Fraud number identification method and device, equipment and storage medium

Abstract

The embodiment of the invention discloses a fraud number identification method and device, equipment and a storage medium, and the method comprises the steps: obtaining the target number attribute information of a target number; wherein the target number attribute information comprises at least one of call frequency information, call duration information, call region information, roaming information and home city dispersion information; determining a target fraud number recognition model corresponding to the target number according to the target number attribute information; and inputting the target number attribute information into the target fraud number recognition model to obtain a target fraud recognition result of the target number. According to the technical scheme provided by the embodiment of the invention, the problem that the recognition accuracy is easy to be insufficient when a general model is used for recognizing the fraud number in the prior art is solved, the corresponding target fraud number recognition model can be matched based on the attribute information of the target number, and then the target number is recognized based on the target fraud number recognition model, so that the recognition accuracy is improved. And the fraud number identification accuracy is improved.

Description

Technical Field

[0001] Embodiments of the present invention relate to the technical field of number analysis, and in particular to a fraudulent number identification method, device, equipment and storage medium. Background Art

[0002] The automated technology for classifying fraudulent number labels is mainly aimed at the current rampant telecommunications fraud, which has begun to gradually endanger social stability. This method is a series of technical means developed to deal with these potential dangers. There are now relatively mature technical response policies for telecommunications fraud, which aim to prevent and combat telecommunications fraud through technical means and management measures.

[0003] Existing technologies can use big data technology to analyze telecommunications network data in real time to identify abnormal behaviors and potential fraudulent activities. Through machine learning models, machine learning models are developed and applied to automatically identify and classify fraudulent numbers. However, existing technologies identify fraudulent numbers based on a general model, but the fraud methods and fraud methods corresponding to the numbers are diverse, and different numbers may have their own unique fraudulent characteristics. The accuracy of fraudulent number identification based on a general model needs to be improved. Summary of the invention

[0004] The embodiments of the present invention provide a fraudulent number identification method, apparatus, device and storage medium. The technical solution of the embodiments of the present invention can match the corresponding target fraudulent number identification model based on the attribute information of the target number, and then identify the target number based on the target fraudulent number identification model, thereby improving the accuracy of fraudulent number identification.

[0005] In a first aspect, an embodiment of the present invention provides a method for identifying a fraudulent number, the method comprising:

[0006] Obtain target number attribute information of the target number; wherein the target number attribute information includes: at least one of call count information, call duration information, call region information, roaming information and home city dispersion information; determine a target fraud number recognition model corresponding to the target number based on the target number attribute information; input the target number attribute information into the target fraud number recognition model to obtain a target fraud recognition result of the target number.

[0007] In a second aspect, an embodiment of the present invention provides a fraudulent number identification device, the device comprising:

[0008] An attribute information acquisition module is used to obtain target number attribute information of a target number; wherein the target number attribute information includes: at least one of call count information, call duration information, call region information, roaming information and home city dispersion information; an identification model determination module is used to determine a target fraud number identification model corresponding to the target number based on the target number attribute information; a fraud number identification module is used to input the target number attribute information into the target fraud number identification model to obtain a target fraud identification result of the target number.

[0009] In a third aspect, an embodiment of the present invention provides a computer device, the computer device comprising:

[0010] one or more processors;

[0011] A memory for storing one or more programs;

[0012] When the one or more programs are executed by the one or more processors, the one or more processors implement the fraudulent number identification method described in any embodiment.

[0013] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the fraudulent number identification method described in any embodiment.

