Multi-party security anti-fraud method, device and storage medium based on blockchain

Abstract

The present application discloses a multi-party security anti-fraud method, device and storage medium based on blockchain. The method includes: encrypting the claim information with a public key, generating first encrypted information, and uploading the first encrypted information to the blockchain system; obtaining the first encrypted information from the blockchain system, determining the second encrypted information based on the first encrypted information, and uploading the second encrypted information to the blockchain system; obtaining the second encrypted information from the blockchain system, and decrypting the second encrypted information with a private key to determine the claim amount; calling a pre-embedded risk prediction model, and using the risk prediction model and the first transaction data information corresponding to the user terminal to predict the fraud risk probability of the current claim transaction, wherein the first transaction data information includes the claim amount information; and when the fraud risk probability corresponding to the current claim transaction is greater than a preset threshold, determining that the fraud risk of the current claim transaction is high, and alerting the insurance company system.

Description

Technical Field

[0001] The present application relates to the field of blockchain transaction technology, and in particular to a blockchain-based multi-party security anti-fraud method, device and storage medium. Background Art

[0002] At present, although the development of blockchain technology has been relatively mature, the application of blockchain technology in the insurance industry is still in the exploratory stage. And although some insurance companies have cooperated with other companies to try to use blockchain technology for policy management and claims processing, most of them are still in the proof-of-concept stage, and there are few mature applications.

[0003] And although a small number of blockchain systems used in insurance companies can realize policy management and claims processing, there are still many shortcomings in actual application.

[0004] For example, since existing policyholders attach great importance to personal privacy, they have high requirements for the privacy of transaction data information. However, transaction data information in existing blockchain systems will be shared by multiple parties and lacks an effective privacy protection mechanism.

[0005] For another example, since most of the transactions of insurance companies are related to money, insurance companies have high requirements for the security of transactions. However, since the existing blockchain systems applied to insurance companies do not have models related to predicting fraud risks, the security of insurance companies' transactions cannot be effectively guaranteed.

[0006] Currently, no effective solution has been proposed for the technical problems that transaction data information in the blockchain system in the above-mentioned prior art is shared by multiple parties, lacks an effective privacy protection mechanism, and cannot guarantee the security of transactions. Summary of the invention

[0007] The embodiments of the present disclosure provide a multi-party security anti-fraud method, device and storage medium based on blockchain, so as to at least solve the technical problems existing in the prior art that transaction data information in the blockchain system is shared by multiple parties, lacks an effective privacy protection mechanism, and cannot guarantee the security of transactions.

[0008] According to one aspect of an embodiment of the present disclosure, a multi-party security anti-fraud method based on blockchain is provided, including: a user terminal encrypts claim information using a public key to generate first encrypted information, and uploads the first encrypted information to a blockchain system; an insurance company system obtains the first encrypted information from the blockchain system, determines second encrypted information based on the first encrypted information, and uploads the second encrypted information to the blockchain system; the user terminal obtains the second encrypted information from the blockchain system, and decrypts the second encrypted information using a private key to determine a claim amount; a smart contract calls a pre-embedded risk prediction model, and uses the risk prediction model and the acquired first transaction data information corresponding to the user terminal to predict the fraud risk probability of the current claim transaction, wherein the first transaction data information includes claim amount information; and when the fraud risk probability corresponding to the current claim transaction is greater than a preset threshold, the smart contract determines that the fraud risk of the current claim transaction is high, and alerts the insurance company system.

[0009] According to another aspect of an embodiment of the present disclosure, a storage medium is further provided, the storage medium including a stored program, wherein when the program is running, a processor executes any one of the methods described above.

[0010] According to another aspect of the embodiment of the present disclosure, a multi-party security anti-fraud device based on blockchain is also provided, including: a first encrypted information generation module, used for the user terminal to encrypt the claim information using a public key, generate first encrypted information, and upload the first encrypted information to the blockchain system; a second encrypted information generation module, used for the insurance company system to obtain the first encrypted information from the blockchain system, determine the second encrypted information based on the first encrypted information, and upload the second encrypted information to the blockchain system; a decryption module, used for the user terminal to obtain the second encrypted information from the blockchain system, and use the private key to decrypt the second encrypted information to determine the claim amount; a risk probability prediction module, used for the smart contract to call a pre-embedded risk prediction model, and use the risk prediction model and the acquired first transaction data information corresponding to the user terminal to predict the fraud risk probability of the current claim transaction, wherein the first transaction data information includes the claim amount information; and an alarm module, used for when the fraud risk probability corresponding to the current claim transaction is greater than a preset threshold, the smart contract determines that the fraud risk of the current claim transaction is high, and alarms the insurance company system.

[0011] According to another aspect of the embodiment of the present disclosure, a multi-party security anti-fraud device based on blockchain is also provided, including: a processor; and a memory, connected to the processor, for providing the processor with instructions for processing the following processing steps: the user terminal uses a public key to encrypt the claim information to generate first encrypted information, and uploads the first encrypted information to the blockchain system; the insurance company system obtains the first encrypted information from the blockchain system, determines the second encrypted information based on the first encrypted information, and uploads the second encrypted information to the blockchain system; the user terminal obtains the second encrypted information from the blockchain system, and decrypts the second encrypted information using a private key to determine the claim amount; the smart contract calls a pre-embedded risk prediction model, and uses the risk prediction model and the first transaction data information corresponding to the user terminal to predict the fraud risk probability of the current claim transaction, wherein the first transaction data information includes the claim amount information; and when the risk probability corresponding to the current claim transaction is greater than a preset threshold, the smart contract determines that the fraud risk of the current claim transaction is high, and alerts the insurance company system.

