Fraud-related violation APP detection method and electronic equipment
By building a knowledge graph and using a graph attention neural network, the problem that traditional detection methods cannot adapt to the rapidly changing APP ecological environment is solved, and high accuracy and strong adaptability of fraud-related violations APP detection is achieved.
Patent Information
- Application Number
- CN202411112187.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-14
- Publication Date
- 2024-11-26
AI Technical Summary
Traditional fraud-related and irregular APP detection methods are difficult to adapt to the rapidly changing APP ecological environment, and cannot comprehensively capture and analyze the complex relationship between APP and its associated entities, resulting in inaccurate detection results.
By obtaining APP data, building a knowledge graph, using graph attention neural network to train and verify the data, generating an APP detection model, and then detecting whether the APP to be detected is a fraud-related violation APP.
APP detection with high accuracy, strong adaptability and multi-dimensional analysis is achieved, providing strong technical support for APP market supervision and user protection.
Smart Images

Figure CN119026123A_ABST
Abstract
Description
Technical Field
[0001] The present disclosure relates to the field of application programs, and more particularly to a method and electronic device for detecting fraudulent and illegal APPs. Background Art
[0002] With the development of science and technology, mobile applications (Application, APP) have fully penetrated into people's lives, and some fraud-related illegal APPs have also appeared frequently, posing a major threat to the security of users and the market. Traditional detection methods mainly rely on static analysis and dynamic behavior detection, but due to their strong regularity and difficulty in adapting to the rapidly changing APP ecological environment, the detection effect is limited, and it is impossible to fully capture and analyze the complex relationship between APP and its related entities, resulting in inaccurate detection results. Summary of the invention
[0003] The present disclosure provides a method and electronic device for detecting fraudulent and illegal APPs, so as to at least solve the above technical problems existing in the prior art.
[0004] According to a first aspect of the present disclosure, a method for detecting fraudulent and illegal APPs is provided, the method comprising:
[0005] Acquire APP data, and obtain data to be processed based on the APP data;
[0006] Identify entities and relationships between entities in the data to be processed, and construct a knowledge graph using the entities as nodes and the relationships between the entities as edges;
[0007] A training set and a validation set are obtained based on the data set of the knowledge graph, and the graph attention neural network is trained and validated using the training set and the validation set to obtain an APP detection model;
[0008] The data of the APP to be detected is input into the APP detection model for detection, and a detection result of whether the APP to be detected is a fraud-related and illegal APP is output.
[0009] In one possible implementation manner, the acquiring APP data and obtaining the data to be processed based on the APP data includes:
[0010] Collect APP data through online channels or APP engines;
[0011] The APP data is cleaned and standardized to obtain data to be processed.
[0012] In one possible implementation, identifying entities and relationships between entities in the data to be processed, and constructing a knowledge graph using the entities as nodes and the relationships between the entities as edges, includes:
[0013] Determine static data of the data to be processed, identify static entities in the static data, and first relationships between the static entities;
[0014] Determining dynamic data of the data to be processed, identifying dynamic entities in the dynamic data, and second relationships between the dynamic entities;
[0015] Graph structure data is constructed with static entities and dynamic entities as nodes and first relationships and second relationships as edges; the graph structure data is used as a knowledge graph.
[0016] In one possible implementation, the data set based on the knowledge graph is used to obtain a training set and a validation set, including:
[0017] A graph attention network is used to perform representation learning on the graph structure data to obtain a high-dimensional vector;
[0018] The high-dimensional vector is labeled to obtain a training set and a validation set.
[0019] In one possible implementation, the training set and the validation set are used to train and validate the graph attention neural network to obtain an APP detection model, including:
[0020] Using the training set to train the graph attention neural network to obtain an initial detection model;
[0021] Using the verification set to verify the initial detection model, to obtain a verification result;
[0022] Based on the verification results, the model parameters of the initial detection model are optimized by cross-validation until the model converges to obtain an APP detection model.
[0023] In one possible implementation, it also includes:
[0024] The detection results are fed back into the knowledge graph to update the knowledge graph.
[0025] In one possible implementation manner, the test result includes a test report.
[0026] In one possible implementation, the graph structure data is stored in a graph database.
