Risk prevention and control method and device based on willingness classification mining model, equipment and storage medium

Abstract

The invention discloses a risk prevention and control method and device based on a willingness classification mining model, equipment and a storage medium, a self-supervision mode is realized by innovatively identifying a classification model based on willingness and utilizing a BERT pre-training mechanism, and a small amount of annotation data in the initial stage is utilized to carry out BERT model training so as to achieve 80% accurate willingness identification. After online production, the real willingness of the user is utilized to continuously accumulate willingness data for labeling, the model is repeatedly trained, and through multiple iterations, the accurate recognition degree of the willingness classification model reaches 90%. A manual call-out scene is completely replaced in an anti-fraud scene in a risk control platform, customer service personnel are liberated, and the risk identification accuracy is greatly improved. According to the application of the method in the anti-fraud scene, the risk identification capability for the fraud scene of the original score is supplemented, and the method plays an important role in capability supplement of the whole risk control platform.

Description

Technical Field

[0001] The present invention relates to the field of anti-fraud prevention and control technology, and in particular to the field of risk control by using a risk control center to decide on user willingness risks, conduct risk warnings and risk disposal for user transactions, and more particularly to a risk prevention and control method, device and storage medium based on a willingness classification mining model. Background Art

[0002] Telecom fraud cases involve huge amounts of money and a wide range of people. The means of fraud are also becoming more and more diverse, from the initial telephone fraud to the current Internet fraud, SMS fraud and other forms. Fraudsters often use false identities and make up various reasons to trick victims into transferring money, remitting money or providing personal information, so as to achieve the purpose of illegal profit.

[0003] In the existing technology, the risk of fraud is generally prevented through the risk control center, which mainly adopts the method of expert rules. For the scenario of "cardholders encountering telecommunications fraud", most of them are induced frauds, and the traditional risk control technology rules are not well implemented by users themselves. It is necessary to collect multi-dimensional data, apply big data and intelligent AI technology to accurately identify customer intentions, conduct in-depth analysis and decision-making on customer intentions, identify customer risks, and dissuade and awaken customers. However, the technology used by the current risk control business center cannot meet the above requirements. Summary of the invention

[0004] To this end, the present invention proposes to use the intention classification mining model to realize in-depth mining of transaction user intentions in fraud scenarios, and obtain the user's real transaction intentions through multiple rounds of human-computer dialogues, voice-text conversion, and classification speech label recognition.

[0005] The problem that the risk control center cannot obtain the user's true intention to operate the transaction in the transaction scenario has been solved. By solving this problem, the risk control center will use customer intention as an important factor in risk decision-making in risk warning, thereby improving the accuracy of risk decision-making.

[0006] To achieve the above objectives, in a first aspect, the present invention provides a risk prevention and control method based on a willingness classification mining model, comprising:

[0007] Obtaining the user's voice call, identifying and extracting key words based on the voiceprint information of the voice call;

[0008] Establishing an intention recognition model, sending the extracted key words to the intention recognition model, and the intention recognition model classifying the risk level according to the extracted key words through expert rules;

[0009] According to the extracted key words, a next question is raised by setting rules and transmitted to the user end, and the above process is repeated;

[0010] All risk level classifications are summarized to form risk level labels which are transmitted to the anti-fraud risk control system, and the anti-fraud risk control system gives risk results based on the risk level labels.

[0011] Furthermore, the set rules use expert rules to perform risk screening in the Drools engine, directly control the results of high risk, and directly release the results of low risk, and configure several categories of rules, Rule 1, Rule 2 to Rule 10, and the levels of the rules are arranged from high risk to low risk. In transaction screening, when the transaction hits Rule 1 to Rule 3, the risk level is determined to be high risk; in transaction screening, when the transaction hits Rule 4 to Rule 8, the risk level is determined to be medium risk; in transaction screening, when the transaction hits Rule 9 to Rule 10, the risk level is determined to be low risk; the category and risk level of the rules are determined by analyzing the operational behavior of the elements involved.