[0012] The present application provides a multi-party security anti-fraud method based on blockchain. First, the user terminal uses a public key to encrypt the claim information, generates a first encrypted information, and sends the first encrypted information to the blockchain system. Then, the insurance company system obtains the first encrypted information from the blockchain system, determines the second encrypted information based on the first encrypted information, and uploads the second encrypted information to the blockchain system. Furthermore, the user terminal obtains the second encrypted information from the blockchain system, and uses the private key to decrypt the second encrypted information to determine the claim amount. After that, the smart contract schedules a pre-embedded risk prediction model, and uses the risk prediction model and the first transaction data information corresponding to the user terminal to predict the fraud risk probability of the current claim transaction. Finally, when the risk probability corresponding to the current claim transaction is greater than the preset threshold, it is determined that the fraud risk of the current claim transaction is high, and an alarm is sent to the insurance company system.

[0013] Since the user terminal in this application uses the public key to encrypt the claim information to generate the first encrypted information, and the insurance company system directly calculates based on the first encrypted information to generate the second encrypted information corresponding to the claim amount, the insurance company system will not know the specific claim information and claim amount corresponding to the user terminal, whether in the process of information transmission or in the process of the insurance company system calculating the second encrypted information. Therefore, the technical solution of this application can effectively protect the privacy of users and improve data processing efficiency.

[0014] Furthermore, since the risk prediction model is pre-embedded in the smart contract in the technical solution disclosed in the present application, during the process of the above-mentioned insurance company system and the user terminal conducting a claim transaction, the smart contract can determine whether the probability of the current claim transaction having a fraud risk is greater than a preset threshold based on the first transaction data information corresponding to the user terminal. And if it is greater than the preset threshold, the insurance company system is promptly alerted, thereby further ensuring the security of the insurance company in the claim transaction.

[0015] This solves the technical problems existing in the prior art that transaction data information in the blockchain system is shared by multiple parties, lacks an effective privacy protection mechanism, and cannot guarantee the security of transactions. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The drawings described herein are used to provide a further understanding of the present disclosure and constitute a part of the present application. The illustrative embodiments of the present disclosure and their descriptions are used to explain the present disclosure and do not constitute an improper limitation on the present disclosure. In the drawings:

[0017] Figure 1 is a hardware structure block diagram of a computing device for implementing the method according to Embodiment 1 of the present application;

[0018] Figure 2 This is a schematic diagram of a multi-party security anti-fraud system based on blockchain according to Example 1 of the present application;

[0019] Figure 3 This is a flowchart of the multi-party security anti-fraud method based on blockchain according to Example 1 of the present application;

[0020] Figure 4 It is a flowchart of predicting the risk probability of the current claim transaction according to the fully homomorphic encryption described in Example 1 of the present application and using the risk prediction model;

[0021] Figure 5 is a schematic diagram of a multi-party security anti-fraud device based on blockchain according to Example 2 of the present application; and

[0022] Figure 6 This is a schematic diagram of a multi-party security anti-fraud device based on blockchain as described in Example 3 of the present application. DETAILED DESCRIPTION

[0023] In order to enable those skilled in the art to better understand the technical solutions of the present disclosure, 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 embodiments of a part of the present disclosure, not all of the embodiments. Based on the embodiments in the present disclosure, all other embodiments obtained by ordinary technicians in this field without making creative work should fall within the scope of protection of the present disclosure.

[0024] It should be noted that the terms "first", "second", etc. in the specification and claims of the present disclosure and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present disclosure described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product, or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products, or devices.

[0025] Example 1

[0026] According to this embodiment, a method embodiment of a multi-party secure anti-fraud based on blockchain is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0027] The method embodiment provided in this embodiment can be executed in a mobile terminal, a computer terminal, a server or a similar computing device. Figure 1 The figure shows a hardware structure block diagram of a computing device for implementing multi-party security anti-fraud based on blockchain. Figure 1 As shown, the computing device may include one or more processors (the processor may include but is not limited to a processing device such as a microprocessor MCU or a programmable logic device FPGA), a memory for storing data, a transmission device for communication functions, and an input / output interface. The memory, the transmission device, and the input / output interface are connected to the processor via a bus. In addition, it may also include: a display, a keyboard, and a cursor control device connected to the input / output interface. A person skilled in the art can understand that Figure 1 The structure shown is only for illustration and does not limit the structure of the above electronic device. Figure 1More or fewer components as shown, or with Figure 1 Different configurations are shown.

[0028] It should be noted that the one or more processors and / or other data processing circuits described above may generally be referred to herein as "data processing circuitry". The data processing circuitry may be embodied in whole or in part as software, hardware, firmware, or any other combination thereof. In addition, the data processing circuitry may be a single independent processing module, or may be incorporated in whole or in part into any of the other components of the computing device. As involved in the embodiments of the present disclosure, the data processing circuitry acts as a processor control (e.g., selection of a variable resistor terminal path connected to an interface).