[0014] The technical solution provided by the embodiment of the present invention obtains the target number attribute information of the target number; wherein the target number attribute information includes at least one of: number of calls information, call duration information, call region information, roaming information and home city dispersion information; according to the target number attribute information, the target fraud number recognition model corresponding to the target number is determined; the target number attribute information is input into the target fraud number recognition model to obtain the target fraud recognition result of the target number. The technical solution of the embodiment of the present invention solves the problem that the recognition of fraud numbers by the common model in the prior art is prone to insufficient recognition accuracy. The corresponding target fraud number recognition model can be matched based on the attribute information of the target number, and then the target number can be recognized based on the target fraud number recognition model, thereby improving the recognition accuracy of fraud numbers. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 This is a flow chart of a fraud number identification method provided by an embodiment of the present invention;

[0016] Figure 2 This is a flow chart of another fraud number identification method provided by an embodiment of the present invention;

[0017] Figure 3This is a workflow diagram for identifying fraudulent numbers provided by an embodiment of the present invention;

[0018] Figure 4 This is another workflow diagram for identifying fraudulent numbers provided by an embodiment of the present invention;

[0019] Figure 5 It is a structural schematic diagram of a fraud number identification device provided by an embodiment of the present invention;

[0020] Figure 6 It is a structural schematic diagram of a computer device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0021] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0022] Figure 1 This is a flow chart of a fraudulent number identification method provided by an embodiment of the present invention. The embodiment of the present invention can be applied to a scenario in which whether an input number is a fraudulent number is identified. The method can be executed by a fraudulent number identification device, which can be implemented by software and / or hardware.

[0023] like Figure 1 As shown, the fraud number identification method includes the following steps:

[0024] S110. Obtain target number attribute information of the target number.

[0025] Among them, the target number may be a number that needs to be identified as a fraudulent number. Specifically, the target number can be determined manually, and the method for determining the target number is not limited here. Further, the target number attribute information may be information about the call attributes of the target number. Specifically, the target number attribute information includes: at least one of call count information, call duration information, call region information, roaming information and home city dispersion information. The call count information at least includes: the number of times the number dials multiple types of service provider numbers; the call duration information at least includes: the single duration and maximum duration of the number dialing the phone; the call region information at least includes: the proportion of same-city calls, the proportion of repeated numbers on the other end, the proportion of passing time and the proportion of fixed call numbers; the roaming information at least includes: roaming call count, the number of roaming cities and the number of roaming provinces; the home city dispersion information at least includes: the calling home city dispersion, the called home city dispersion, the call city dispersion and the number dispersion. Exemplarily, the target number attribute information of the target number can be read from the preset database.

[0026] S120. Determine a target fraud number recognition model corresponding to the target number according to the target number attribute information.

[0027] The target fraud number recognition model may be a model for identifying whether the target number is a fraud number. Specifically, the target number attribute information may be analyzed to determine the fraud type of the target number, and then the corresponding fraud number recognition model may be matched according to the fraud type of the target number, and the matched fraud number recognition model may be used as the target fraud number recognition model.

[0028] S130: Input the target number attribute information into the target fraud number identification model to obtain a target fraud identification result of the target number.

[0029] The target fraud identification result may be an identification result of whether the target number is a fraud number. Specifically, the target fraud identification result may include a fraud number and a non-fraud number. Specifically, the target number feature may be input into a target fraud number identification model so that the model analyzes the target number feature to obtain the target fraud identification result.

[0030] The technical solution provided by the embodiment of the present invention obtains the target number attribute information of the target number; wherein the target number attribute information includes at least one of: number of calls information, call duration information, call region information, roaming information and home city dispersion information; according to the target number attribute information, the target fraud number recognition model corresponding to the target number is determined; the target number attribute information is input into the target fraud number recognition model to obtain the target fraud recognition result of the target number. The technical solution of the embodiment of the present invention solves the problem that the recognition of fraud numbers by the common model in the prior art is prone to insufficient recognition accuracy. The corresponding target fraud number recognition model can be matched based on the attribute information of the target number, and then the target number can be recognized based on the target fraud number recognition model, thereby improving the recognition accuracy of fraud numbers.