[0012] Furthermore, when the risk level is medium risk, it includes:

[0013] If the risk level is determined to be medium risk, when it is detected that the user is not the user himself, the operation scenario is authenticated; when it is detected that the user is the user himself, the user's risk willingness is identified.

[0014] Furthermore, when it is detected that the user is himself, the user's risk willingness is identified, including: reaching the user through voice or outbound calling scenarios within the app, establishing an outbound calling mechanism with the user, and obtaining the user's transaction willingness through willingness classification mining models, TTS, ASR and other technologies, and making the final risk decision based on the collection of transaction intentions.

[0015] In a second aspect, the present invention also provides a risk prevention and control device based on a willingness classification mining model, the device comprising: a data extraction unit, a data screening unit, a calculation unit, and an output unit;

[0016] The data extraction unit is used to obtain a voice call of the user, identify key words according to the voiceprint information of the voice call, and extract the key words;

[0017] The data screening unit: establishes an intention recognition model, sends the extracted key words to the intention recognition model, and the intention recognition model classifies the risk level according to the extracted key words through expert rules;

[0018] The operation unit: according to the extracted key words, the next question is raised by setting rules and transmitted to the user end, and the above process is repeated;

[0019] The output unit summarizes all risk level classifications to form risk level labels and transmits them to the anti-fraud risk control system. The anti-fraud risk control system gives risk results based on the risk level labels.

[0020] Furthermore, the set rules use expert rules to perform risk screening in the Drools engine, directly control the results of high risk, and directly release the results of low risk, and configure several categories of rules, Rule 1, Rule 2 to Rule 10, and the levels of the rules are arranged from high risk to low risk. In transaction screening, when the transaction hits Rule 1 to Rule 3, the risk level is determined to be high risk; in transaction screening, when the transaction hits Rule 4 to Rule 8, the risk level is determined to be medium risk; in transaction screening, when the transaction hits Rule 9 to Rule 10, the risk level is determined to be low risk; the category and risk level of the rules are determined by analyzing the operational behavior of the elements involved.

[0021] Furthermore, when the risk level is medium risk, it includes:

[0022] If the risk level is determined to be medium risk, when it is detected that the user is not the user himself, the operation scenario is authenticated; when it is detected that the user is the user himself, the user's risk willingness is identified.

[0023] Furthermore, when it is detected that the user is himself, the user's risk willingness is identified, including: reaching the user through voice or outbound calling scenarios within the app, establishing an outbound calling mechanism with the user, and obtaining the user's transaction willingness through willingness classification mining models, TTS, ASR and other technologies, and making the final risk decision based on the collection of transaction intentions.

[0024] In a third aspect, the present invention provides a computing device, comprising:

[0025] A memory for storing program instructions;

[0026] The processor is used to call the program instructions stored in the memory and execute the risk prevention and control method based on the willingness classification mining model according to the obtained program instructions.

[0027] In a fourth aspect, the present invention provides a computer-readable storage medium, including computer-readable instructions. When a computer reads and executes the computer-readable instructions, the above-mentioned risk prevention and control method based on the willingness classification mining model is implemented. The readable medium may be multiple, and the multiple readable media can operate independently of each other.

[0028] Compared with the existing technology, this solution has realized a "self-supervision" method through the innovative classification model based on intention recognition and the use of BERT's pre-training mechanism. It uses a small amount of annotated data in the early stage to train the BERT model to achieve 80% accurate intention recognition. After going online, it uses the user's real intention to continuously accumulate intention data for annotation, and repeatedly trains the model. After multiple iterations, the intention classification model has achieved an accurate recognition rate of 90%. It completely replaces the manual outbound call scenario in the anti-fraud scenario of the risk control center, liberates customer service personnel, and greatly improves the accuracy of risk identification. The application of this invention in the anti-fraud scenario supplements the risk identification capability for the "honest fraud scenario", and plays an important role in supplementing the capabilities of the entire risk control center. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] In order to more clearly illustrate the embodiments of the present disclosure or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present disclosure. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0030] Figure 1 A flow chart of an embodiment of a risk prevention and control method based on a willingness classification mining model provided by the present invention;