[0029] The memory can be used to store software programs and modules of application software, such as the program instructions / data storage device corresponding to the multi-party security anti-fraud method based on blockchain in the embodiment of the present disclosure. The processor executes various functional applications and data processing by running the software programs and modules stored in the memory, that is, the multi-party security anti-fraud method based on blockchain of the above-mentioned application program is realized. The memory may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory may further include a memory remotely arranged relative to the processor, and these remote memories may be connected to the computing device via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0030] The transmission device is used to receive or send data via a network. The specific example of the above network may include a wireless network provided by a communication provider of the computing device. In one example, the transmission device includes a network adapter (Network Interface Controller, NIC), which can be connected to other network devices through a base station so as to communicate with the Internet. In one example, the transmission device can be a radio frequency (Radio Frequency, RF) module, which is used to communicate with the Internet wirelessly.

[0031] The display may be, for example, a touch screen liquid crystal display (LCD) that may enable a user to interact with a user interface of the computing device.

[0032] It should be noted that, in some optional embodiments, the above Figure 1 The computing device shown may include hardware elements (including circuits), software elements (including computer code stored on a computer-readable medium), or a combination of both hardware elements and software elements. Figure 1This is merely one example of a particular embodiment and is intended to illustrate the types of components that may be present in the computing devices described above.

[0033] Figure 2 Schematic diagram of a multi-party security anti-fraud system based on blockchain according to this embodiment. Figure 2 As shown, the system includes: a blockchain network layer, a smart contract layer, a data privacy protection layer, an application interface layer, and an application terminal layer.

[0034] Among them, the blockchain network layer includes the alliance chain network. The alliance chain network supports multiple channels and smart contracts, and the alliance chain network includes insurance companies, hospitals, repair shops and other related institutions and manufacturers that may generate claims transactions. The smart contract layer is mainly used to implement the core insurance business logic, such as policy management, claims processing and fraud risk detection. The data privacy protection layer is mainly used to protect the data privacy on the chain based on technologies such as zero-knowledge proof and fully homomorphic encryption, so as to realize the secure sharing and calculation of privacy data on the chain. The application interface layer is mainly used to provide an interface for third-party applications to interact with the blockchain. The application terminal layer provides an operation interface for users (for example, policyholders, claims adjusters or insurance company employees).

[0035] Under the above operating environment, according to the first aspect of this embodiment, a multi-party security anti-fraud method based on blockchain is provided. The method comprises: Figure 2 The system implementation shown in . Figure 3 A schematic diagram showing the process of the method is shown in FIG. Figure 3 As shown, the method includes:

[0036] S302: The user terminal encrypts the claim information using the public key to generate first encrypted information, and sends the first encrypted information to the blockchain system;

[0037] S304: The insurance company system obtains the first encrypted information sent by the blockchain system, determines the second encrypted information based on the first encrypted information, and sends the second encrypted information to the blockchain system;

[0038] S306: The user terminal obtains the second encrypted information sent by the blockchain system, and decrypts the second encrypted information using the private key to determine the claim amount;

[0039] S308: the smart contract calls a pre-embedded risk prediction model, and uses the risk prediction model and the acquired first transaction data information corresponding to the user terminal to predict the fraud risk probability of the current claim transaction, wherein the first transaction data information includes claim amount information; and

[0040] S310: When the fraud risk probability corresponding to the current claim transaction is greater than a preset threshold, it is determined that the fraud risk of the current claim transaction is high, and an alarm is sent to the insurance company system.

[0041] Specifically, Figure 4 This is a flowchart of predicting the risk probability of the current claim transaction based on the fully homomorphic encryption described in the embodiment of the present application and using the risk prediction model. Figure 4 As shown, first, the user terminal uses the public key PK to encrypt the claim information q to generate the first encrypted information Enc PK (q), and the first encrypted information Enc PK (q) is uploaded to the blockchain system (S302). The first encrypted information Enc PK (q) includes the first sub-encrypted information Enc corresponding to the claim expense q1 PK (q1), and the second sub-encrypted information Enc corresponding to the claim ratio q2 PK (q2).

[0042] Then, the insurance company system obtains the first encrypted information Enc from the blockchain system PK (q), based on the first encrypted information Enc PK (q), determine the second encrypted information Enc PK (p), and the second encrypted information Enc PK (p) Upload to the blockchain system (S304). The second encrypted information Enc PK (p) corresponds to the claim amount. Specifically, upon receiving the first sub-encrypted information Enc sent by the blockchain system, the insurance company system PK (q1) and the second sub-encrypted information Enc PK (q2) After that, the first sub-encryption information Enc PK (q1) and the second sub-encrypted information Enc PK (q2) Perform homomorphic multiplication operation to determine the second encrypted information Enc PK (p). The above content will be described in detail later, so it will not be repeated here.

[0043] Therefore, it can be seen from the above that the insurance company system directly calculates the second encrypted information based on the first encrypted information corresponding to the user terminal, and uploads the second encrypted information to the blockchain system. Therefore, during the entire process of the insurance company system calculating the second encrypted information corresponding to the claim amount, the claim amount and claim information corresponding to the user terminal will not be known. In other words, during the entire calculation process, the insurance company system cannot disclose the privacy information of the insured corresponding to the user terminal. In this way, the privacy of the insured can be effectively protected and the data processing efficiency can be further improved.

[0044] After that, the user terminal obtains the second encrypted information Enc from the blockchain system PK (p), and use the private key to encrypt the second information Enc PK (p) is decrypted to determine the claim amount p (S306).