[0031] Figure 2 It is another flow chart of a fraud number identification method provided by an embodiment of the present invention. The embodiment of the present invention can be applied to the scenario of identifying whether an input number is a fraud number. Based on the above embodiment, this embodiment further illustrates how to determine a target fraud number identification model corresponding to a target number according to target number attribute information; and how to obtain a target fraud identification result of a target number based on the target fraud number identification model. The device can be implemented by software and / or hardware and integrated into a computer device with application development function.

[0032] like Figure 2 As shown, the fraud number identification method includes the following steps:

[0033] S210. Obtain target number attribute information of the target number.

[0034] S220. Determine the target fraud type of the target number according to the target number attribute information.

[0035] Among them, the target fraud type may be a fraud type that may be associated with the target number. Specifically, multiple preset fraud types may be set in advance, and then the target number attribute information of the target number may be analyzed to classify the target number into the corresponding preset fraud type, and the classified preset fraud type is used as the target fraud type. By determining the target fraud type corresponding to the target number, it is convenient to subsequently match the adaptive fraud number recognition model according to the target fraud type, thereby improving the accuracy of fraud number recognition.

[0036] S230, matching the sample fraud number recognition model corresponding to the target fraud-related type, and using the matched sample fraud number recognition model as the target fraud number recognition model.

[0037] The sample fraud number recognition model may be a model for identifying whether an input number is a fraud number. Specifically, each fraud type may have a corresponding sample fraud number recognition model, which may have a higher recognition accuracy when identifying numbers of the corresponding fraud type. The target fraud number recognition model includes: a number feature determination network and a number feature analysis network.

[0038] Optionally, the training process of the sample fraud number recognition model includes: obtaining a preliminary fraud number recognition model and sample number attribute information of multiple preset sample numbers; classifying the fraud types of the preset sample numbers based on the sample number attribute information to obtain multiple sample fraud types; for each sample fraud type, training the preliminary fraud number recognition model based on the sample number attribute information of the preset sample numbers in the sample fraud type to obtain a sample fraud number recognition model corresponding to the sample fraud type.

[0039] Among them, the preset sample number may be a preset sample number for training a fraud number recognition model. The sample number attribute information may be information about the call attributes of the sample number. Specifically, the sample number attribute information may also include at least one of call count information, call duration information, call region information, roaming information, and home city dispersion information. Furthermore, the sample fraud type may be a fraud type associated with the sample number. Specifically, the sample number attribute information of the sample number may be analyzed, the sample number may be classified into a corresponding preset fraud type, and the classified preset fraud type may be used as the sample fraud type.

[0040] Furthermore, the preliminary fraudulent number recognition model may be an original fraudulent number recognition model that has not been pre-trained. The preset sample numbers may contain corresponding number labels for evaluating the recognition results of the initial fraudulent number recognition model and adjusting the model according to the evaluation results.

[0041] Optionally, a preliminary fraud number recognition model is trained based on sample number attribute information of preset sample numbers in sample fraud-related fraud types to obtain a sample fraud number recognition model corresponding to the sample fraud-related fraud types, including: adding the sample number attribute information of the preset sample numbers to the preliminary fraud number recognition model to obtain a sample fraud recognition result; determining a recognition loss function based on the sample fraud recognition result and the preset number label of the preset sample number; and adjusting the preliminary fraud number recognition model based on the recognition loss function to obtain a sample fraud number recognition model.

[0042] The sample fraud identification result may be the identification result of the preliminary fraud number identification model on whether the preset sample number is a fraud number. The preset number label may be a label in the preset sample number that is pre-marked as to whether it is a fraud number. By comparing the sample fraud identification result with the preset number label of the preset sample number, the error of the preliminary fraud number identification model in identifying the fraud number can be known.

[0043] Further, the recognition loss function can be determined according to the comparison result. The recognition loss function can be a loss function corresponding to the recognition error. The error size can be known through the recognition loss function. Subsequently, the preliminary fraudulent number recognition model can be adjusted according to the recognition loss function to obtain the target fraudulent number recognition model. Specifically, the model parameters in the preliminary fraudulent number recognition model can be adjusted according to the recognition loss function to obtain the target fraudulent number recognition model.

