Process behavior identification method and device and computer readable storage medium

By using blacklist and whitelist models to detect the underlying logical data of the target process at the operating system level, the problems of interference and data leakage in the existing technology are solved, and accurate identification and security detection of process behavior are achieved.

CN119026122AInactive Publication Date: 2024-11-26SICHUAN MEDITATION SOFTWARE CO LTD
View PDF 5 Cites 0 Cited by

Patent Information

Application Number
CN202411034807.7
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-30
Publication Date
2024-11-26
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

The existing machine learning-based abnormal behavior detection methods mainly stay at the network level, resulting in the collection and modeling of user service flow data, interfering with normal services and leaking sensitive data, and unable to effectively identify process behavior.

Method used

By obtaining the underlying logical data of the target process, using blacklist models and whitelist models for detection, avoiding direct collection of business data, ensuring data confidentiality, and improving detection accuracy through dual-model detection.

Benefits of technology

It realizes accurate identification of process behavior, avoids interference to user business and data leakage risks, and improves detection accuracy and security.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119026122A_ABST
    Figure CN119026122A_ABST
Patent Text Reader

Abstract

The invention discloses a process behavior identification method and device and a computer readable storage medium, and aims to solve the problem that the normal service of a user is inevitably interfered due to the fact that most detection schemes stay on a network level and service flow data of the user needs to be collected and modeled. And the privacy and sensitive data of the user face the risk of leakage. The invention provides a process behavior identification method, which is characterized in that a result obtained by detecting bottom-layer logic data through a blacklist model and a white list model according to the bottom-layer logic data of a target process is used as a basis for determining whether the process behavior of the target process is abnormal or not, so that direct acquisition and use of user service data can be avoided; zero interference and transparency to user services are achieved, the risk of data leakage of the security model is eradicated from the source, and confidentiality of user data is ensured. In addition, the data is detected through double models, so that the detection accuracy can be greatly improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present disclosure generally relates to the field of machine learning technology, and more particularly to a method and device for identifying process behavior, and a computer-readable storage medium. Background Art

[0002] Important pillar industries such as finance, energy, and transportation are extremely vulnerable to ransomware attacks. Since machine learning algorithms can explore deeper connections between data features, they can make more effective use of information from normal data and threat data, and show higher accuracy and better generalization capabilities. Even if the data content in two normal flows is completely different, they can still achieve good recognition capabilities. Therefore, abnormal behavior detection methods based on machine learning are often used to detect ransomware.

[0003] However, most of the current abnormal behavior detection solutions based on machine learning remain at the network level, and often require the collection and modeling of the user's business flow data itself, which inevitably causes certain interference to the user's normal business and puts the user's privacy and sensitive data at risk of leakage. Summary of the invention

[0004] In view of the above defects or deficiencies in the prior art, it is desired to provide a method for identifying process behavior that can avoid and improve the above problems.

[0005] A first aspect of the present application provides a method for identifying process behavior, the method comprising:

[0006] Acquire the underlying logic data of the target process, where the underlying logic data is data generated by the user business data of the target process in the operating system;

[0007] The underlying logic data is detected by a blacklist model and a whitelist model respectively to obtain a first detection result and a second detection result, wherein the blacklist model is obtained by training based on the blacklist data of the operating system layer, and the whitelist model is obtained by training based on the whitelist data of the operating system layer, and the first detection result is used to characterize whether the underlying logic data is blacklist data, and the second detection result is used to characterize whether the underlying logic data is whitelist data;

[0008] The process behavior of the target process is determined according to the first detection result and the second detection result.

[0009] The method for identifying the mapping process behavior provided by the present application, considering that most detection schemes remain at the network level, often requires the collection and modeling of the user's business flow data itself, which inevitably causes certain interference to the user's normal business and exposes the user's privacy and sensitive data to the risk of leakage. The present application provides a method for identifying process behavior, which detects the underlying logical data of the target process through a blacklist model and a whitelist model respectively, and obtains a first detection result and a second detection result. The first detection result and the second detection result are used as the basis for determining whether the process behavior of the target process is abnormal, which can avoid the direct collection and use of user business data, zero interference and transparency in user business, and eliminate the risk of data leakage of the security model itself from the source, ensuring the confidentiality of user data. In addition, the present application detects data through dual models, which can greatly improve the accuracy of detection.

