the third recognition result to form analysis results in multiple dimensions; Performing correlation analysis on the analysis results of the multiple dimensions to generate the access relationship map, wherein the access relationship map includes an asset portrait map and a personnel portrait map; Perform graph analysis on the asset portrait graph and the personnel portrait graph to identify abnormal access behaviors and threat events.

5. The analysis method according to claim 4, in, Performing graph analysis on the asset portrait graph and the personnel portrait graph to identify abnormal access behaviors and threat events, including: According to the user's keywords obtained from the search engine, the user's portrait information is obtained from the access relationship graph and displayed; The degree of deviation of the user's access behavior from the baseline rule is counted with a preset time length as the statistical granularity, and abnormal access behavior and threat events whose deviation exceeds the upper and lower limits are identified and displayed.

6. The analysis method according to any one of claims 1 to 5, in, Use the operator models in the operator library to build a baseline model, including: Acquire multiple operator models from the operator library to build an initial baseline model; Training the initial baseline model using training samples; According to the training results and business scenarios, the concatenation of operator models in the initial baseline model is corrected to obtain the baseline model.

7. The analysis method according to claim 6, in, Use the operator models in the operator library to build a baseline model, including: Building the baseline model through a model building function; The model building function includes an operator management unit, a program management unit, a model management unit, a task management unit, a service management unit, and a shortcut management unit; The model building function also includes a visualization interface, and the baseline model is built through the model building function, including: In response to the operation instruction obtained through the visual interface, the following operations are performed: The operator management unit performs operations such as acquiring the operator model, customizing the operator model, etc. to manage the operator model; An operation of managing a program for training the initial baseline model is performed by the program management unit; The model management unit executes operations such as operator model concatenation to manage the baseline model; Execute task management operations through the task management unit; Executing service management operations through the service management unit; The shortcut management operation is performed through the shortcut management unit.

8. The analysis method according to any one of claims 1 to 5, in, Obtain access behavior data, including: Acquiring the access behavior data through a plurality of receiving servers deployed in a distributed manner, wherein different receiving servers correspond to different log sources; Converting the access behavior data collected from the plurality of receiving servers into a standard data format; Monitor the data collection status of each log source to determine the health status of the log source; Each of the receiving servers is monitored to determine the health status of the receiving server.

9. An electronic device, include: one or more processors; A memory having one or more programs stored thereon, wherein when the one or more programs are executed by the one or more processors, the one or more processors implement the abnormal access behavior analysis method according to any one of claims 1 to 8.

10. A computer-readable medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the method for analyzing abnormal access behavior according to any one of claims 1 to 8 is implemented.

11. A computer program product, comprising a computer program or instructions, wherein when the computer program or instructions are executed by a processor, the method for analyzing abnormal access behavior according to any one of claims 1 to 8 is implemented.

Citation Information

Patent Citations

  • User entity behavior analysis method and system of continuous immune safety system

    CN113918938A

  • User abnormal operation behavior identification method and device

    CN115577348A

  • System and method for monitoring and recording unknown virus behavior trajectory in real time

    CN117313095A

  • Identifying insider-threat security incidents via recursive anomaly detection of user behavior

    US20170126710A1

Cited By

  • Sensitive data security protection method and system

    CN120162809A

  • Security protection system of trusted data space

    CN120200864A

  • Access security protection method and system based on user behavior portrait

    CN121396653A

  • A user behavior portrait-based access security protection method and system

    CN121396653B