[0027] According to a second aspect of the present disclosure, a device for detecting fraudulent and illegal APPs is provided, the device comprising:
[0028] An acquisition module, used to acquire APP data and obtain data to be processed based on the APP data;
[0029] A construction module, used to identify entities and relationships between entities in the data to be processed, and to construct a knowledge graph using the entities as nodes and the relationships between the entities as edges;
[0030] A training module, used to obtain a training set and a verification set based on the data set of the knowledge graph, and use the training set and the verification set to train and verify the graph attention neural network to obtain an APP detection model;
[0031] The detection module is used to input the data of the APP to be detected into the APP detection model for detection, and output the detection result of whether the APP to be detected is a fraud-related and illegal APP.
[0032] According to a third aspect of the present disclosure, there is provided an electronic device, including:
[0033] at least one processor; and
[0034] a memory communicatively connected to the at least one processor; wherein,
[0035] The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method described in the present disclosure.
[0036] According to a fourth aspect of the present disclosure, a non-transitory computer-readable storage medium storing computer instructions is provided, wherein the computer instructions are used to cause the computer to execute the method described in the present disclosure.
[0037] The disclosed fraud-related illegal APP detection method and electronic device, the application first obtains multi-dimensional APP data, pre-processes the multi-dimensional APP data to obtain the data to be processed, and then constructs a knowledge graph through the relationship between the entities of the data to be processed, and trains and verifies the graph attention neural network through the training set and verification set composed of the knowledge graph to obtain the APP detection model, so as to calculate the data of the APP to be detected through the APP detection model to determine whether the APP to be detected is a fraud-related illegal APP. The application constructs a knowledge graph through multi-dimensional data, and then uses the knowledge graph to train the graph attention neural network to obtain the APP detection model, so that the obtained APP detection model has the advantages of high accuracy, strong adaptability and multi-dimensional analysis, and provides strong technical support for the supervision of the APP market and user protection.
[0038] It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present disclosure, nor is it intended to limit the scope of the present disclosure. Other features of the present disclosure will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS
[0039] The above and other objects, features and advantages of the exemplary embodiments of the present disclosure will become readily understood by reading the detailed description below with reference to the accompanying drawings. In the accompanying drawings, several embodiments of the present disclosure are shown in an exemplary and non-limiting manner, in which:
[0040] In the drawings, the same or corresponding reference numerals represent the same or corresponding parts.
[0041] Figure 1 A schematic diagram of the implementation process of the fraud-related and illegal APP detection method according to an embodiment of the present disclosure is shown;
[0042] Figure 2 A flow chart of constructing a knowledge graph according to an embodiment of the present disclosure is shown;
[0043] Figure 3 A simplified diagram of the APP entity relationship in an embodiment of the present disclosure is shown;
[0044] Figure 4 A schematic diagram of the structure of a device for detecting fraudulent and illegal APPs according to an embodiment of the present disclosure is shown;
[0045] Figure 5 A schematic diagram of the structure of an electronic device according to an embodiment of the present disclosure is shown. DETAILED DESCRIPTION
[0046] In order to make the purpose, features, and advantages of the present disclosure more obvious and easy to understand, the technical solutions in the embodiments of the present disclosure will be clearly and completely described below in conjunction with the drawings in the embodiments of the present disclosure. Obviously, the described embodiments are only part of the embodiments of the present disclosure, not all of the embodiments. Based on the embodiments in the present disclosure, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present disclosure.
[0047] The following is an introduction to a fraud-related and illegal APP detection method and electronic device provided by this application in conjunction with the accompanying drawings.
[0048] like Figure 1 As shown, the embodiment of the present application provides a method for detecting fraudulent and illegal APPs, the method comprising:
[0049] S101, acquiring APP data, and obtaining data to be processed based on the APP data;
[0050] It is understandable that in the present application, APP data can be obtained through multiple channels, for example, it can be from various network channels, such as application stores, web crawlers, user reports, etc., and APP data can also be collected by APP dynamic and static detection engines.
[0051] In some optional embodiments, the APP data includes:
[0052] APP description, user comments, version history, development team, operation and maintenance team, download path, package name, application name, permission request, signature characteristics, class name characteristics, string characteristics, network communication information, application sensitive behavior, user privacy data access information, and integrated third-party SDK information. It is understandable that the APP in this application may also include APP data obtained by other means, and APP data may also be other types of data, which is not limited in this application.
[0053] S102, identifying entities and relationships between entities in the data to be processed, and constructing a knowledge graph using the entities as nodes and the relationships between the entities as edges;
[0054] It should be noted that entity and relationship extraction is a key step in building a knowledge graph. In this application, by identifying entities (nodes) and relationships (edges) between entities from the data to be processed, complex data information can be structured into graph data, which is convenient for subsequent storage, query and analysis.