[0010] A second aspect provides a process behavior identification device, the device comprising:

[0011] An acquisition module, used to acquire the underlying logic data of the target process, wherein the underlying logic data is data generated by the user business data of the target process in the operating system;

[0012] A detection module, used to detect the underlying logic data through a blacklist model and a whitelist model respectively to obtain a first detection result and a second detection result, wherein the blacklist model is obtained by training based on the blacklist data of the operating system layer, and the whitelist model is obtained by training based on the whitelist data of the operating system layer, and the first detection result is used to characterize whether the underlying logic data is blacklist data, and the second detection result is used to characterize whether the underlying logic data is whitelist data;

[0013] A determination module is used to determine the process behavior of the target process according to the first detection result and the second detection result.

[0014] A third aspect provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program implements the method of the first aspect when executed by a processor. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Other features, objects and advantages of the present application will become more apparent by reading the detailed description of non-limiting embodiments made with reference to the following drawings:

[0016] Figure 1 A flowchart of a process behavior identification method provided by the present application;

[0017] Figure 2 A flowchart of a process behavior identification method provided by the present application;

[0018] Figure 3 A flowchart of a blacklist model training provided for this application;

[0019] Figure 4 A flowchart of a process behavior identification method provided by the present application;

[0020] Figure 5 A flowchart of a whitelist model training provided for this application;

[0021] Figure 6 A flowchart of a process behavior identification method provided by the present application;

[0022] Figure 7 A flowchart of the steps of a process behavior identification method provided by this application;

[0023] Figure 8 A structural block diagram of a process behavior identification device provided by the present application;

[0024] Fig. 9 A schematic diagram of the structure of a computer system of a computer device provided in this application. DETAILED DESCRIPTION

[0025] The present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It is to be understood that the specific embodiments described herein are only used to explain the relevant invention, rather than to limit the invention. It is also necessary to explain that, for ease of description, only the parts related to the invention are shown in the accompanying drawings.

[0026] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0027] Please refer to Figure 1 , Figure 1 The present application provides a method for identifying process behavior, which is described by taking the method executed on a computer device as an example. The method includes the following steps:

[0028] Step S20, obtaining the underlying logic data of the target process, where the underlying logic data is the data generated by the user business data of the target process in the operating system;

[0029] The target process refers to any running activity of a program in a computer device on a certain data set. The target process represents the execution process of a program in the operating system. In this process, the program is executed sequentially on the processor, accompanied by the allocation and management of the resources required by the program. The process includes not only executable code, but also process-specific data (such as input and output), call stack, and possible security features.

[0030] The computer device of the present application determines whether the process behavior of the target process is normal or abnormal based on the underlying logic data of the program corresponding to the target process during execution. The underlying logic data is the user business data of the target process generated in the operating system, that is, the user business data corresponding to the target process can be generated after running at the operating system layer and undergoing certain logic operations.

[0031] Therefore, the underlying logical data is data collected from the operating system of the computer device. For example, the underlying logical data includes resource consumption data, natural language logs, assembly instruction streams, system call sequences, CPU occupancy, memory, hard disk, etc.

[0032] The process behavior is judged through the underlying logical data of the target process. There is no need for computer equipment to provide a special interface to collect data, so it will not interfere with normal business. There is no need to collect business data, which can avoid the leakage of important data (user personal identity information, account information, password information, etc.).

[0033] In an optional embodiment, after acquiring the underlying logic data, the present application further performs the following operations on the underlying logic data:

[0034] Perform cleaning operations on the underlying logical data. Since the collected underlying logical data may be incomplete, incorrect, inaccurate, irrelevant, etc., advanced cleaning operations on the underlying logical data through cleaning tools can obtain the data required for subsequent judgments of this application, avoiding the influence of unnecessary data on the judgment results.

[0035] In addition, after cleaning the underlying logic data, you can also perform corresponding characteristic analysis operations on the underlying logic data to obtain the specific statistical distribution of the underlying logic data and the main information of the data. This will facilitate subsequent judgments based on more refined and accurate data and improve the speed of judgment.