[0031] Figure 2 A structural block diagram of an embodiment of a risk prevention and control device based on a willingness classification mining model provided by the present invention;

[0032] Figure 3 The present invention provides a schematic structural diagram of an electronic device. DETAILED DESCRIPTION

[0033] The embodiments of the present application are described in detail below, and examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present application, and cannot be interpreted as limiting the present application.

[0034] It will be understood by those skilled in the art that, unless otherwise stated, the singular forms "a", "an" and "the" used herein may also include plural forms. It should be further understood that the term "comprising" used in the specification of the present application refers to the presence of features, integers, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or groups thereof. The term "and / or" used herein includes all or any unit and all combinations of one or more associated listed items.

[0035] For the convenience of description, some nouns or terms involved in the embodiments of the present application are explained below:

[0036] TTS text-to-speech: A text-to-speech platform that obtains text information, recognizes text information and converts it into corresponding audio information.

[0037] ASR audio to text: An audio to file platform that obtains audio files and converts them into corresponding text information.

[0038] Open source BERT model: BERT (Bidirectional Encoder Representations from Transformers) is a pre-trained model proposed by Google AI Research in 2018. This model has excellent machine reading comprehension capabilities. It abandons traditional RNN and CNN, optimizes the word distance conversion to 1 through the Attention mechanism, effectively solves the thorny long-term dependency problem in NLP, and is widely used.

[0039] Drools rule open source engine: It is a rule engine based on Java pre-research and development provided by the JBOSS organization. It can free complex and changeable business rules from hard coding and store them in files or specific storage media (such as databases) in the form of rule scripts, so that changes to business rules can take effect immediately in the production environment without modifying the project code or restarting the server.

[0040] In order to make the objectives, technical solutions and advantages of the present application more clear, the specific implementation methods of the present application will be further described in detail below with reference to the accompanying drawings.

[0041] The core of the present invention is to provide a risk prevention and control method based on the willingness classification mining model. Figure 1 The flowchart of the risk prevention and control method based on the willingness classification mining model provided in this embodiment is shown. In order to better understand the method, Figure 1 The technical solution of this embodiment is described in detail:

[0042] Step S1: obtaining a user's voice call, identifying and extracting key words based on the voiceprint information of the voice call;

[0043] Step S2: Establish an intention recognition model, send the extracted key words to the intention recognition model, and the intention recognition model classifies the risk level according to the extracted key words through expert rules;

[0044] Step S3: according to the extracted key words, the next question is raised by setting rules and transmitted to the user end, and the above process is repeated;

[0045] Step S4: Summarize all risk level classifications to form risk level labels and transmit them to the anti-fraud risk control system. The anti-fraud risk control system gives risk results based on the risk level labels.

[0046] In a preferred embodiment, the set rules use expert rules to perform risk screening in the Drools engine, directly control the results of high risk, and directly release the results of low risk, and configure several categories of rules, Rule 1, Rule 2 to Rule 10, and the levels of the rules are arranged from high risk to low risk. In transaction screening, when the transaction hits Rule 1 to Rule 3, the risk level is determined to be high risk; in transaction screening, when the transaction hits Rule 4 to Rule 8, the risk level is determined to be medium risk; in transaction screening, when the transaction hits Rule 9 to Rule 10, the risk level is determined to be low risk; the category and risk level of the rules are determined by analyzing the operational behavior of the elements involved.

[0047] In a preferred embodiment, when the risk level is medium risk, it includes: if the risk level is determined to be medium risk, when it is detected that the user is not the user himself, the operation scenario is authenticated; when it is detected that the user is the user himself, the user's risk willingness is identified.