[0055] It can be understood that entity extraction refers to identifying and extracting meaningful entities from the data to be processed as nodes. For fraud-related and illegal APP detection tools, common entities include but are not limited to applications (APP), developers (Developer), users (User), permissions (Permission), communication information (URL), behavior (Behavior), etc.
[0056] S103, obtaining a training set and a verification set based on the data set of the knowledge graph, and using the training set and the verification set to train and verify the graph attention neural network to obtain an APP detection model;
[0057] The knowledge graph is the graph structure data. It should be noted that this application annotates the graph structure data and then uses the annotated graph structure data as a data set. In this application, the data set is divided into a training set and a verification set. The graph attention neural network is trained with the training set to obtain an initial detection model, and then the performance of the initial detection model is verified. Then, when the performance of the initial detection model does not reach the target, the initial detection model is optimized. When the model parameters are optimized, the APP detection model is output.
[0058] S104, inputting the data of the APP to be detected into the APP detection model for detection, and outputting a detection result of whether the APP to be detected is a fraud-related and illegal APP.
[0059] The data of the APP to be detected is input into the APP detection model for detection, and the detection result of whether the APP to be detected is a fraud-related or illegal APP is obtained. In this application, the trained model is used to perform real-time detection on the new APP to determine whether it is fraud-related or illegal. The APP detection model can generate a detection report to explain the detection results.
[0060] In some optional embodiments, the acquiring APP data and obtaining the data to be processed based on the APP data includes:
[0061] Collect APP data through online channels or APP engines;
[0062] The APP data is cleaned and standardized to obtain data to be processed.
[0063] Specifically, this application collects APP data from various network channels (such as app stores, web crawlers, user reports, etc.) and APP dynamic and static detection engines, including: APP description, user comments, version history, development team, operation and maintenance team, download path, package name, application name, permission request, signature features, class name features, string features, network communication information, application sensitive behavior, user privacy data access information, integrated third-party SDK information, and other multi-dimensional data. Then the acquired APP data is preprocessed, specifically cleaning and standardizing the data, processing missing values and outliers, and performing text preprocessing (word segmentation, removing stop words, word vectorization, etc.).
[0064] In some optional embodiments, identifying entities in the data to be processed and the relationships between entities, and constructing a knowledge graph using the entities as nodes and the relationships between the entities as edges, includes:
[0065] Determine static data of the data to be processed, identify static entities in the static data, and first relationships between the static entities;
[0066] Determining dynamic data of the data to be processed, identifying dynamic entities in the dynamic data, and second relationships between the dynamic entities;
[0067] Graph structure data is constructed with static entities and dynamic entities as nodes and first relationships and second relationships as edges; the graph structure data is used as a knowledge graph.
[0068] Specifically, the entity extraction in the data to be processed in this application includes entities extracted from APP static data and entities extracted from APP dynamic data;
[0069] Among them, Figure 2 As shown in the figure, the entities extracted from the APP static data include:
[0070] APP: APP metadata obtained from the application market, such as name, version, size, developer information, release time, icon and package name.
[0071] Developer: Extract developer information from APP metadata, such as company name, contact information, registered address, etc.
[0072] System permissions: Extract the declared permissions from the APP's Manifest file, whether they are sensitive and whether they are applied dynamically.
[0073] Third-party libraries: Extract the used third-party libraries and frameworks from the decompiled code.
[0074] The entities extracted from the APP dynamic data include:
[0075] User: Extract user information from user behavior data, such as user name, user ID, device information, location, etc. The device includes account information, system information, location information, and hardware information.
[0076] Behavior: Extract behavioral information from runtime data, such as system calls, permission usage, etc.
[0077] Communication: Network information accessed during the APP operation, including Http, DNS, IP, port, return data, etc.
[0078] In this application, relationship extraction refers to identifying and extracting the relationship between entities from the original data. Relationship types can include: developers developing apps, users downloading apps, apps requesting permissions, apps accessing specific URLs, apps calling system APIs: specifically, system API call parameters and determining whether the system APIs are sensitive, etc. Similarly, the relationships between entities in the data to be processed include the relationships between entities extracted from the static data of the app and the relationships between entities extracted from the dynamic data of the app.