[0036] In addition, the underlying logical data can also be processed, for example, by data format conversion, data feature extraction, and labeling. By converting the underlying logical data into data format, data of different forms can be converted into a unified form, which facilitates unified judgment. Performing data feature extraction operations on the underlying logical data can reduce the dimension of the data without losing data information, so as to reduce the size of a single data, thereby reducing the amount of data for the entire judgment, and highlighting effective data features to improve the efficiency of judgment. If a variety of different models are used in the subsequent judgment of the underlying logical data, the underlying logical data can be labeled to select a more suitable judgment tool for it through the label of the underlying logical data to improve the accuracy of the judgment. The label of the underlying logical data can be a type used to characterize the data, such as a number, text, time series, etc.

[0037] Step S30, respectively detecting the underlying logic data by a blacklist model and a whitelist model to obtain a first detection result and a second detection result, wherein the blacklist model is obtained by training based on the blacklist data of the operating system layer, and the whitelist model is obtained by training based on the whitelist data of the operating system layer, and the first detection result is used to characterize whether the underlying logic data is blacklist data, and the second detection result is used to characterize whether the underlying logic data is whitelist data;

[0038] Among them, the blacklist model and the whitelist model can be pre-trained and stored in the memory address corresponding to the computer device. When detection is required, they can be called from the corresponding memory address.

[0039] The blacklist model is used to detect whether the underlying logical data is blacklist data, and the whitelist model is used to detect whether the underlying logical data is whitelist data.

[0040] In an optional embodiment, if Figure 2 As shown, Figure 2 An optional method embodiment of obtaining a blacklist model provided by the present application includes the following steps:

[0041] Step S201, obtaining a blacklist sample data set from a third party;

[0042] The third party may be, for example, other computer devices, host computers, etc., or historical data stored in a memory by a third party, etc. The blacklist sample data set may include data of various types, such as numerical data, text data, time series data, etc.

[0043] It should be noted here that the data in the blacklist sample data set also comes from the underlying logical data of the operating system. The underlying logical data and user business data are very different in content and form, but they are similar in structure, and the similarity of the structural relationship means the similarity of the operating paradigms of the two. The traditional data modeling method is a process of continuously optimizing data through fixed logic. It is difficult to reconstruct the behavioral logic of the underlying logical data based on user business data. Therefore, this application adopts machine learning to train the model to obtain a blacklist model. Machine learning is not limited to the type of data, so the obtained blacklist sample data set includes different types of data, making the data richer, so that it can avoid overfitting problems caused by relying on a single data and enhance the robustness of the model.

[0044] In an optional embodiment, after obtaining the blacklist sample data set, the present application may also preprocess the data in the blacklist sample data set, and the preprocessing includes at least one of data cleaning, extracting the main information of the data, converting the data form, extracting data features, and marking labels. The reasons for performing the above processing on the sample data have been introduced above and will not be repeated here.

[0045] Step S202, classifying the data in the blacklist sample data set according to data type to obtain a plurality of different types of first sample data sets;

[0046] Among them, since the blacklist sample data set includes many different types of data, in order to conduct more targeted model training, it is first necessary to classify the data in the blacklist sample data set according to the data type, that is, to divide the data of the same type into a data subset to facilitate subsequent classification model training.

[0047] Step S203, inputting different types of first sample data sets into different first initial models for model training to obtain a plurality of different first models;

[0048] The first initial model may also be different models set according to different data types. The number of the first initial models is the same as the number of data types, that is, there are as many first initial models as there are types of data.

[0049] like Figure 3 As shown, Figure 3 This is a flowchart for training a blacklist model. As can be seen from the flowchart, by inputting different types of first sample data sets into different models respectively and performing unsupervised learning through different methods, initial blacklist models corresponding to various types of data can be obtained in a more targeted manner. In other words, different initial blacklist models can be obtained by training different types of data.

[0050] Step S204: Screen out a blacklist model from a plurality of different first models by using an ensemble learning method.

[0051] Among them, after different initial blacklist models can be obtained based on the above-mentioned training of different types of data, in order to facilitate subsequent identification, the present application also needs to determine a model with comprehensive recognition capabilities based on multiple different initial blacklist models. Here, the present application processes multiple different initial blacklist models through an integrated learning method to obtain the final required blacklist model.