[0048] In a preferred embodiment, when it is detected that the user is himself, the user's risk willingness is identified, including: reaching the user through voice within the app or an outbound call scenario, establishing an outbound call mechanism with the user, and obtaining the user's transaction willingness through willingness classification mining models, TTS, ASR and other technologies, and making the final risk decision based on the collection of transaction intentions.

[0049] In the scheme of this embodiment, through the innovative classification model based on intention recognition and the use of BERT's pre-training mechanism, a "self-supervision" method is implemented. The BERT model is trained using a small amount of annotated data in the early stage to achieve 80% accurate intention recognition. After going online, the intention data is continuously accumulated using the user's real intention for annotation, and the model is repeatedly trained. After multiple iterations, the accuracy of the intention classification model reaches 90%. In the anti-fraud scenario of the risk control center, it completely replaces the manual outbound call scenario, liberates customer service personnel, and greatly improves the accuracy of risk identification. The application of this invention in the anti-fraud scenario supplements the risk identification capability for the "honest fraud scenario", and plays an important role in supplementing the capabilities of the entire risk control center.

[0050] In one embodiment, the present invention provides a risk prevention and control device based on a willingness classification mining model. Figure 2The figure shows a structural block diagram of a risk prevention and control device based on a willingness classification mining model provided by this embodiment, wherein the device comprises: a data extraction unit 10, a data screening unit 20, a calculation unit 30, and an output unit 40;

[0051] The data extraction unit 10 is used to obtain the user's voice call, identify key words according to the voiceprint information of the voice call, and extract them;

[0052] The data screening unit 20: establishes an intention recognition model, sends the extracted key words to the intention recognition model, and the intention recognition model classifies the risk level according to the extracted key words through expert rules;

[0053] The operation unit 30: proposes the next question according to the extracted key words by setting rules and transmits it to the user end, and repeats the above process;

[0054] The output unit 40 summarizes all risk level classifications to form risk level labels and transmits them to the anti-fraud risk control system. The anti-fraud risk control system gives risk results based on the risk level labels.

[0055] In a preferred implementation of the present embodiment, the set rules use expert rules to perform risk screening in the Drools engine, directly control the results of high risk, and directly release the results of low risk, and configure several categories of rules, Rule 1, Rule 2 to Rule 10, and the levels of the rules are arranged from high risk to low risk. In transaction screening, when the transaction hits Rule 1 to Rule 3, the risk level is determined to be high risk; in transaction screening, when the transaction hits Rule 4 to Rule 8, the risk level is determined to be medium risk; in transaction screening, when the transaction hits Rule 9 to Rule 10, the risk level is determined to be low risk; the category and risk level of the rules are determined by analyzing the operational behavior of the elements involved.

[0056] In a preferred implementation of this embodiment, when the risk level is medium risk, it includes: if the risk level is determined to be medium risk, when it is detected that the user is not the user himself, the operation scenario is authenticated; when it is detected that the user is the user himself, the user's risk willingness is identified.

[0057] In a preferred implementation of the present embodiment, when it is detected that the user is himself, the user risk willingness identification is performed, including: reaching the user by voice within the app or by outbound calling, establishing an outbound calling mechanism with the user, obtaining the user's transaction willingness by using willingness classification mining models, TTS, ASR and other technologies, and making a final risk decision based on the collection of transaction willingness.

[0058] In the scheme of this embodiment, through the innovative classification model based on intention recognition and the use of BERT's pre-training mechanism, a "self-supervision" method is implemented. The BERT model is trained using a small amount of annotated data in the early stage to achieve 80% accurate intention recognition. After going online, the intention data is continuously accumulated using the user's real intention for annotation, and the model is repeatedly trained. After multiple iterations, the accuracy of the intention classification model reaches 90%. In the anti-fraud scenario of the risk control center, it completely replaces the manual outbound call scenario, liberates customer service personnel, and greatly improves the accuracy of risk identification. The application of this invention in the anti-fraud scenario supplements the risk identification capability for the "honest fraud scenario", and plays an important role in supplementing the capabilities of the entire risk control center.