[0079] Among them, the relationships between entities extracted from APP static data include:
[0080] Developer-APP: Developer develops or maintains an APP.
[0081] APP-PERMISSIONS: APP declares and uses certain permissions.
[0082] APP-third-party library: APP uses some third-party libraries and frameworks. APP integrates third-party SDK to call system API or access devices.
[0083] The relationships between entities extracted from APP dynamic data include:
[0084] User-APP: A user downloads, installs or uses an APP.
[0085] APP-behavior: The behavior generated by the APP when it is running.
[0086] APP-URL: APP accesses certain specified URLs or DNS.
[0087] like Figure 3 As shown, this application first obtains APP samples and APP download pages of application stores through APP crawlers, and then sends APP samples to APP static engine and APP dynamic engine for judgment through APP judgment platform, and obtains APP static analysis results after passing through APP static engine; if APP is a shelled sample, it is unshelled after passing through APP dynamic engine to obtain APP original code, and the APP original code is input into APP static engine to obtain APP static analysis results; at the same time, the sample passes through APP dynamic engine to obtain APP operation data, where APP operation data includes network data, behavior data, etc. Then, the APP static analysis results and APP operation data are integrated and processed to obtain APP entity relationship information.
[0088] The page processing program (text recognition, NLP) of the app store APP download page obtains the APP developer information and APP release information. After integrating the APP developer information, APP release information, fraud-related information feature library and APP entity relationship information data, model training is performed to obtain the APP knowledge graph.
[0089] In some optional embodiments, the fraud-related illegal APP detection method provided by the present application further includes:
[0090] The graph structure data is stored in a graph database.
[0091] Specifically, in this application, graph structure data is constructed by using entities as nodes and relationships between entities as edges, and stored in a graph database. The graph structure database in this application can be Neo4j, which can efficiently store and manage complex graph structure data. Therefore, the knowledge graph in this application is a graph database that includes static information, dynamic information, and extension information of the APP.
[0092] In some optional embodiments, the data set based on the knowledge graph is used to obtain a training set and a validation set, including:
[0093] A graph attention network is used to perform representation learning on the graph structure data to obtain a high-dimensional vector;
[0094] The high-dimensional vector is labeled to obtain a training set and a validation set.
[0095] Specifically, in this application, the entities and relationships in the knowledge graph are represented as a graph structure, and a graph attention network (GAT) is used for representation learning to obtain a high-dimensional vector representation. Then, based on known fraud-related and illegal APPs, training data and verification data are prepared for supervised learning.
[0096] In some optional embodiments, the training set and the validation set are used to train and validate the graph attention neural network to obtain the APP detection model, including:
[0097] Using the training set to train the graph attention neural network to obtain an initial detection model;
[0098] Using the verification set to verify the initial detection model, to obtain a verification result;
[0099] Based on the verification results, the model parameters of the initial detection model are optimized by cross-validation until the model converges to obtain an APP detection model.
[0100] In this application, a training set is used to train the graph attention neural network to obtain an initial detection model, and then the initial detection model is verified using a validation set to obtain a verification result. Based on the verification result, the model parameters of the initial detection model are optimized by cross-validation until the model converges to obtain an APP detection model. It can be understood that the detection model is obtained after each cross-training, and then the detection model is evaluated. In this application, indicators such as accuracy, recall rate, and F1 value are used to evaluate the model performance until the optimal model parameters are obtained, thereby obtaining the optimal APP detection model, and outputting the optimal APP detection model when the optimal APP detection model is obtained.
[0101] In some optional embodiments, the fraud-related illegal APP detection method provided by the present application further includes:
[0102] The detection results are fed back into the knowledge graph to update the knowledge graph.
[0103] Specifically, in the present application, after the data of the APP to be detected is input into the APP detection model for detection and the detection result of whether the APP to be detected is a fraud-related and illegal APP is output, the detection result can be fed back to the knowledge graph, thereby updating the knowledge graph.
[0104] In some optional embodiments, the test result includes a test report.
[0105] In this application, the trained model is used to perform real-time detection of new apps to determine whether they are involved in fraud or violations. The trained model generates a detection report and explains the detection results.
[0106] The fraud-related and illegal APP detection method provided in this application uses crawlers, static detection, dynamic detection and other means to obtain APP feature information, and then cleans, sorts, classifies, disassembles, associates, and combines the APP feature information to construct a knowledge graph; then this application uses a graph neural network model to build and train the knowledge graph; finally, the APP detection model is used to detect fraud-related issues on new APPs.
