Joint pricing method and device, computer equipment and storage medium

Abstract

The invention discloses a joint pricing method and device, computer equipment and a storage medium. The method comprises the steps that after acquired pricing query information triggered by a user side is input to a first operation side, a joint pricing instruction is sent to a second operation side through the first operation side, and a joint pricing privacy calculation instruction is obtained; performing privacy calculation on the joint pricing privacy calculation instruction through a preset privacy platform model to obtain joint pricing query features; and an operation result corresponding to the joint pricing query feature is fed back to the user side through a preset privacy platform model. The data between the first operation end and the second operation end are integrated through the preset privacy platform model, a suitable model structure is trained, data sharing and intercommunication between different operation ends are achieved, the accuracy and fairness of subsequent product pricing are improved, meanwhile, the effect of privacy calculation can be achieved through the preset privacy platform model, and the user experience is improved. And the data privacy and the data security are protected.

Description

Technical Field

[0001] The present invention relates to the fields of artificial intelligence technology and financial technology, and in particular to a joint pricing method, device, computer equipment and storage medium. Background Art

[0002] Accurate product pricing has always been a common situation for companies in various industries, especially in the insurance industry. With the booming insurance industry, the accuracy and fairness of auto insurance product pricing have also become a focus of widespread attention both inside and outside the industry. However, the traditional auto insurance product pricing model is limited by a single data source and a rigid pricing framework. It is difficult to fully and deeply capture the unique risk characteristics of vehicles and car owners, thus affecting the accuracy and fairness of product pricing. At the same time, in previous technologies, insurance companies usually rely on direct data docking to obtain plaintext information for modeling. This practice also has significant security risks in terms of data privacy protection and data security.

[0003] Therefore, those skilled in the art are in urgent need of finding a new technical solution to solve the above problems. Summary of the invention

[0004] Based on this, a joint pricing method, apparatus, computer equipment and storage medium are provided to solve the technical problems in the prior art of insufficient accuracy and fairness of product pricing and potential security risks in data privacy protection and data security.

[0005] A joint pricing approach comprising: After inputting the acquired pricing query information triggered by the user terminal into the first operating terminal, a joint pricing instruction is issued to the second operating terminal through the first operating terminal to obtain a joint pricing privacy computing instruction; After performing privacy calculation on the joint pricing privacy calculation instruction through a preset privacy platform model, a joint pricing query feature is obtained; the preset privacy platform model is pre-trained by jointly training the data fed back by the first operation terminal and the second operation terminal; The calculation result of the joint pricing query feature response is fed back to the user end through the preset privacy platform model.

[0006] A joint pricing device, comprising: An input module, configured to input the acquired pricing query information triggered by the user terminal into the first operation terminal, and then send a joint pricing instruction to the second operation terminal through the first operation terminal to obtain a joint pricing privacy calculation instruction; a privacy computing module, configured to obtain a joint pricing query feature after performing privacy computing on the joint pricing privacy computing instruction through a preset privacy platform model; the preset privacy platform model is pre-trained by jointly training the data fed back by the first operating terminal and the second operating terminal; A feedback module is used to feed back the calculation result of the joint pricing query feature to the user end through the preset privacy platform model.

[0007] A computer device comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor implements the above-mentioned joint pricing method when executing the computer program.

[0008] A computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the joint pricing method is implemented.

[0009] The above-mentioned joint pricing method, device, computer equipment and storage medium input the obtained pricing query information triggered by the user end into the first operating end, and then send a joint pricing instruction to the second operating end through the first operating end to obtain a joint pricing privacy calculation instruction; after performing privacy calculation on the joint pricing privacy calculation instruction through a preset privacy platform model, a joint pricing query feature is obtained; the preset privacy platform model is pre-trained by jointly training the data fed back by the first operating end and the second operating end; the calculation result of the joint pricing query feature is fed back to the user end through the preset privacy platform model. This scheme integrates the data between the first operating end and the second operating end through the preset privacy platform model, trains a suitable model structure, realizes data sharing and intercommunication between different operating ends, and improves the accuracy and fairness of subsequent product pricing; at the same time, the preset privacy platform model can realize the role of privacy calculation, that is, on the premise of protecting data privacy and data security, a comprehensive assessment of the risk characteristics corresponding to the user is achieved, and the scientificity and rationality of product pricing are improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments of the present invention will be briefly introduced below. Obviously, the accompanying drawings in the following description are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative labor.