[0059] The risk control device based on the willingness classification mining model of this embodiment is used to implement the aforementioned risk control method based on the willingness classification mining model. Therefore, the specific implementation method of the risk control device based on the willingness classification mining model can be seen in the specific implementation method of the risk control method based on the willingness classification mining model in the previous text. For example, the data extraction unit 10 is used to implement step S1 in the above-mentioned risk control method based on the willingness classification mining model, and the data screening unit 20 is used to implement step S2 in the above-mentioned risk control method based on the willingness classification mining model, etc. Therefore, its specific implementation method can refer to the description of the corresponding embodiments of each part. The device provided by the present invention is attached to the method provided by the present invention, and its specific effects can refer to the above-mentioned method, which will not be repeated here.

[0060] In one embodiment, a computing device is provided, comprising:

[0061] A memory for storing program instructions;

[0062] A processor is used to implement the steps of any of the above-mentioned risk prevention and control methods based on the willingness classification mining model when executing the computer program.

[0063] As an example, Figure 3 A schematic diagram of the structure of an electronic device applicable to the embodiment of the present application is shown. Figure 3 As shown, the electronic device 2000 includes: a processor 2001 and a memory 2003. The processor 2001 and the memory 2003 are connected, such as through a bus 2002. Optionally, the electronic device 2000 may also include a transceiver 2004. It should be noted that in actual applications, the transceiver 2004 is not limited to one, and the structure of the electronic device 2000 does not constitute a limitation on the embodiments of the present application.

[0064] The processor 2001 is applied in the embodiment of the present application to implement the method shown in the above method embodiment. The transceiver 2004 may include a receiver and a transmitter. The transceiver 2004 is applied in the embodiment of the present application to implement the function of the electronic device of the embodiment of the present application communicating with other devices when executed.

[0065] Processor 2001 may be a CPU (Central Processing Unit), a general purpose processor, a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array) or other programmable logic devices, transistor logic devices, hardware components or any combination thereof. It may implement or execute various exemplary logic blocks, modules and circuits described in conjunction with the disclosure of this application. Processor 2001 may also be a combination that implements computing functions, such as a combination of one or more microprocessors, a combination of a DSP and a microprocessor, etc.

[0066] The bus 2002 may include a path to transmit information between the above components. The bus 2002 may be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus. The bus 2002 may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 3 Only one thick line is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0067] The memory 2003 can be a ROM (Read Only Memory) or other types of static storage devices that can store static information and instructions, a RAM (Random Access Memory) or other types of dynamic storage devices that can store information and instructions, or an EEPROM (Electrically Erasable Programmable Read Only Memory), a CD-ROM (Compact Disc Read Only Memory) or other optical disk storage, optical disk storage (including compressed optical disk, laser disk, optical disk, digital versatile disk, Blu-ray disk, etc.), a magnetic disk storage medium or other magnetic storage device, or any other medium that can be used to carry or store the desired program code in the form of instructions or data structures and can be accessed by a computer, but is not limited to these.

[0068] Optionally, the memory 2003 is used to store application code for executing the solution of the present application, and the execution is controlled by the processor 2001. The processor 2001 is used to execute the application code stored in the memory 2003 to implement the investment and financing project recommendation method provided in any embodiment of the present application.

[0069] The electronic device provided in the embodiment of the present application is applicable to any embodiment of the above method and will not be described in detail here.

[0070] The risk prevention and control method based on the intention classification mining model provided by the present invention has obvious advantages and beneficial effects compared with the existing technology. Through the innovative intention identification classification model, the pre-training mechanism of BERT is used to realize a "self-supervision" method. The BERT model is trained with a small amount of annotated data in the early stage to achieve 80% accurate intention identification. After going online, the intention data is continuously accumulated using the user's real intention for annotation, and the model is repeatedly trained. After multiple iterations, the intention classification model has an accurate recognition rate of 90%. It completely replaces the manual outbound call scenario in the anti-fraud scenario of the risk control center, liberates customer service personnel, and greatly improves the accuracy of risk identification. The application of this invention in the anti-fraud scenario supplements the risk identification capability for the "duty fraud scenario", which plays an important role in the capacity supplement of the entire risk control center.