[0045] Furthermore, the smart contract calls a pre-embedded risk prediction model, and uses the risk prediction model and the first transaction data information corresponding to the user terminal to predict the fraud risk probability of the current claim transaction (S308). Specifically, first, the smart contract obtains the first transaction data information corresponding to the user terminal, and extracts the risk prediction feature information corresponding to the current claim transaction from the first transaction data information. Then, the smart contract quantifies and normalizes the risk prediction feature information corresponding to the current claim transaction, thereby generating a risk prediction feature vector corresponding to the current claim transaction. Further, the smart contract calls the risk prediction model, and based on the risk prediction feature vector corresponding to the current claim transaction, predicts the risk probability of the current claim transaction. Thus, the risk prediction model can determine whether there is a fraud risk in the current claim transaction.

[0046] Finally, when the fraud risk probability corresponding to the current claim transaction is greater than a preset threshold, the fraud risk of the current claim transaction is determined to be high, and an alarm is sent to the insurance company system (S310). The preset threshold may be, for example, a safety threshold calculated by the insurance company system based on historical claim transaction data, or may be a safety threshold input into the insurance company system by insurance company personnel based on experience.

[0047] For example, the preset threshold is 0.9. Then the smart contract inputs the risk prediction feature vector Y corresponding to the current claim transaction extracted from the first transaction data information into the risk prediction model h θ (X) = f(θ T ·X), the fraud risk probability corresponding to the current claim transaction can be determined. Y =h θ (Y) = f(θ T·Y). And, in the risk probability P corresponding to the current claim transaction Y If it is greater than 0.9, the fraud risk of the current claim transaction is judged to be high, and an alarm is sent to the insurance company system; when the risk probability P corresponding to the current claim transaction Y When it is less than 0.9, it is judged that the fraud risk of the current claim transaction is low and the current claim transaction is relatively safe.

[0048] Therefore, it can be seen from the above that the insurance company system can predict the risk probability of the current claim transaction based on the first transaction data information corresponding to the user terminal and the risk prediction model pre-embedded in the smart contract during the current claim transaction. And when the risk probability of the current claim transaction is greater than the preset threshold, an alarm can be received in time, thereby greatly improving the security of the claim transaction.

[0049] As described in the background technology, although a small number of blockchain systems applied to insurance companies can achieve policy management and claims processing, there are still many deficiencies in practical applications. For example, since existing policyholders attach great importance to personal privacy, policyholders have high requirements for the privacy of transaction data information, but transaction data information in existing blockchain systems will be shared by multiple parties, lacking an effective privacy protection mechanism. For another example, since most of the transactions of insurance companies are related to money, insurance companies have high requirements for the security of transactions. However, since the existing blockchain systems applied to insurance companies do not have models related to predicting fraud risks, the security of insurance companies when conducting transactions cannot be effectively guaranteed.

[0050] In view of this, the present application provides a multi-party security anti-fraud method based on blockchain. And because in the present application, the user terminal uses the public key to encrypt the claim information to generate the first encrypted information, and the insurance company system directly calculates on the basis of the first encrypted information to generate the second encrypted information corresponding to the claim amount, therefore, whether in the process of information transmission or in the process of the insurance company system calculating the second encrypted information, the insurance company system will not know the specific claim information and claim amount corresponding to the user terminal. Therefore, the technical solution of the present application can effectively protect the privacy of users and improve data processing efficiency.

[0051] Furthermore, since the risk prediction model is pre-embedded in the smart contract in the technical solution disclosed in the present application, during the process of the above-mentioned insurance company system and the user terminal conducting a claim transaction, the smart contract can determine whether the probability of the current claim transaction having a fraud risk is greater than a preset threshold based on the first transaction data information corresponding to the user terminal. And if it is greater than the preset threshold, the insurance company system is promptly alerted, thereby further ensuring the security of the insurance company in the claim transaction.

[0052] This solves the technical problems existing in the prior art that transaction data information in the blockchain system is shared by multiple parties, lacks an effective privacy protection mechanism, and cannot guarantee the security of transactions.

[0053] Optionally, the user terminal encrypts the claim information using the public key to generate the first encrypted information, including: the user terminal encrypts the claim fee using the public key to generate the first sub-encrypted information, and the user terminal encrypts the claim ratio using the public key to generate the second sub-encrypted information. Further optionally, the insurance company system obtains the first encrypted information sent by the blockchain system, and based on the first encrypted information, determines the second encrypted information, including: the insurance company system obtains the first sub-encrypted information and the second sub-encrypted information sent by the blockchain system, and performs a homomorphic multiplication operation to determine the second encrypted information.

[0054] Specifically, first, the user terminal uses the public key to encrypt the claim fee q1 to generate the first sub-encrypted information Enc PK (q1), and the user terminal uses the public key to encrypt the claim ratio q2 to generate the second sub-encrypted information Enc PK (q2). Then, the user terminal sends the first sub-encrypted information Enc PK (q1) and the second sub-encrypted information Enc PK (q2) Upload to the blockchain system. Further, the insurance company system obtains the first sub-encrypted information Enc sent by the blockchain system. PK (q1) and the second sub-encrypted information Enc PK (q2), and perform homomorphic multiplication Enc PK (q1×q2), thereby determining the second encrypted information Enc PK (p). The second encrypted information Enc PK (p) = Enc PK (q1×q2), and p is the claim amount.