[0052] In an optional embodiment, if Figure 4 As shown, Figure 4 An optional method embodiment of obtaining a whitelist model provided by the present application includes the following steps:

[0053] Step S401, collecting a whitelist sample data set from the operating system layer;

[0054] The data in the whitelist sample data set is the underlying logical data collected from the operating system layer of the computer device. The whitelist sample data set may also include a variety of different types of data, such as numerical data, text data, time series data, etc.

[0055] This application uses machine learning to perform model training to obtain a whitelist model. Machine learning is not limited to the type of data, so the obtained whitelist sample data set includes different types of data, making the data richer. This avoids overfitting problems caused by relying on a single data and enhances the robustness of the model.

[0056] In an optional embodiment, after obtaining the whitelist sample data set, the present application may also preprocess the data in the whitelist sample data set, and the preprocessing includes at least one of data cleaning, extracting main data information, data format conversion, data feature extraction, and labeling. The reasons for performing the above processing on the sample data have been introduced above and will not be repeated here.

[0057] Step S402, classifying the data in the whitelist sample data set according to data type to obtain a plurality of different types of second sample data sets;

[0058] Among them, since the whitelist sample data set includes many different types of data, in order to conduct more targeted model training, it is first necessary to classify the data in the whitelist sample data set according to the data type, that is, to divide the data of the same type into a data subset to facilitate subsequent classification model training.

[0059] Step S403, inputting different types of second sample data sets into different second initial models for model training to obtain a plurality of different second models;

[0060] The second initial model may also be different models set according to different data types. The number of the second initial models is the same as the number of data types, that is, there are as many second initial models as there are types of data.

[0061] like Figure 5 As shown, Figure 5 This is a flowchart for training a whitelist model. As can be seen from the flowchart, by inputting different types of second sample data sets into different models and performing unsupervised learning through different methods, initial whitelist models corresponding to various types of data can be obtained in a more targeted manner. In other words, different initial whitelist models can be obtained by training different types of data.

[0062] Step S404: Screen out a whitelist model from a plurality of different second models by using an ensemble learning method.

[0063] Among them, after different initial whitelist models can be obtained based on the above-mentioned training of different types of data, in order to facilitate subsequent identification, the present application also needs to determine a model with comprehensive recognition capabilities based on multiple different initial whitelist models. Here, the present application processes multiple different initial whitelist models through an integrated learning method to obtain the final required black and white list model.

[0064] After training the blacklist model and the whitelist model according to the above process, the present application will detect the underlying logical data of the target process obtained from the operating system of the computer device in real time through the blacklist model and the whitelist model respectively to obtain the first detection result and the second detection result.

[0065] Among them, the underlying logic data of the target process can be simultaneously input into the blacklist model and the whitelist model for detection to obtain the first detection result and the second detection result. The first detection result includes that the underlying logic data belongs to the blacklist data and the underlying logic data does not belong to the blacklist data; the second detection result includes that the underlying logic data belongs to the whitelist data and the underlying logic data does not belong to the whitelist data. Since the data used in the training of the two models is limited, any model obtained through limited data training has certain limitations. The present application uses two different models to detect the same data separately, which can improve the accuracy of detection.

[0066] The present application may also first input the underlying logic data into a whitelist model for detection to obtain a second detection result. When the second detection result is that the underlying logic data belongs to the whitelist data, the underlying logic data is then input into a blacklist model for detection to obtain the first detection result.

[0067] In an optional embodiment, if Figure 6 As shown, Figure 6 An optional method embodiment for detecting by a blacklist model and a whitelist model provided by the present application includes the following steps:

[0068] Step S601, inputting the underlying logic data into the blacklist model to perform blacklist detection to obtain a first detection result;

[0069] Step S602: If the first detection result is that the underlying logic data does not belong to the blacklist data, the underlying logic data is input into the whitelist model to perform a whitelist detection to obtain a second detection result.

[0070] Among them, Figure 7 As shown, the present application may be to first input the underlying logic data into a blacklist model for detection to obtain a first detection result, and when the first detection result is that the underlying logic data does not belong to the blacklist data, then input the underlying logic data into a whitelist model for detection to obtain a second detection result.