[0044] S240: Input the target number attribute information into the number feature determination network of the target fraud number identification model to obtain the target number feature.

[0045] The number feature determination network may be a network for extracting number features of an input number. The target number feature may be a common feature corresponding to the target number. Specifically, the target number feature includes at least one of a number of calls feature, a call duration feature, a call region feature, a roaming feature, and a home city dispersion feature. By inputting the target number attribute information into the number feature determination network, the network may analyze the target number attribute information and obtain the target number feature.

[0046] S250: Input the target number feature into the number feature analysis network of the target fraud number recognition model to obtain the target fraud recognition result of the target number.

[0047] The number feature analysis model may be a network for analyzing the correlation between number features and fraud features. The target fraud identification result may be an identification result of whether the target number is a fraud number. Specifically, the target fraud identification result may include a fraud number and a non-fraud number. Specifically, the target number features may be input into the number feature analysis network so that the network analyzes the target number features, thereby obtaining the target fraud identification result.

[0048] Optionally, after obtaining the target fraud identification result of the target number, if the target fraud identification result is a fraud number, the target number is sent to a preset number management system so that the preset number management system adds the target number to a fraud number list.

[0049] The preset number management system may be a preset system for managing fraudulent numbers. If the target fraud identification result is a fraudulent number, the target number may be added to a fraudulent number list, so that when the number fraud is identified later, the number to be identified is first matched with the number in the fraudulent number list. If the match is successful, the identification result may be directly determined, thereby improving the efficiency of fraudulent number identification.

[0050] For example, in order to better understand the technical solution provided by the present invention, a specific embodiment is introduced below. Figure 3 FIG. 1 is a flowchart of a method for identifying fraudulent numbers provided by an embodiment of the present invention. Figure 3 As shown, the workflow for fraud number identification includes the following steps:

[0051] Data collection: Collect information on fraudulent numbers involved in cases, 12321 reports, etc., and collect whitelist users. Analyze the call records of related numbers to extract features such as number of calls, call duration, roaming information, etc.

[0052] Data preprocessing: clean, label, normalize, perform variance test, importance test, outlier detection, and null value processing on the collected data.

[0053] Construct a fraud-related feature tree: Based on the features of fraud-related numbers and whitelist features, a fraud-related feature label tree is formed.

[0054] Feature extraction: Extract the following features from the fraudulent numbers: a. Number of calls: the number of times the fraudulent numbers dialed numbers of various types of service providers. b. Call duration: the single duration and maximum duration of calls made by the fraudulent numbers. c. Call location ratio: the proportion of the same city involved, the proportion of repeated numbers on the other end, the proportion of call duration, and the proportion of fixed-line phones. d. Roaming information: roaming call count, the number of roaming cities, and the number of roaming provinces. e. Dispersion of the city of origin: the dispersion of the calling city of origin, the dispersion of the called city of origin, the dispersion of the city of origin, and the dispersion of the number. The construction of data features is a key step that affects the training effect of the model. How to extract the required features from the original data and construct new features is the biggest technical challenge. The required features can be constructed according to the actual business scenarios, and the variables can be fused in combination with the professional knowledge of the corresponding business to improve the richness of the information contained in the features. Pay attention to data security when extracting features, and desensitize sensitive data.

[0055] Constructing a feature label tree: Constructing a feature label tree (also called a classification tree) is to classify numbers by gradually analyzing the relationships between various features.

[0056] Fraud-related intelligence correlation analysis: This is one of the important technical means to combat telecommunications network fraud. By conducting a comprehensive analysis of multi-source heterogeneous intelligence data, it reveals hidden fraud patterns and behavioral characteristics, providing strong support for preventing and combating telecommunications network fraud. Specifically include: a. Feature selection: Feature selection is performed through the Gini coefficient. B. Decision tree classification: Use the decision tree algorithm to construct a tree model, grow and split the tree through the relationship between different features, classify the numbers in turn, and construct a feature model of fraud-related numbers.