[0011] Figure 1 is a schematic diagram of an application environment of a joint pricing method in one embodiment of the present invention; Figure 2is a process diagram of a joint pricing method in one embodiment of the present invention; Figure 3 It is a structural schematic diagram of a joint pricing device in one embodiment of the present invention; Figure 4 is a schematic diagram of a computer device in one embodiment of the present invention; Figure 5 is a schematic diagram of a system in one embodiment of the present invention. DETAILED DESCRIPTION

[0012] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0013] The joint pricing method provided by the present invention can be applied in Figure 1 In an application environment, a client communicates with a server via a network. The client may be, but is not limited to, various personal computers, laptops, smart phones, tablet computers, and portable wearable devices. The server may be implemented as an independent server or a server cluster consisting of multiple servers.

[0014] In one embodiment, if Figure 2 As shown, a joint pricing method is provided, which is applied in Figure 1 The server in the example is used as an example to illustrate the process, including the following steps: S10, after inputting the obtained pricing query information triggered by the user end into the first operating end, a joint pricing instruction is issued to the second operating end through the first operating end to obtain a joint pricing privacy computing instruction.

[0015] Understandably, the user terminal may be an operating terminal directly associated with the user, which may display the required query information for the user; the pricing query information may be the price information of the product that the user needs to check, such as querying the insurance company's pricing information about a certain insurance product. In addition, if the user triggers a button or other key with the same function corresponding to the user terminal, the pricing query information may be triggered to the first operating terminal; the first operating terminal may be the operating terminal where the second operating terminal's cooperative enterprise is located, which may be a terminal or an information processing terminal, such as a cooperative enterprise that may have other information (car, house, money, bank account, etc.) in addition to vehicle information, personal network, geographic location information, and personal behavior data; the second operating terminal may be the operating terminal where the enterprise corresponding to the product is located , it can also be a terminal or an information processing terminal. For example, an insurance company has original data related to auto insurance (including vehicle model, vehicle age, vehicle parts-to-whole ratio, vehicle owner information, historical accident records, etc.); the joint pricing instruction can be a communication instruction between the first operating terminal and the second operating terminal (which is the model result output by the first operating terminal and also the input of the preset privacy platform model), specifically, the first operating terminal converts the pricing query information fed back by the user terminal into an instruction that can be received by the second operating terminal, including requiring the second operating terminal to provide specific product quotation information; the joint pricing privacy computing instruction can be a communication instruction fed back by the second operating terminal to the first operating terminal, specifically used for the first operating terminal to issue a privacy computing instruction to the preset privacy platform model.

[0016] S20, after performing privacy calculation on the joint pricing privacy calculation instruction through a preset privacy platform model, a joint pricing query feature is obtained; the preset privacy platform model is pre-trained by jointly training the data fed back by the first operating end and the second operating end.

[0017] It can be understood that the preset privacy platform model can exchange information with the first operating terminal and the second operating terminal at the same time, that is, in the process of training the platform model, it can obtain the information fed back by the two operating terminals in advance, and after training the information of the two operating terminals, it can form a data encryption for information from the same source that can be fed back to different operating terminals at the same time. In addition, model structures such as GLM, K-MEANS, GMM, XGBOOST, etc. can be selected for data modeling training; the joint pricing query feature can be the target feature information obtained in the platform model based on the pricing query information required by the user, that is, the two parts divided by the platform model (including the cooperative enterprise prediction node and the product corresponding enterprise prediction node) can parse and process the information separately to obtain accurate output results.

[0018] S30, feeding back the calculation result of the joint pricing query feature to the user terminal through the preset privacy platform model.