[0071] It is understandable that if the first detection result is that the underlying logical data belongs to the blacklist model, the present application may not need to input the underlying logical data into the whitelist model for detection, and directly determine the process behavior of the target process based on the first detection result, thereby improving detection efficiency.

[0072] Step S40: determining the process behavior of the target process according to the first detection result and the second detection result.

[0073] Among them, after the first detection result and the second detection result are obtained according to the above method, the present application may determine that the process behavior of the target process is abnormal as long as the first detection result is that the underlying logical data belongs to the blacklist data;

[0074] The present application may also be to determine that the process behavior of the target process is abnormal as long as the second detection result is that the underlying logic data does not belong to the whitelist data;

[0075] The present application may also be to determine that the process behavior of the target process is abnormal when the first detection result is that the underlying logic data belongs to the blacklist data and the second detection result is that the underlying logic data does not belong to the whitelist data;

[0076] The present application may also determine that the process behavior of the target process is normal when the first detection result is that the underlying logic data does not belong to the blacklist data and the second detection result is that the underlying logic data belongs to the whitelist data.

[0077] In an optional embodiment, the computer device of the present application may be to determine that the process behavior of the target process is abnormal if the first detection result is that the underlying logic data belongs to blacklist data, and the second detection result is that the underlying logic data does not belong to whitelist data;

[0078] If the first detection result is that the underlying logic data does not belong to the blacklist data, and the second detection result is that the underlying logic data belongs to the whitelist data, it is determined that the process behavior of the target process is normal.

[0079] The present application performs double judgment through the first detection result and the second detection result, so as to improve the accuracy of judging the process behavior of the target process.

[0080] In an optional embodiment, the present application executes a blocking process when it is determined that the process behavior of the target process is abnormal, and the blocking process is used to block the target process. Through this operation, direct blocking can be given in a timely manner at the operating system level to reduce losses.

[0081] It should be noted that although the operations of the method of the present invention are described in a specific order in the accompanying drawings, this does not require or imply that these operations must be performed in this specific order, or that all the operations shown must be performed to achieve the desired results. On the contrary, the steps depicted in the flow chart can change the execution order.

[0082] Further references Figure 8 , which shows an exemplary structural block diagram of a process behavior identification device 800 according to an embodiment of the present application. The device 800 includes an acquisition module 801, a detection module 802 and a determination module 803.

[0083] The acquisition module 801 is used to acquire the underlying logic data of the target process, where the underlying logic data is the data generated by the user business data of the target process in the operating system;

[0084] The detection module 802 is used to detect the underlying logic data through a blacklist model and a whitelist model respectively to obtain a first detection result and a second detection result. The blacklist model is obtained by training based on the blacklist data of the operating system layer, and the whitelist model is obtained by training based on the whitelist data of the operating system layer. The first detection result is used to characterize whether the underlying logic data is blacklist data, and the second detection result is used to characterize whether the underlying logic data is whitelist data;

[0085] The determination module 803 is used to determine the process behavior of the target process according to the first detection result and the second detection result.

[0086] In an optional embodiment, the detection module 802 is specifically used to input the underlying logic data into the blacklist model to perform blacklist detection and obtain a first detection result;

[0087] If the first detection result is that the underlying logic data does not belong to the blacklist data, the underlying logic data is input into the whitelist model to perform a whitelist detection to obtain a second detection result.

[0088] In an optional embodiment, the determination module 803 is specifically configured to determine that the process behavior of the target process is abnormal if the first detection result is that the underlying logic data belongs to the blacklist data, and the second detection result is that the underlying logic data does not belong to the whitelist data;

[0089] If the first detection result is that the underlying logic data does not belong to the blacklist data, and the second detection result is that the underlying logic data belongs to the whitelist data, it is determined that the process behavior of the target process is normal.

[0090] In an optional embodiment, the above device further includes a blocking module.

[0091] The blocking module is used to execute a blocking process when it is determined that the process behavior of the target process is abnormal, and the blocking process is used to block the target process.

[0092] In an optional embodiment, the above device further includes a training module.