[0057] Result output: The identified fraudulent numbers are provided to the management department through the fraudulent number feature model, and the persons associated with the fraudulent intelligence are cracked down and dealt with.

[0058] For example, Figure 4 This is another workflow diagram for identifying fraudulent numbers provided by an embodiment of the present invention. Figure 4 As shown, the workflow for fraud number identification includes the following steps: first, select models for different business scenarios, and then collect data through various channels; then adjust the model parameters, and then perform data segmentation and selection; then, perform machine learning algorithm and neural network algorithm analysis; finally, provide multiple ways to output the results.

[0059] The technical solution provided by the embodiment of the present invention obtains the target number attribute information of the target number; determines the target fraud type of the target number according to the target number attribute information; matches the sample fraud number recognition model corresponding to the target fraud type, and uses the matched sample fraud number recognition model as the target fraud number recognition model; inputs the target number attribute information into the number feature determination network of the target fraud number recognition model to obtain the target number feature; inputs the target number feature into the number feature analysis network of the target fraud number recognition model to obtain the target fraud recognition result of the target number. The technical solution of the embodiment of the present invention solves the problem that the recognition of fraud numbers by the general model in the prior art is prone to insufficient recognition accuracy. The corresponding target fraud number recognition model can be matched based on the attribute information of the target number, and then the target number can be recognized based on the target fraud number recognition model, thereby improving the recognition accuracy of fraud numbers.

[0060] Figure 5 It is a structural schematic diagram of a fraud number identification device provided by an embodiment of the present invention. The embodiment of the present invention can be applied to the scenario of identifying whether an input number is a fraud number. The device can be implemented by software and / or hardware and integrated into a computer device with application development function.

[0061] like Figure 5As shown, the fraudulent number identification device includes: an attribute information acquisition module 310, a recognition model determination module 320 and a fraudulent number identification module 330.

[0062] Among them, the attribute information acquisition module 310 is used to obtain the target number attribute information of the target number; wherein the target number attribute information includes: at least one of call count information, call duration information, call region information, roaming information and home city dispersion information; the recognition model determination module 320 is used to determine the target fraud number recognition model corresponding to the target number according to the target number attribute information; the fraud number recognition module 330 is used to input the target number attribute information into the target fraud number recognition model to obtain the target fraud recognition result of the target number.

[0063] The technical solution provided by the embodiment of the present invention obtains the target number attribute information of the target number; wherein the target number attribute information includes at least one of: number of calls information, call duration information, call region information, roaming information and home city dispersion information; according to the target number attribute information, the target fraud number recognition model corresponding to the target number is determined; the target number attribute information is input into the target fraud number recognition model to obtain the target fraud recognition result of the target number. The technical solution of the embodiment of the present invention solves the problem that the recognition of fraud numbers by the common model in the prior art is prone to insufficient recognition accuracy. The corresponding target fraud number recognition model can be matched based on the attribute information of the target number, and then the target number can be recognized based on the target fraud number recognition model, thereby improving the recognition accuracy of fraud numbers.

[0064] In an optional implementation, the recognition model determination module 320 is specifically used to: determine the target fraud type of the target number based on the target number attribute information; match the sample fraud number recognition model corresponding to the target fraud type, and use the matched sample fraud number recognition model as the target fraud number recognition model.

[0065] In an optional implementation, the target fraud number identification model includes: a number feature determination network and a number feature analysis network, and the fraud number identification module 330 is specifically used to: input the target number attribute information into the number feature determination network of the target fraud number identification model to obtain the target number features; wherein the target number features include: at least one of call number features, call duration features, call region features, roaming features and home city dispersion features; input the target number features into the number feature analysis network of the target fraud number identification model to obtain the target fraud identification result of the target number.