[0019] It can be understood that the preset privacy platform model can be a central server, which is used to aggregate the results output by the first operation end and the second operation end to obtain the operation result; after obtaining the joint pricing query feature, it is also necessary to obtain the operation result that can be output to the user end through data processing of the first operation end and the second operation end, such as applying the joint pricing query feature of the preset privacy platform model to the premium calculation to obtain the operation result. The operation result can specifically show the price corresponding to the product, such as the risk premium acting on the insurance company's calculation process, thereby affecting the final discount calculation. The final joint pricing premium of the auto insurance product is X yuan.

[0020] In the embodiment of steps S10 to S30, the data between the first operating terminal and the second operating terminal are integrated through the preset privacy platform model, and a suitable model structure is trained to achieve data sharing and intercommunication between different operating terminals, thereby improving the accuracy and fairness of subsequent product pricing. At the same time, the preset privacy platform model can realize the role of privacy computing, that is, under the premise of protecting data privacy and data security, a comprehensive assessment of the risk characteristics corresponding to the user can be achieved, thereby improving the scientificity and rationality of product pricing.

[0021] Furthermore, if Figure 5 As shown, the operation result corresponding to the joint pricing query feature is fed back to the user end through the preset privacy platform model, including: Feedback the joint pricing query feature to the second operating end through the preset privacy platform model; After the second operation end user terminal calculates the joint pricing query feature and the calculation factor, the calculation result is fed back to the first operation end; The operation result is fed back to the user end through the first operation end.

[0022] Understandably, the preset privacy platform model is connected to the first operation terminal and the second operation terminal respectively. Specifically, after the client initially transmits the pricing query information, the information transmitted by different operation terminals arrives at the preset privacy platform model, and then the calculation results are transmitted to different operation terminals through the preset privacy platform model until the final result is fed back to the user terminal; The different processing ends in this embodiment play their respective roles in different nodes of information transmission. They can use their processing methods at their corresponding processing ends, separate the different processing mechanisms, and train a preset privacy platform model for federated learning. This not only improves the efficiency of data processing, but also ensures privacy and data security during data processing.

[0023] Furthermore, before the first operating terminal issues a joint pricing instruction to the second operating terminal, the method further includes: The pricing query information triggered by the user terminal is split by the first operating terminal according to the data type requested, and the split joint pricing instructions are diverted to the second operating terminal.

[0024] Understandably, the data type of the request corresponding to the pricing query data may be a pricing query for the products of Company A or a pricing query for the products of Company B, or a pricing query for the products of auto insurance or a pricing query for life insurance, or a pricing query for the products of Product A or a pricing query for the products of Product B, wherein the second operation terminals of the corresponding enterprises of the products corresponding to different products may be different, and the second operation terminals of the enterprises corresponding to each product are independent, and the corresponding data processing means are also independent; the specific method of splitting can identify the character type corresponding to the pricing query information to determine the specific data type, and in the specific operation process, multiple data types in the information can be split, such as pre-split the numeric type, character type and date type, and then identify the data type corresponding to the character type; diversion can be to disperse the joint pricing instructions to different second operation terminals or different processing nodes of the second operation terminals, and the specific method of diversion can queue the data according to the transmission timing, and process them in sequence according to the queues, which can ensure the sequentiality of data processing and avoid data loss or repeated processing, and can also perform layered and diversion management on the data, which can reduce the amount of simultaneous access to data at each layer and reduce the pressure on the server; This embodiment can split the corresponding pricing query information into joint pricing instructions that are suitable for processing by the second operating end (based on data type splitting) to improve the efficiency of data processing at the second operating end and reduce the pressure on system services.

[0025] Furthermore, before performing privacy calculation on the joint pricing privacy calculation instruction through a preset privacy platform model, the method further includes: Obtain business data through preset middleware; The business data transmitted by the preset middleware is obtained through a preset distributed system, and after analyzing and processing the business data, target training data that can be used as training for the preset privacy platform model is extracted.