[0093] The training module is used to obtain a blacklist sample dataset from a third party;

[0094] Classifying the data in the blacklist sample data set according to data type to obtain a plurality of different types of first sample data sets;

[0095] Inputting different types of first sample data sets into different first initial models for model training to obtain a plurality of different first models;

[0096] The blacklist model is screened out from multiple different first models through an integrated learning method.

[0097] In an optional embodiment, the training module is also used to preprocess the data in the blacklist sample data set, and the preprocessing includes at least one of data cleaning, extracting main information of the data, data format conversion, data feature extraction and labeling.

[0098] In an optional embodiment, the training module is further used to collect a whitelist sample data set from the operating system layer;

[0099] Classifying the data in the whitelist sample data set according to data type to obtain a plurality of different types of second sample data sets;

[0100] Inputting different types of second sample data sets into different second initial models for model training to obtain a plurality of different second models;

[0101] A whitelist model is screened out from a plurality of different second models through an integrated learning method.

[0102] In an optional embodiment, the training module is also used to preprocess the data in the whitelist sample data set, and the preprocessing includes at least one of data cleaning, extracting main information of the data, data format conversion, data feature extraction and labeling.

[0103] It should be understood that the units or modules described in the apparatus 800 are similar to those described in the reference Figure 2 The various steps in the described method correspond to each other. Therefore, the operations and features described above for the method are also applicable to the device 800 and the units contained therein, and will not be repeated here. The device 800 can be pre-implemented in a browser or other security application of an electronic device, or loaded into a browser or its security application of an electronic device by downloading or the like. The corresponding units in the device 800 can cooperate with the units in the electronic device to implement the solution of the embodiment of the present application.

[0104] Reference below Fig. 9 , which shows a schematic diagram of the structure of a computer system 900 suitable for implementing a service device of an embodiment of the present application.

[0105] like Fig. 9 As shown, the computer system 900 includes a central processing unit (CPU) 901, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 902 or a program loaded from a storage part 908 into a random access memory (RAM) 903. In the RAM 903, various programs and data required for the operation of the system 900 are also stored. The CPU 901, the ROM 902, and the RAM 903 are connected to each other via a bus 904. An input / output (I / O) interface 905 is also connected to the bus 904.

[0106] The following components are connected to the I / O interface 909: an input section 906 including a keyboard, a mouse, etc.; an output section 907 including a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker, etc.; a storage section 908 including a hard disk, etc.; and a communication section 909 including a network interface card such as a LAN card, a modem, etc. The communication section 909 performs communication processing via a network such as the Internet. A drive 910 is also connected to the I / O interface 905 as needed. A removable medium 911, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 910 as needed, so that a computer program read therefrom is installed into the storage section 908 as needed.

[0107] In particular, according to the embodiments of the present disclosure, the above reference Figure 2-7 The described processes may be implemented as computer software programs. For example, embodiments of the present disclosure include a computer program product comprising a computer program tangibly embodied on a machine-readable medium, the computer program comprising a computer program for executing Figure 2-7 In such an embodiment, the computer program may be downloaded and installed from a network via the communication section 909 and / or installed from the removable medium 911.

[0108] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present invention. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, program segment or a part of a code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the square box can also occur in a different order from the order marked in the accompanying drawings. For example, two square boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs the specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0109] The units or modules involved in the embodiments described in the present application may be implemented by software or hardware. The units or modules described may also be arranged in a processor, for example, may be described as: a processor includes an XX unit, a YY unit, and a ZZ unit. The names of these units or modules do not, in some cases, constitute a limitation on the unit or module itself, for example, the XX unit may also be described as a "unit for XX".

[0110] As another aspect, the present application further provides a computer-readable storage medium, which may be a computer-readable storage medium included in the device described in the above embodiment; or a computer-readable storage medium that exists independently and is not assembled into the device. The computer-readable storage medium stores one or more programs, and the programs are used by one or more processors to execute the formula input method described in the present application.

[0111] The above description is only a preferred embodiment of the present application and an explanation of the technical principles used. Those skilled in the art should understand that the scope of the invention involved in the present application is not limited to the technical solution formed by a specific combination of the above technical features, but should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the inventive concept. For example, the above features are replaced with (but not limited to) technical features with similar functions disclosed in the present application.