[0107] like Figure 4 As shown, the embodiment of the present application provides a fraud-related illegal APP detection device, the device comprising:
[0108] The acquisition module 401 is used to acquire APP data and obtain data to be processed based on the APP data;
[0109] A construction module 402 is used to identify entities and relationships between entities in the data to be processed, and to construct a knowledge graph using the entities as nodes and the relationships between the entities as edges;
[0110] A training module 403 is used to obtain a training set and a verification set based on the data set of the knowledge graph, and use the training set and the verification set to train and verify the graph attention neural network to obtain an APP detection model;
[0111] The detection module 404 is used to input the data of the APP to be detected into the APP detection model for detection, and output a detection result of whether the APP to be detected is a fraud-related and illegal APP.
[0112] In some optional embodiments, the acquisition module 401 includes:
[0113] A collection unit is used to collect APP data through network channels or APP engines;
[0114] The processing unit is used to perform data cleaning and standardization on the APP data to obtain data to be processed.
[0115] In some optional embodiments, the building block 402 includes:
[0116] A first identification unit, configured to determine static data of the data to be processed, identify static entities in the static data, and first relationships between the static entities;
[0117] A second identification unit, configured to determine dynamic data of the data to be processed, identify dynamic entities in the dynamic data, and second relationships between the dynamic entities;
[0118] A construction unit is used to construct graph structure data using static entities and dynamic entities as nodes and first relationships and second relationships as edges; and use the graph structure data as a knowledge graph.
[0119] In some optional embodiments, the training module 403 includes:
[0120] A learning unit, used for performing representation learning on the graph structure data using a graph attention network to obtain a high-dimensional vector;
[0121] The labeling unit is used to label the high-dimensional vector to obtain a training set and a validation set.
[0122] In some optional embodiments, the training module 403 further includes:
[0123] A training unit, used to train the graph attention neural network using the training set to obtain an initial detection model;
[0124] A verification unit, used to verify the initial detection model using the verification set to obtain a verification result;
[0125] The optimization unit is used to optimize the model parameters of the initial detection model by cross-validation based on the verification result until the model converges to obtain the APP detection model.
[0126] Some optional embodiments further include:
[0127] An updating module is used to feed back the detection results to the knowledge graph to update the knowledge graph.
[0128] In some optional embodiments, the test result includes a test report.
[0129] Some optional embodiments further include:
[0130] A storage unit is used to store the graph structure data in a graph database.
[0131] The APP data includes:
[0132] APP description, user comments, version history, development team, operation and maintenance team, download path, package name, application name, permission request, signature characteristics, class name characteristics, string characteristics, network communication information, application sensitive behavior, user privacy data access information and integrated third-party SDK information.
[0133] According to an embodiment of the present disclosure, the present disclosure also provides an electronic device and a readable storage medium.
[0134] Figure 5A schematic block diagram of an example electronic device 500 that can be used to implement an embodiment of the present disclosure is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present disclosure described and / or required herein.
[0135] like Figure 5 As shown, the device 500 includes a computing unit 501, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 502 or a computer program loaded from a storage unit 508 into a random access memory (RAM) 503. In the RAM 503, various programs and data required for the operation of the device 500 can also be stored. The computing unit 501, the ROM 502, and the RAM 503 are connected to each other via a bus 504. An input / output (I / O) interface 505 is also connected to the bus 504.
[0136] A number of components in the device 500 are connected to the I / O interface 505, including: an input unit 506, such as a keyboard, a mouse, etc.; an output unit 507, such as various types of displays, speakers, etc.; a storage unit 508, such as a disk, an optical disk, etc.; and a communication unit 509, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 509 allows the device 500 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.
[0137] The computing unit 501 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the computing unit 501 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, digital signal processors (DSPs), and any appropriate processors, controllers, microcontrollers, etc. The computing unit 501 performs the various methods and processes described above, such as a fraud-related illegal APP detection method. For example, in some embodiments, the fraud-related illegal APP detection method may be implemented as a computer software program, which is tangibly contained in a machine-readable medium, such as a storage unit 508. In some embodiments, part or all of the computer program may be loaded and / or installed on the device 500 via ROM 502 and / or communication unit 509. When the computer program is loaded into RAM 503 and executed by the computing unit 501, one or more steps of the fraud-related illegal APP detection method described above may be performed. Alternatively, in other embodiments, the computing unit 501 may be configured to perform the fraud-related illegal APP detection method in any other appropriate manner (e.g., by means of firmware).