[0026] Understandably, the preset middleware may be Kafka, which is a distributed stream processing platform developed by the Apache Software Foundation. It is widely used for its high throughput, persistence, horizontal scalability, support for stream data processing and other features; the preset distributed system is Hadoop, which is a distributed system infrastructure developed by the Apache Foundation. Users can develop distributed programs without understanding the underlying details of the distribution, and make full use of the power of the cluster for high-speed computing and storage; business data may be data from specific businesses of product-corresponding enterprises and cooperative enterprises, such as various insurance product data in insurance business data; Specifically, the preset middleware Kafka cluster is used to process the data through the stream processing framework (such as Kafka Streams) or batch processing framework (such as Apache Flink) in real time or in batches, and the processed data can be further stored in the preset distributed system Hadoop's HDFS for subsequent processing and analysis; the data stored in HDFS is processed, analyzed, feature screened, and feature extracted through the MapReduce, Spark and other tools and technologies in the preset distributed system Hadoop ecosystem to obtain the target training data that meets the preset privacy platform model training requirements; The middleware and big data technology in this embodiment play an important role in the industry's data collection process. Through the integration and collaboration of the two, business data can be efficiently collected, stored, processed and analyzed to provide target feature data to support business decisions.

[0027] Furthermore, before performing privacy calculation on the joint pricing privacy calculation instruction through a preset privacy platform model, the method further includes: Decomposing the same associated target feature data to obtain first target feature data and second target feature data; The first processing model corresponding to the first operating terminal completes encryption processing after training the first target feature data through a first preset privacy computing method; The second processing model corresponding to the second operating end completes encryption processing after training the second target feature data through a second preset privacy computing method.

[0028] It can be understood that the first processing model and the second processing model are part of the preset privacy platform model, and each processing model corresponds to its own operation end. In this way, the data of each operation end is directly fed back to the corresponding processing model during training or actual application, and each encryption processing also exists independently. For example, the processing models corresponding to the first target feature data and the second target feature data mentioned above are different, where the same associated target feature data refers to data related to the same product in the same application scenario, such as data related to the pricing of the same auto insurance product; encryption processing refers to each model using corresponding data encryption, data desensitization and access control methods to achieve encryption processing, and the encrypted content can also be encrypted by methods such as differential privacy and federated learning. Federated learning specifically includes converting the original data of the participants into model information, and then aggregating and training the model information to ensure that the original data does not go out of the domain. The first preset privacy calculation method and the second preset privacy calculation method preferably select the above different methods. The different operation mechanisms of the two parties can improve data security, that is, according to the selected privacy protection technology, the corresponding privacy calculation protocol is implemented. In secure multi-party computing, each party can perform calculations through encryption and secret sharing. It should be noted that the model processing in this embodiment is mainly the processing and modeling of the original data. The expected model results are obtained by data modeling. Different enterprises have different types of data, so the model processing and the corresponding model processing results are also different. At the same time, cross-industry cooperation also occurs frequently. The mode of cooperation is changed from the original plaintext data docking to the model processing obtained by the corresponding operation segments to process the original data and then dock, so as to ensure that the sensitive information of the original plaintext data will not be leaked; This embodiment splits the associated target feature data into two parts and processes them at different operating ends, which can effectively reduce the risk of data leakage. Even if the data or system at one end is attacked, the data at the other end remains safe. In the subsequent stage, the output results of the first processing model and the second processing model can be fused to fully utilize the advantages of the two models, which can further improve the performance of the overall model and improve the accuracy of prediction or classification. Distributing data processing tasks to two different operating ends helps to achieve load balancing and avoid performance degradation or failure of a single operating end due to processing too much data.

[0029] Furthermore, before the first processing model corresponding to the first operating end trains the first target feature data through a first preset privacy computing method and completes the encryption processing, and before the second processing model corresponding to the second operating end trains the second target feature data through a second preset privacy computing method and completes the encryption processing, it also includes: Noise parameters are respectively added to the model parameters in the first initial processing model and the second initial processing model in a preset differential privacy manner to obtain the first processing model and the second processing model.