[0071] In one embodiment, a computer-readable storage medium is provided, including computer-readable instructions. When a computer reads and executes the computer-readable instructions, the steps of the risk prevention and control method based on the willingness classification mining model described above are implemented. The risk prevention and control method based on the willingness classification mining model provided by the present invention has obvious advantages and beneficial effects compared with the prior art. Through the innovative willingness identification classification model, the BERT pre-training mechanism is used to implement a "self-supervision" method. The BERT model is trained using a small amount of annotated data in the early stage to achieve 80% accurate willingness identification. After the online production, the willingness data is continuously accumulated using the user's real willingness for annotation, and the model is repeatedly trained. After multiple iterations, the willingness classification model reaches 90% accurate identification. In the anti-fraud scenario of the risk control middle platform, the manual outbound call scenario is completely replaced, customer service personnel are liberated, and the risk identification accuracy is greatly improved. The application of this invention in the anti-fraud scenario supplements the risk identification capability for the "duty fraud scenario", which plays an important role in the capacity supplement of the entire risk control middle platform.

[0072] Through the description of the above implementation methods, technical personnel in the relevant field can understand that for the convenience and simplicity of description, only the division of the above-mentioned functional modules is used as an example. In actual applications, the above-mentioned functions can be assigned to different functional modules as needed, that is, the internal structure of the specific device can be divided into different functional modules to complete all or part of the functions described above.

[0073] Those of ordinary skill in the art will appreciate that the modules and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in terms of function in the above description. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0074] In several embodiments covered by the present application, it should be understood that the disclosed devices, equipment and methods can be implemented in other ways. For example, the division of modules is only a logical function division, and there may be other division methods in actual implementation, such as multiple modules or components can be combined or integrated into another device, or some features can be ignored or not executed. In addition, the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or modules, or it can be an electrical, mechanical or other form of connection.

[0075] The modules described as separate components may or may not be physically separated, and the components displayed as modules may or may not be physical modules, that is, they may be located in one place or distributed on multiple network modules. Some or all of the modules may be selected according to actual needs to achieve the purpose of the embodiments of the present application.

[0076] In addition, each functional module in each embodiment of the present application can be implemented in the form of hardware or in the form of software functional modules. If these functional modules are implemented in the form of software functional modules and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present application is essentially or part of the prior art that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a computer program product, which includes one or more computer instructions. When loading and executing computer program instructions on a computer, the process or function according to the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer program product is stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium, for example, computer instructions can be transmitted from a website site, computer, server or data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) mode to another website site, computer, server or data center. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more available media integrations. The available media may be magnetic media (eg, floppy disks, hard disks, magnetic tapes), optical media (eg, DVDs), or semiconductor media (eg, solid state disks (SSDs)).

[0077] The above contents are merely examples and explanations of the structure of the present invention. The technicians in this technical field may make various modifications or additions to the specific embodiments described or replace them in a similar manner. As long as they do not deviate from the structure of the invention or exceed the scope defined by the claims, they should all fall within the protection scope of the present invention.

Claims

1. A risk prevention and control method based on a willingness classification mining model, characterized in that: include: Obtaining the user's voice call, identifying and extracting key words based on the voiceprint information of the voice call; Establishing an intention recognition model, sending the extracted key words to the intention recognition model, and the intention recognition model classifying the risk level according to the extracted key words through expert rules; According to the extracted key words, a next question is raised by setting rules and transmitted to the user end, and the above process is repeated; All risk level classifications are summarized to form risk level labels which are transmitted to the anti-fraud risk control system, and the anti-fraud risk control system gives risk results based on the risk level labels.