[0066] In an optional embodiment, the fraud number identification device also includes: a sample fraud number identification model training module, which is used to: obtain a preliminary fraud number identification model and sample number attribute information of multiple preset sample numbers; classify the fraud types of the preset sample numbers based on the sample number attribute information to obtain multiple sample fraud types; for each sample fraud type, based on the sample number attribute information of the preset sample numbers in the sample fraud type, train the preliminary fraud number identification model to obtain a sample fraud number identification model corresponding to the sample fraud type.

[0067] In an optional embodiment, the sample fraud number recognition model training module includes: a model adjustment unit, used to: add the sample number attribute information of the preset sample number to the preliminary fraud number recognition model to obtain a sample fraud recognition result; determine an identification loss function based on the sample fraud recognition result and the preset number label of the preset sample number; adjust the preliminary fraud number recognition model according to the identification loss function to obtain the sample fraud number recognition model.

[0068] In an optional embodiment, the fraud number identification device also includes: a fraud number processing module, which is used to: after obtaining the target fraud identification result of the target number, if the target fraud identification result is a fraud number, send the target number to a preset number management system, so that the preset number management system adds the target number to the fraud number list.

[0069] In an optional implementation, the call frequency information includes at least: the number of times a number dials multiple types of service provider numbers; the call duration information includes at least: the single duration and maximum duration of a call made by the number; the call region information includes at least: the proportion of same-city calls, the proportion of repeated numbers on the other end, the proportion of passing time and the proportion of fixed call numbers; the roaming information includes at least: roaming call counts, the number of roaming cities and the number of roaming provinces; the home city dispersion information includes at least: the calling city dispersion, the called city dispersion, the call city dispersion and the number dispersion.

[0070] The fraudulent number identification device provided in the embodiment of the present invention can execute the fraudulent number identification method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0071] Figure 6 A schematic diagram of the structure of a computer device provided in an embodiment of the present invention. Figure 6 A block diagram of an exemplary computer device 12 suitable for use in implementing embodiments of the present invention is shown. Figure 6The computer device 12 shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present invention. The computer device 12 can be any terminal device with computing capability and can be configured in the fraud number identification device.

[0072] like Figure 6 As shown, the computer device 12 is in the form of a general-purpose computing device. The components of the computer device 12 may include, but are not limited to: one or more processors or processing units 16, a system memory 28, and a bus 18 that connects various system components (including the system memory 28 and the processing unit 16).

[0073] The bus 18 may be one or more of several types of bus structures, including a memory bus or memory controller, a peripheral bus, an accelerated graphics port, a processor or a local bus using any of a variety of bus architectures. For example, these architectures include, but are not limited to, an Industry Standard Architecture (ISA) bus, a Micro Channel Architecture (MAC) bus, an Enhanced ISA bus, a Video Electronics Standards Association (VESA) local bus, and a Peripheral Component Interconnect (PCI) bus.

[0074] The computer device 12 typically includes a variety of computer system readable media. These media can be any available media that can be accessed by the computer device 12, including volatile and non-volatile media, removable and non-removable media.

[0075] The system memory 28 may include computer system readable media in the form of volatile memory, such as random access memory (RAM) 30 and / or cache 32. The computer device 12 may further include other removable / non-removable, volatile / non-volatile computer system storage media. By way of example only, the storage system 34 may be used to read and write non-removable, non-volatile magnetic media ( Figure 6 not shown, usually called a "hard drive"). Although Figure 6 Not shown in the figure, a disk drive for reading and writing to a removable non-volatile disk (e.g., a "floppy disk"), and an optical disk drive for reading and writing to a removable non-volatile optical disk (e.g., a CD-ROM, a DVD-ROM, or other optical media) may be provided. In these cases, each drive may be connected to the bus 18 via one or more data medium interfaces. The system memory 28 may include at least one program product having a set (e.g., at least one) of program modules that are configured to perform the functions of the various embodiments of the present invention.