[0138] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.
[0139] The program code for implementing the method of the present disclosure may be written in any combination of one or more programming languages. These program codes may be provided to a processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that the program code, when executed by the processor or controller, enables the functions / operations specified in the flow chart and / or block diagram to be implemented. The program code may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.
[0140] In the context of the present disclosure, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or equipment. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or equipment, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.
[0141] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user can provide input to the computer. Other types of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).
[0142] The systems and techniques described herein may be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such back-end components, middleware components, or front-end components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), and the Internet.
[0143] A computer system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The relationship of client and server is generated by computer programs running on respective computers and having a client-server relationship with each other. The server may be a cloud server, a server of a distributed system, or a server combined with a blockchain.
[0144] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps recorded in this disclosure can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solutions disclosed in this disclosure can be achieved, and this document does not limit this.
[0145] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of the features. In the description of the present disclosure, the meaning of "plurality" is two or more, unless otherwise clearly and specifically defined.
[0146] The above is only a specific embodiment of the present disclosure, but the protection scope of the present disclosure is not limited thereto. Any person skilled in the art who is familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed in the present disclosure, which should be included in the protection scope of the present disclosure. Therefore, the protection scope of the present disclosure should be based on the protection scope of the claims.
Claims
1. A method for detecting fraudulent and illegal APPs, characterized in that: The method comprises: Acquire APP data, and obtain data to be processed based on the APP data; Identify entities and relationships between entities in the data to be processed, and construct a knowledge graph using the entities as nodes and the relationships between the entities as edges; A training set and a validation set are obtained based on the data set of the knowledge graph, and the graph attention neural network is trained and validated using the training set and the validation set to obtain an APP detection model; The data of the APP to be detected is input into the APP detection model for detection, and a detection result of whether the APP to be detected is a fraud-related and illegal APP is output.
2. The method according to claim 1, characterized in that The acquiring of APP data and obtaining the data to be processed based on the APP data includes: Collect APP data through online channels or APP engines; The APP data is cleaned and standardized to obtain data to be processed.
3. The method according to claim 1, characterized in that The identifying entities and the relationships between entities in the data to be processed, and constructing a knowledge graph using the entities as nodes and the relationships between the entities as edges, includes: Determine static data of the data to be processed, identify static entities in the static data, and first relationships between the static entities; Determining dynamic data of the data to be processed, identifying dynamic entities in the dynamic data, and second relationships between the dynamic entities; Graph structure data is constructed with static entities and dynamic entities as nodes and first relationships and second relationships as edges; the graph structure data is used as a knowledge graph.
4. The method according to claim 3, characterized in that The data set based on the knowledge graph obtains a training set and a validation set, including: A graph attention network is used to perform representation learning on the graph structure data to obtain a high-dimensional vector; The high-dimensional vector is labeled to obtain a training set and a validation set.
5. The method according to claim 4, characterized in that The method of using the training set and the validation set to train and validate the graph attention neural network to obtain an APP detection model includes: Using the training set to train the graph attention neural network to obtain an initial detection model; Using the verification set to verify the initial detection model, to obtain a verification result; Based on the verification results, the model parameters of the initial detection model are optimized by cross-validation until the model converges to obtain an APP detection model.
6. The method according to claim 1, characterized in that Also includes: The detection results are fed back into the knowledge graph to update the knowledge graph.
7. The method according to claim 1, characterized in that The test results include a test report.
8. The method according to claim 3, characterized in that Also includes: The graph structure data is stored in a graph database.
9. The method according to claim 1, characterized in that: The APP data includes: APP description, user comments, version history, development team, operation and maintenance team, download path, package name, application name, permission request, signature characteristics, class name characteristics, string characteristics, network communication information, application sensitive behavior, user privacy data access information and integrated third-party SDK information.
10. An electronic device, characterized in that: include: at least one processor; as well as a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method according to any one of claims 1 to 9.
Citation Information
Patent Citations
Internet financial gang fraud behavior detection method based on knowledge graph
CN112053221A
Application program detection method and device, electronic equipment and storage medium
CN113760764A
Application program detection method and device and electronic equipment
CN113849812A
Malicious software behavior detection and classification system based on deep learning
CN113961922A
Application program detection method and device and electronic equipment
CN118228258A