[0055] Thus, the insurance company system can determine the second encrypted information based on the first sub-encrypted information and the second sub-encrypted information by directly performing homomorphic multiplication operations on the first sub-encrypted information and the second sub-encrypted information. That is, the insurance company system can calculate the encrypted claim amount (i.e., the second encrypted information) without knowing the claim expense, claim ratio, and claim amount (i.e., the private information corresponding to the policyholder), thereby effectively protecting the privacy of the policyholder.

[0056] Optionally, the smart contract calls a pre-embedded risk prediction model, and uses the risk prediction model and the first transaction data information corresponding to the user terminal to predict the fraud risk probability of the current claim transaction, including: collecting the second transaction data information, and constructing a sample set based on the collected information, wherein the second transaction data information corresponds to the historical claim transaction; training a binary classification model using the sample set, and generating a risk prediction model; and obtaining the first transaction data information corresponding to the user terminal, and based on the obtained first transaction data information and the risk prediction model, predicting the risk probability of the current claim transaction. Further optionally, the operation of collecting the second transaction data information and constructing a sample set based on the collected information includes: collecting the second transaction data information corresponding to different historical claim transactions, and extracting multiple groups of risk prediction feature information corresponding to different historical claim transactions from the second transaction data information; quantifying and normalizing the multiple groups of risk prediction feature information to generate multiple risk prediction feature vectors; and constructing a sample set based on the generated multiple risk prediction feature vectors.

[0057] Specifically, first, the second transaction data information corresponding to different historical claims transactions is collected, and multiple groups of risk prediction feature information corresponding to different historical claims transactions are extracted from the second transaction data information. Among them, the risk prediction feature information may include, for example, policyholder attribute feature information, policy feature information, claim feature information, and policyholder behavior feature information. And among them, the policyholder attribute feature information may include, for example, the policyholder age and the policyholder gender. Policy feature information may include, for example, premium, insured amount, insurance type, and effective time. Claim feature information includes, for example, claim time, claim amount, and claim reason. Policyholder behavior feature information includes, for example, insurance frequency, claim frequency, and accident interval. For example, the risk prediction feature information includes n feature information corresponding to policyholder attribute feature information, policy feature information, claim feature information, and / or policyholder behavior feature information.

[0058] When extracting multiple sets of risk prediction feature information corresponding to different historical claims transactions, the smart contract quantifies and normalizes the multiple sets of risk prediction feature information, thereby generating multiple risk prediction feature vectors corresponding to different historical claims transactions. For example, risk prediction feature information corresponding to m sets of historical claims transactions is extracted, and after quantification and normalization, m risk prediction feature vectors are generated.

[0059] Thus, a sample set can be constructed based on the generated multiple risk prediction feature vectors. For example, the sample set includes the risk prediction feature vector (x 11 ,x 12 ......x 1n );The risk prediction feature vector (x 21 ,x 22 ......x 2n );......;and historical claims transactions x m The corresponding risk prediction feature vector (x m1 ,x m2 ......x mn ).

[0060] Then, the smart contract uses the constructed sample set to train a binary classification model to generate a risk prediction model. The binary classification model can be, for example, a logistic regression model, a random forest, or XGBoost, etc. In this application, a logistic regression model is used as an example.

[0061] For example, define a prediction function:

[0062] h θ (X) = f(θ T ·X) (Formula 1)

[0063] Among them, θ is the model parameter, θ T Represents the transpose of the model parameters.

[0064] Define the sigmoid activation function:

[0065]

[0066] Define the loss function:

[0067]

[0068] Where m is the number of samples, and i = 1 to m. j is the jth element in the model parameter θ. (i) is the true label of sample i. θ (X (i)) is the predicted value of sample i. λ is the regularization coefficient.

[0069] The smart contract then uses the samples in the sample set to train the above logistic regression model to obtain a risk prediction model.

[0070] Finally, the smart contract obtains the first transaction data information corresponding to the user terminal and calls the pre-trained risk prediction model, so that it can predict the fraud risk probability of the current claim transaction based on the first transaction data information.

[0071] Therefore, through the above method, it is possible to determine whether there is a fraud risk in the current claims transaction through the risk prediction model, thereby achieving the technical effect of ensuring the security of the claims transaction.

[0072] Optionally, it also includes: obtaining third transaction data information from the blockchain system according to a preset time period, and using the third transaction data information to retrain the risk prediction model, thereby updating the parameters of the risk prediction model.

[0073] Specifically, as time goes by, the fraudster may adopt new means to reduce the performance of the risk prediction model. In order to ensure the timeliness of the risk prediction model, the smart contract needs to retrain the risk prediction model according to a predetermined time period and update the parameters of the risk prediction model.

[0074] Let the current time be t, X t represents the feature matrix corresponding to the risk prediction vector at time t, y t represents the label vector corresponding to the risk prediction vector at time t, θ t represents the model parameters at time t. Thus, the model parameters θ t Initialize and set:

[0075] θ t+1 =θ t (Formula 4)

[0076] Among them, θ t+1 Represents the model parameters of the risk prediction model at time t+1.

[0077] Therefore, based on the above formula 3 and formula 4, the following formula can be obtained:

[0078]

[0079] Among them, the given function J t (θ t+1 ) is the representation of the regularized logistic regression cost function at time t.

[0080] Then, the blockchain system minimizes J through optimization algorithms such as gradient descentt (θ t+1 ), thus obtaining the updated model parameters θ t+1 .

[0081] Thus, by updating the model parameters of the risk prediction model according to a predetermined period, the technical effect of avoiding the degradation of the model performance due to interference from external factors and ensuring the timeliness of the risk prediction model is achieved.