[0076] A program / utility 40 having a set (at least one) of program modules 42 may be stored, for example, in system memory 28, such program modules 42 including, but not limited to, an operating system, one or more application programs, other program modules, and program data, each of which or some combination may include an implementation of a network environment. Program modules 42 generally perform the functions and / or methods of the embodiments described herein.

[0077] The computer device 12 may also communicate with one or more external devices 14 (e.g., keyboards, pointing devices, displays 24, etc.), one or more devices that enable a user to interact with the computer device 12, and / or any device that enables the computer device 12 to communicate with one or more other computing devices (e.g., network cards, modems, etc.). Such communication may be performed through an input / output (I / O) interface 22. Furthermore, the computer device 12 may also communicate with one or more networks (e.g., local area networks (LANs), wide area networks (WANs), and / or public networks, such as the Internet) through a network adapter 20. Figure 6 As shown, the network adapter 20 communicates with other modules of the computer device 12 via the bus 18. It should be understood that although Figure 6 Not shown, other hardware and / or software modules may be used in conjunction with computer device 12, including but not limited to microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0078] The processing unit 16 executes various functional applications and data processing by running the program stored in the system memory 28, for example, implementing the fraudulent number identification method provided in the embodiment of the present invention, which includes:

[0079] Obtain target number attribute information of the target number; wherein the target number attribute information includes: at least one of call count information, call duration information, call region information, roaming information and home city dispersion information; determine a target fraud number recognition model corresponding to the target number based on the target number attribute information; input the target number attribute information into the target fraud number recognition model to obtain a target fraud recognition result of the target number.

[0080] This embodiment provides a computer-readable storage medium, on which a computer program is stored. When the program is executed by a processor, the fraud number identification method provided by any embodiment of the present invention is implemented, including:

[0081] Obtain target number attribute information of the target number; wherein the target number attribute information includes: at least one of call count information, call duration information, call region information, roaming information and home city dispersion information; determine a target fraud number recognition model corresponding to the target number based on the target number attribute information; input the target number attribute information into the target fraud number recognition model to obtain a target fraud recognition result of the target number.

[0082] The computer storage medium of the embodiment of the present invention can adopt any combination of one or more computer-readable media. The computer-readable medium can be a computer-readable signal medium or a computer-readable storage medium. The computer-readable storage medium can be, for example, but not limited to: an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples (non-exhaustive list) of computer-readable storage media include: an electrical connection with one or more wires, 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 above. In this document, a computer-readable storage medium can be any tangible medium containing or storing a program, which can be used by an instruction execution system, device or device or used in combination with it.

[0083] Computer-readable signal media may include data signals propagated in baseband or as part of a carrier wave, which carry computer-readable program code. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Computer-readable signal media may also be any computer-readable medium other than a computer-readable storage medium, which may send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device.

[0084] The program code embodied on the computer readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.

[0085] Computer program code for performing the operations of the present invention may be written in one or more programming languages ​​or combinations thereof, including object-oriented programming languages ​​such as Java, Smalltalk, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0086] It should be understood by those skilled in the art that the modules or steps of the present invention described above can be implemented by a general-purpose computing device, they can be concentrated on a single computing device, or distributed on a network composed of multiple computing devices, optionally, they can be implemented by a program code executable by a computer device, so that they can be stored in a storage device and executed by the computing device, or they can be made into individual integrated circuit modules, or multiple modules or steps therein can be made into a single integrated circuit module for implementation. Thus, the present invention is not limited to any specific combination of hardware and software.

[0087] Note that the above are only preferred embodiments of the present invention and the technical principles used. Those skilled in the art will understand that the present invention is not limited to the specific embodiments herein, and that various obvious changes, readjustments and substitutions can be made by those skilled in the art without departing from the scope of protection of the present invention. Therefore, although the present invention has been described in more detail through the above embodiments, the present invention is not limited to the above embodiments, and may include more other equivalent embodiments without departing from the concept of the present invention, and the scope of the present invention is determined by the scope of the appended claims.