Claims

1. A method for identifying process behavior, It is characterized in that The method comprises: Acquire the underlying logic data of the target process, where the underlying logic data is data generated by the user business data of the target process in the operating system; The underlying logic data is detected by a blacklist model and a whitelist model respectively to obtain a first detection result and a second detection result, wherein the blacklist model is obtained by training based on the blacklist data of the operating system layer, and the whitelist model is obtained by training based on the whitelist data of the operating system layer, and the first detection result is used to characterize whether the underlying logic data is blacklist data, and the second detection result is used to characterize whether the underlying logic data is whitelist data; The process behavior of the target process is determined according to the first detection result and the second detection result.

2. The identification method according to claim 1, It is characterized in that The detecting the underlying logic data by using a blacklist model and a whitelist model respectively to obtain a first detection result and a second detection result includes: Inputting the underlying logic data into the blacklist model to perform blacklist detection to obtain the first detection result; If the first detection result is that the underlying logic data does not belong to the blacklist data, the underlying logic data is input into the whitelist model to perform a whitelist detection to obtain the second detection result.

3. The identification method according to claim 1, It is characterized in that The determining the process behavior of the target process according to the first detection result and the second detection result includes: If the first detection result is that the underlying logic data belongs to blacklist data, and the second detection result is that the underlying logic data does not belong to whitelist data, then determining that the process behavior of the target process is abnormal; If the first detection result is that the underlying logic data does not belong to the blacklist data, and the second detection result is that the underlying logic data belongs to the whitelist data, it is determined that the process behavior of the target process is normal.

4. The identification method according to claim 2 or 3, It is characterized in that The method further comprises: In the case where it is determined that the process behavior of the target process is abnormal, a blocking process is executed, where the blocking process is used to block the target process.

5. The identification method according to claim 1, It is characterized in that The training process of the blacklist model includes: Obtain a blacklist sample dataset from a third party; Classifying the data in the blacklist sample data set according to data type to obtain a plurality of different types of first sample data sets; Inputting the different types of first sample data sets into different first initial models for model training to obtain a plurality of different first models; The blacklist model is screened out from the multiple different first models by an integrated learning method.

6. The identification method according to claim 5, It is characterized in that The method further comprises: The data in the blacklist sample data set is preprocessed, and the preprocessing includes at least one of data cleaning, extracting main data information, data form conversion, data feature extraction and labeling.

7. The identification method according to claim 1, It is characterized in that The training process of the whitelist model includes: Collect whitelist sample datasets from the operating system layer; Classifying the data in the whitelist sample data set according to data type to obtain a plurality of different types of second sample data sets; Inputting the different types of second sample data sets into different second initial models for model training to obtain a plurality of different second models; The whitelist model is screened out from the multiple different second models by an integrated learning method.

8. The identification method according to claim 7, It is characterized in that The method further comprises: The data in the whitelist sample data set is preprocessed, and the preprocessing includes at least one of data cleaning, extracting main data information, data format conversion, data feature extraction and labeling.

9. A device for identifying process behavior, It is characterized in that The device comprises: An acquisition module, used to acquire the underlying logic data of the target process, wherein the underlying logic data is data generated by the user business data of the target process in the operating system; A detection module, used to detect the underlying logic data through a blacklist model and a whitelist model respectively to obtain a first detection result and a second detection result, wherein the blacklist model is obtained by training based on the blacklist data of the operating system layer, and the whitelist model is obtained by training based on the whitelist data of the operating system layer, and the first detection result is used to characterize whether the underlying logic data is blacklist data, and the second detection result is used to characterize whether the underlying logic data is whitelist data; A determination module is used to determine the process behavior of the target process according to the first detection result and the second detection result.

10. A computer-readable storage medium having a computer program stored thereon, It is characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 8 are implemented.

Citation Information

Patent Citations

  • Operation state abnormity detection method, device and equipment

    CN111738467A

  • Program white list updating method based on TCM software server

    CN112417436A

  • Illegal software identification method

    CN113569242A

  • Method, device and equipment for identifying illegal privilege lifting of process and storage medium

    CN115659327A

  • Process controlling system for managing computer in incorporation using gray list and method thereof

    KR1020110104373A