[0082] Optionally, it also includes: the insurance company system encrypts the first transaction data information corresponding to the user terminal, and generates proof information based on a zero-knowledge proof algorithm; and the insurance company system sends the proof information to the supervision system, and the supervision system verifies the proof information.

[0083] Specifically, in this embodiment, zero-knowledge proof is used to ensure the privacy protection and compliance sharing of transaction data information related to the policyholder. For example, the insurance company system can transmit the encrypted information related to the policyholder to the blockchain system, and use zero-knowledge proof to generate proof related to the encrypted information, so that the insurance company system can prove to the regulator that the encrypted information related to the policyholder meets certain conditions without disclosing specific data.

[0084] Therefore, through the above operations, the technical effect of realizing privacy protection and compliant sharing of transaction data related to the policyholder is achieved.

[0085] In addition, the risk prediction model can be embedded in the smart contract. The implementation steps are as follows:

[0086] 1. When the smart contract is initialized, load the pre-trained model parameters;

[0087] 2. Define smart contract functions to extract feature vectors for current claims transactions;

[0088] 3. Call the prediction function and sigmoid function to calculate the fraud risk probability;

[0089] 4. If the fraud risk probability is greater than a predetermined threshold, the current claim transaction is marked as suspicious and an alarm is issued;

[0090] 5. Regularly call the smart contract function and use the new on-chain transaction data increment to retrain the risk prediction model, thereby updating the model parameters;

[0091] 6. Write the updated model parameters into the database for later use.

[0092] Furthermore, the technical solution disclosed in this application can also be stored off-chain. The IPFS distributed storage system is used to store files with large data such as policy attachments, and only the file hash is saved on the chain, taking into account both system performance and data integrity.

[0093] Furthermore, the technical solution disclosed in this application can also be cross-chain docking. Through cross-chain protocols such as blockchain integration framework, it can achieve interconnection with blockchains in other industries such as banking and medical care.

[0094] Furthermore, the technical solution disclosed in this application introduces a regulatory node, allowing regulatory agencies to audit on-chain data in real time and supporting the reporting of on-chain data to regulatory agencies after desensitization.

[0095] Advantages of this application:

[0096] 1. It improves the security, transparency and credibility of insurance business data and can effectively prevent insurance fraud;

[0097] 2. Simplified the claims process, shortened the claims settlement time from weeks to days, and greatly improved the customer experience;

[0098] 3. It realizes the privacy protection and compliance sharing of insurance data, and promotes the collaboration between the insurance industry and other industries;

[0099] 4. Introduce smart contracts to automatically execute claims, reduce labor costs, and improve claims efficiency and accuracy;

[0100] 5. Using the immutable and traceable characteristics of blockchain, single point failures and data quality issues in traditional systems are avoided; and

[0101] 6. Through pluggable authentication mechanisms and fine-grained permission control, seamless integration with traditional systems is achieved.

[0102] Therefore, according to the first aspect of this embodiment, the technical effect of being able to effectively protect the privacy of the policyholder and ensure the security of the claim transaction is achieved.

[0103] In addition, reference Figure 1 As shown, according to the second aspect of this embodiment, a storage medium is provided, wherein the storage medium includes a stored program, wherein when the program is run, a processor executes any one of the above methods.

[0104] Therefore, according to this embodiment, the technical effect of being able to effectively protect the privacy of the policyholder and ensure the security of the claim transaction is achieved.

[0105] It should be noted that, for the above-mentioned method embodiments, for the sake of simplicity, they are all described as a series of action combinations, but those skilled in the art should know that the present invention is not limited by the described action sequence, because according to the present invention, certain steps can be performed in other sequences or simultaneously. Secondly, those skilled in the art should also know that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by the present invention.

[0106] Through the description of the above implementation methods, those skilled in the art can clearly understand that the method according to the above embodiment can be implemented by means of software plus a necessary general hardware platform, and of course can also be implemented by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, a magnetic disk, or an optical disk), and includes a number of instructions for enabling a terminal device (which can be a mobile phone, a computer, a server, or a network device, etc.) to execute the methods described in each embodiment of the present invention.

[0107] Example 2

[0108] Figure 5 The multi-party security fraud device 500 based on blockchain according to the first aspect of this embodiment is shown, and the device 500 corresponds to the method according to embodiment 1. Figure 5 As shown, the device 500 includes: a first encrypted information generation module 510, which is used for the user terminal to encrypt the claim information using a public key to generate the first encrypted information, and upload the first encrypted information to the blockchain system; a second encrypted information generation module 520, which is used for the insurance company system to obtain the first encrypted information from the blockchain system, determine the second encrypted information based on the first encrypted information, and upload the second encrypted information to the blockchain system; a decryption module 530, which is used for the user terminal to obtain the second encrypted information from the blockchain system, and use the private key to decrypt the second encrypted information to determine the claim amount; a risk probability prediction module 540, which is used for the smart contract to call a pre-embedded risk prediction model, and use the risk prediction model and the first transaction data information corresponding to the user terminal to predict the fraud risk probability of the current claim transaction, wherein the first transaction data information includes the claim amount information; and an alarm module 550, which is used to determine that the fraud risk of the current claim transaction is high when the fraud risk probability corresponding to the current claim transaction is greater than a preset threshold, and to alarm the insurance company system.