2. The risk prevention and control method based on the intention classification mining model according to claim 1 is characterized in that: The setting rules use expert rules to perform risk screening in the Drools engine, directly control the results of high risk, and directly release the results of low risk. Configure several categories of rules, including rule 1, rule 2 to rule 10. The levels of the rules are arranged from high risk to low risk. In transaction screening, when the transaction hits rule 1 to rule 3, the risk level is determined to be high risk; In transaction screening, when the transaction hits rules 4 to 8, the risk level is determined to be medium risk; In transaction screening, when the transaction hits Rule 9 to Rule 10, the risk level is determined to be low risk; the category and risk level of the rule are determined by analyzing the operational behavior of the elements involved.

3. The risk prevention and control method based on the intention classification mining model according to claim 2 is characterized in that: When the risk level is medium risk, it includes: If the risk level is determined to be medium risk, when it is detected that the user is not the user himself, the operation scenario is authenticated; when it is detected that the user is the user himself, the user's risk willingness is identified.

4. The risk prevention and control method based on the intention classification mining model according to claim 3 is characterized in that: When it is detected that the user is himself, the user's risk willingness is identified, including: using the voice within the app or the outbound call scenario to reach the user, establishing an outbound call mechanism with the user, using the willingness classification mining model, TTS, ASR and other technologies to obtain the user's transaction willingness, and making the final risk decision based on the transaction willingness collection.

5. A risk prevention and control device based on a willingness classification mining model, characterized in that: The device comprises: a data extraction unit, a data screening unit, a calculation unit, and an output unit; The data extraction unit is used to obtain a voice call of the user, identify key words according to the voiceprint information of the voice call, and extract the key words; The data screening unit: establishes an intention recognition model, sends the extracted key words to the intention recognition model, and the intention recognition model classifies the risk level according to the extracted key words through expert rules; The operation unit: according to the extracted key words, the next question is raised by setting rules and transmitted to the user end, and the above process is repeated; The output unit summarizes all risk level classifications to form risk level labels and transmits them to the anti-fraud risk control system. The anti-fraud risk control system gives risk results based on the risk level labels.

6. The risk prevention and control method based on the intention classification mining model according to claim 5 is characterized in that: The setting rules use expert rules to perform risk screening in the Drools engine, directly control the results of high risk, and directly release the results of low risk. Configure several categories of rules, including rule 1, rule 2 to rule 10. The levels of the rules are arranged from high risk to low risk. In transaction screening, when the transaction hits rule 1 to rule 3, the risk level is determined to be high risk; In transaction screening, when the transaction hits rules 4 to 8, the risk level is determined to be medium risk; In transaction screening, when the transaction hits Rule 9 to Rule 10, the risk level is determined to be low risk; the category and risk level of the rule are determined by analyzing the operational behavior of the elements involved.

7. The risk prevention and control method based on the intention classification mining model according to claim 6 is characterized in that: When the risk level is medium risk, it includes: If the risk level is determined to be medium risk, when it is detected that the user is not the user himself, the operation scenario is authenticated; when it is detected that the user is the user himself, the user's risk willingness is identified.

8. The risk prevention and control method based on the intention classification mining model according to claim 7 is characterized in that: When it is detected that the user is himself, the user's risk willingness is identified, including: using the voice within the app or the outbound call scenario to reach the user, establishing an outbound call mechanism with the user, using the willingness classification mining model, TTS, ASR and other technologies to obtain the user's transaction willingness, and making the final risk decision based on the transaction willingness collection.

9. A computing device, characterized in that include: A memory for storing program instructions; A processor is used to call the program instructions stored in the memory, and execute the method according to any one of claims 1 to 4 according to the obtained program instructions.

10. A computer-readable storage medium, characterized in that: It includes computer-readable instructions. When a computer reads and executes the computer-readable instructions, it implements the method as claimed in any one of claims 1 to 4. There can be multiple readable media, and the multiple readable media can operate independently of each other.

Images