Claims

1. A fraud number identification method, characterized in that: include: Acquire target number attribute information of the target number; wherein the target number attribute information includes: at least one of call number information, call duration information, call region information, roaming information and home city dispersion information; Determine a target fraud number recognition model corresponding to the target number according to the target number attribute information; The target number attribute information is input into the target fraud number identification model to obtain the target fraud identification result of the target number.

2. The method according to claim 1, characterized in that The target fraud number recognition model includes: a number feature determination network and a number feature analysis network. The target number attribute information is input into the target fraud number recognition model to obtain the target fraud recognition result of the target number, including: Inputting the target number attribute information into the number feature determination network of the target fraud number identification model to obtain the target number feature; wherein the target number feature includes: at least one of the following features: number of calls feature, call duration feature, call region feature, roaming feature and home city dispersion feature; The target number feature is input into the number feature analysis network of the target fraud number recognition model to obtain the target fraud recognition result of the target number.

3. The method according to claim 1, characterized in that The step of determining a target fraud number identification model corresponding to the target number according to the target number attribute information includes: Determining the target fraud type of the target number according to the target number attribute information; The sample fraud number recognition model corresponding to the target fraud-related type is matched, and the matched sample fraud number recognition model is used as the target fraud number recognition model.

4. The method according to claim 3, characterized in that The training process of the sample fraud number recognition model includes: Obtaining a preliminary fraud number recognition model and sample number attribute information of a plurality of preset sample numbers; Classifying the fraud-related types of the preset sample numbers based on the sample number attribute information to obtain a plurality of sample fraud-related types; For each sample fraud type, the preliminary fraud number recognition model is trained based on the sample number attribute information of the preset sample number in the sample fraud type to obtain a sample fraud number recognition model corresponding to the sample fraud type.

5. The method according to claim 4, characterized in that The training of the preliminary fraud number recognition model based on the sample number attribute information of the preset sample number in the sample fraud type to obtain the sample fraud number recognition model corresponding to the sample fraud type includes: Adding the sample number attribute information of the preset sample number to the preliminary fraud number recognition model to obtain a sample fraud recognition result; Determine an identification loss function according to the sample fraud identification result and the preset number label of the preset sample number; The preliminary fraudulent number recognition model is adjusted according to the recognition loss function to obtain the sample fraudulent number recognition model.

6. The method according to claim 5, characterized in that After obtaining the target fraud identification result of the target number, the method further includes: In the case where the target fraud identification result is a fraud number, the target number is sent to a preset number management system so that the preset number management system adds the target number to a fraud number list.

7. The method according to claim 1, characterized in that The call frequency information includes at least: the number of times the number dials multiple types of service provider numbers; the call duration information includes at least: the single duration and maximum duration of the number dialing a call; the call region information includes at least: the proportion of same-city calls, the proportion of repeated numbers on the other end, the proportion of passing time and the proportion of fixed call numbers; the roaming information includes at least: roaming call count, the number of roaming cities and the number of roaming provinces; the home city dispersion information includes at least: the calling city dispersion, the called city dispersion, the call city dispersion and the number dispersion.

8. A fraud number identification device, characterized in that: The device comprises: An attribute information acquisition module, used to acquire target number attribute information of a target number; wherein the target number attribute information includes: at least one of call number information, call duration information, call region information, roaming information and home city dispersion information; An identification model determination module, used to determine a target fraud number identification model corresponding to the target number according to the target number attribute information; The fraud number identification module is used to input the target number attribute information into the target fraud number identification model to obtain the target fraud identification result of the target number.

9. A computer device, characterized in that: The computer device comprises: one or more processors; A memory for storing one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the fraudulent number identification method as described in any one of claims 1-7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the fraud number identification method as described in any one of claims 1 to 7 is implemented.

Images