[0109] Optionally, the first encrypted information generation module 510 includes: a sub-encrypted information generation module, which is used for the user terminal to encrypt the claim fee using the public key to generate the first sub-encrypted information, and the user terminal to encrypt the claim ratio using the public key to generate the second sub-encrypted information.

[0110] Optionally, the second encrypted information generation module 520 includes: a second encrypted information calculation module, which is used for the insurance company system to obtain the first sub-encrypted information and the second sub-encrypted information sent by the blockchain system, and perform homomorphic multiplication operations to determine the second encrypted information.

[0111] Optionally, the risk probability prediction module 540 includes: a sample set construction module, used to collect second transaction data information and construct a sample set based on the collected information, wherein the second transaction data information corresponds to historical claims transactions; a risk prediction model generation module, used to train a binary classification model using the sample set and generate a risk prediction model; and a risk probability prediction submodule, used to obtain first transaction data information corresponding to the user terminal, and perform risk probability prediction on the current claims transaction based on the obtained first transaction data information and the risk prediction model.

[0112] Optionally, the sample set construction module includes: a feature information collection module, used to collect second transaction data information corresponding to different historical claims transactions, and extract multiple groups of risk prediction feature information corresponding to different historical claims transactions from the second transaction data information; a feature vector generation module, used to quantify and normalize multiple groups of risk prediction feature information to generate multiple risk prediction feature vectors; and a sample set construction sub-module, used to construct a sample set based on the generated multiple risk prediction feature vectors.

[0113] Optionally, the device 500 also includes: a parameter updating module, which is used for the smart contract to obtain third transaction data information from the blockchain system according to a preset time period, and use the third transaction data information to retrain the risk prediction model, thereby updating the parameters of the risk prediction model.

[0114] Optionally, the device 500 also includes: a proof information generation module, which is used by the insurance company system to encrypt the first transaction data information corresponding to the user terminal and generate proof information based on a zero-knowledge proof algorithm; and a proof information verification module, which is used by the insurance company system to send the proof information to the supervision system, and the supervision system verifies the proof information.

[0115] Therefore, according to this embodiment, the technical effect of being able to effectively protect the privacy of the policyholder and ensure the security of the claim transaction is achieved.

[0116] Example 3

[0117] Figure 6A multi-party security anti-fraud device 600 based on blockchain according to this embodiment is shown, and the device 600 corresponds to the method according to embodiment 1. Figure 6 As shown, the device 600 includes: a processor; and a memory, connected to the processor, for providing the processor with instructions for processing the following processing steps: the user terminal uses a public key to encrypt the claim information to generate first encrypted information, and uploads the first encrypted information to the blockchain system; the insurance company system obtains the first encrypted information from the blockchain system, determines the second encrypted information based on the first encrypted information, and uploads the second encrypted information to the blockchain system; the user terminal obtains the second encrypted information from the blockchain system, and decrypts the second encrypted information using a private key to determine the claim amount; the smart contract calls a pre-embedded risk prediction model, and uses the risk prediction model and the first transaction data information corresponding to the user terminal to predict the fraud risk probability of the current claim transaction, wherein the first transaction data information includes the claim amount information; and when the fraud risk probability corresponding to the current claim transaction is greater than a preset threshold, it is determined that the fraud risk of the current claim transaction is high, and an alarm is sent to the insurance company system.

[0118] Optionally, the user terminal uses a public key to encrypt the claim information to generate a first encrypted information operation, including: the user terminal uses a public key to encrypt the claim fee to generate a first sub-encrypted information, and the user terminal uses a public key to encrypt the claim ratio to generate a second sub-encrypted information.

[0119] Optionally, the insurance company system obtains the first encrypted information sent by the blockchain system, and determines the operation of the second encrypted information based on the first encrypted information, including: the insurance company system obtains the first sub-encrypted information and the second sub-encrypted information sent by the blockchain system, and performs a homomorphic multiplication operation to determine the second encrypted information.

[0120] Optionally, the smart contract calls a pre-embedded risk prediction model, and uses the risk prediction model and the first transaction data information corresponding to the user terminal to perform a fraud risk probability prediction operation on the current claim transaction, including: collecting second transaction data information, and constructing a sample set based on the collected information, wherein the second transaction data information corresponds to historical claim transactions; using the sample set to train a binary classification model and generate a risk prediction model; and obtaining the first transaction data information corresponding to the user terminal, and based on the obtained first transaction data information and the risk prediction model, performing a risk probability prediction on the current claim transaction.

[0121] Optionally, the operation of collecting second transaction data information and constructing a sample set based on the collected information includes: collecting second transaction data information corresponding to different historical claims transactions, and extracting multiple groups of risk prediction feature information corresponding to different historical claims transactions from the second transaction data information; quantifying and normalizing the multiple groups of risk prediction feature information to generate multiple risk prediction feature vectors; and constructing a sample set based on the generated multiple risk prediction feature vectors.

[0122] Optionally, it also includes: the smart contract obtains third transaction data information from the blockchain system according to a preset time period, and uses the third transaction data information to retrain the risk prediction model, thereby updating the parameters of the risk prediction model.

[0123] Optionally, it also includes: the insurance company system encrypts the first transaction data information corresponding to the user terminal, and generates proof information based on a zero-knowledge proof algorithm; and the insurance company system sends the proof information to the supervision system, and the supervision system verifies the proof information.

[0124] Therefore, according to this embodiment, the technical effect of being able to effectively protect the privacy of the policyholder and ensure the security of the claim transaction is achieved.

[0125] The serial numbers of the above embodiments of the present invention are only for description and do not represent the advantages or disadvantages of the embodiments.

[0126] In the above embodiments of the present invention, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0127] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of units or modules, which can be electrical or other forms.

[0128] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0129] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0130] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, magnetic disk or optical disk and other media that can store program codes.

[0131] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A multi-party security anti-fraud method based on blockchain, characterized in that: include: The user terminal encrypts the claim information using the public key to generate first encrypted information, and uploads the first encrypted information to the blockchain system; The insurance company system obtains the first encrypted information from the blockchain system, determines the second encrypted information based on the first encrypted information, and uploads the second encrypted information to the blockchain system; The user terminal obtains the second encrypted information from the blockchain system, and decrypts the second encrypted information using the private key to determine the claim amount; The smart contract calls a pre-embedded risk prediction model, and uses the risk prediction model and the acquired first transaction data information corresponding to the user terminal to predict the fraud risk probability of the current claim transaction, wherein the first transaction data information includes claim amount information; as well as When the fraud risk probability corresponding to the current claim transaction is greater than a preset threshold, the fraud risk of the current claim transaction is determined to be high, and an alarm is sent to the insurance company system, wherein The user terminal encrypts the claim settlement information using the public key to generate the first encrypted information, including: The user terminal encrypts the claim fee using the public key to generate the first sub-encrypted information, and the user terminal encrypts the claim ratio using the public key to generate the second sub-encrypted information, and wherein The insurance company system obtains the first encrypted information from the blockchain system, and based on the first encrypted information, determines the operation of the second encrypted information, including: The insurance company system obtains the first sub-encrypted information and the second sub-encrypted information from the blockchain system, and performs a homomorphic multiplication operation to determine the second encrypted information.

2. The method according to claim 1, characterized in that The smart contract calls a pre-embedded risk prediction model, and uses the risk prediction model and the acquired first transaction data information corresponding to the user terminal to perform a fraud risk probability prediction operation on the current claim transaction, including: Collecting second transaction data information and constructing a sample set based on the collected information, wherein the second transaction data information corresponds to historical claim transactions; Using the sample set to train a binary classification model and generate the risk prediction model; and First transaction data information corresponding to the user terminal is obtained, and based on the obtained first transaction data information and the risk prediction model, a risk probability prediction is performed on the current claim transaction.

3. The method according to claim 2, characterized in that The operation of collecting the second transaction data information and constructing a sample set based on the collected information includes: Collecting second transaction data information corresponding to different historical claim settlement transactions, and extracting multiple groups of risk prediction feature information corresponding to the different historical claim settlement transactions from the second transaction data information; quantizing and normalizing the multiple sets of risk prediction feature information to generate multiple risk prediction feature vectors; and The sample set is constructed based on the generated multiple risk prediction feature vectors.

4. The method according to claim 3, characterized in that Also includes: The smart contract obtains the third transaction data information from the blockchain system according to a preset time period, and uses the third transaction data information to retrain the risk prediction model, thereby updating the parameters of the risk prediction model.

5. The method according to claim 1, characterized in that Also includes: The insurance company system encrypts the first transaction data information corresponding to the user terminal and generates proof information based on a zero-knowledge proof algorithm; as well as The insurance company system sends the certification information to the supervision system, and the supervision system verifies the certification information.

6. A storage medium, characterized in that: The storage medium includes a stored program, wherein when the program is run, the processor executes the method according to any one of claims 1 to 5.

7. A multi-party security anti-fraud device based on blockchain, characterized in that: include: A first encrypted information generation module, used for the user terminal to encrypt the claim information using a public key to generate first encrypted information, and upload the first encrypted information to the blockchain system; A second encrypted information generation module, used for the insurance company system to obtain the first encrypted information from the blockchain system, determine the second encrypted information based on the first encrypted information, and upload the second encrypted information to the blockchain system; A decryption module, used for the user terminal to obtain the second encrypted information from the blockchain system, and use the private key to decrypt the second encrypted information to determine the claim amount; A risk probability prediction module, which is used for the smart contract to call a pre-embedded risk prediction model, and use the risk prediction model and the acquired first transaction data information corresponding to the user terminal to predict the fraud risk probability of the current claim transaction, wherein the first transaction data information includes claim amount information; as well as The alarm module is used to determine that the fraud risk of the current claim transaction is high and to alarm the insurance company system when the fraud risk probability corresponding to the current claim transaction is greater than a preset threshold.

8. A multi-party security anti-fraud device based on blockchain, characterized in that: include: processor; as well as A memory, connected to the processor, configured to provide the processor with instructions for processing the following processing steps: The user terminal encrypts the claim information using the public key to generate first encrypted information, and uploads the first encrypted information to the blockchain system; The insurance company system obtains the first encrypted information from the blockchain system, determines the second encrypted information based on the first encrypted information, and uploads the second encrypted information to the blockchain system; The user terminal obtains the second encrypted information from the blockchain system, and decrypts the second encrypted information using the private key to determine the claim amount; The smart contract calls a pre-embedded risk prediction model, and uses the risk prediction model and the acquired first transaction data information corresponding to the user terminal to predict the fraud risk probability of the current claim transaction, wherein the first transaction data information includes claim amount information; as well as When the fraud risk probability corresponding to the current claim transaction is greater than a preset threshold, the smart contract determines that the fraud risk of the current claim transaction is high and alerts the insurance company system.

Images