[0030] Understandably, the preset differential privacy methods may include Laplace mechanism and Gaussian mechanism, wherein the noise added by Laplace mechanism obeys Laplace distribution, and its standard deviation is related to the privacy budget ε and the sensitivity Δf of the function; the noise added by Gaussian mechanism obeys Gaussian distribution, and its standard deviation is also related to the privacy budget (ε, δ) and the sensitivity Δf of the function; Specifically, after determining the differential privacy mechanism to be used, the sensitivity of the real-valued function is calculated, and the sensitivity Δf can be defined as: Δf=maxD,D IIf(D)−f(D′)II, where D and D′ are adjacent data sets; determine the privacy budget, where the privacy budget ε and δ (in the Gaussian mechanism) are parameters that control the strength of privacy protection; calculate the size of the noise to be added according to the selected differential privacy mechanism and the calculated sensitivity Δf, as well as the determined privacy budgets ε and δ, where for the Laplace mechanism: the size of the noise is determined by the following formula: Noise∼Laplace(0,Δf / ε); for the Gaussian mechanism: the size of the noise is determined by the following formula: Noise∼N(0,σ2), where σ is a parameter, and its calculation formula is: σ=Δf2ln(1.25 / δ) / ε; add the calculated noise to the model parameters of the first initial processing model and the second initial processing model, thereby obtaining the first processing model and the second processing model; The process of adding noise parameters to model parameters in this embodiment through a preset differential privacy method ensures that the resulting processing model can maintain good model performance while meeting the privacy protection requirements.

[0031] Furthermore, before the first processing model corresponding to the first operating end trains the first target feature data through a first preset privacy computing method and completes the encryption processing, and before the second processing model corresponding to the second operating end trains the second target feature data through a second preset privacy computing method and completes the encryption processing, it also includes: Importing target training data into an input module of the model through a preset database, and performing data exploration through the preset database to obtain exploration data; Performing data preprocessing on the exploration data to obtain preprocessed data; Performing feature screening and extraction on the pre-processed data by a preset data screening method to obtain screening data; According to the screened data, a preset model algorithm is selected to perform data modeling training to obtain a preliminary model structure; After the preliminary model structure is evaluated and optimized by a preset model evaluation and optimization method, a first processing model and / or a second processing model is obtained.

[0032] Understandably, the input module of the model is used to receive the initial training data; the preset database can be various commonly used databases, wherein, during the import process, it is necessary to pre-establish the connection between the database and the model input module, such as through a specific read_sql function or a specific database Python database; data exploration can be performed through libraries such as Pandas, NumPy, Seaborn, etc., which includes viewing the statistical summary of the data, visualizing the distribution of the data, detecting missing values ​​and outliers, calculating correlations, etc.; data preprocessing includes processing missing values ​​(such as filling, deleting or interpolating), converting data types, encoding categorical variables, standardizing or normalizing numerical features, etc.; the feature screening and extraction method can use PCA to extract specific principal components in the data; the preset model algorithm can be a logistic regression algorithm or a K-MEANS clustering algorithm or a Gaussian mixture model clustering algorithm or a gradient boosting decision tree; the preset model evaluation and optimization method can be to verify the results of the model output, and to continuously adjust the model parameters after comparing the loss value output by the loss function; This embodiment may be a training process of the first processing model or / and the second processing model, and continuously optimizes the structure of the model through a special training method so that the obtained model results are suitable for use in the current scenario and improve the decision-making performance of the model.

[0033] In summary, the above provides a joint pricing method, after the obtained pricing query information triggered by the user end is input into the first operation end, a joint pricing instruction is issued to the second operation end through the first operation end to obtain a joint pricing privacy calculation instruction; after the joint pricing privacy calculation instruction is privacy-calculated through the preset privacy platform model, a joint pricing query feature is obtained; the preset privacy platform model is pre-trained by the data fed back by the first operation end and the second operation end; the calculation result of the joint pricing query feature is fed back to the user end through the preset privacy platform model; this scheme integrates the data between the first operation end and the second operation end through the preset privacy platform model, trains a suitable model structure, realizes data sharing and intercommunication between different operation ends, and improves the accuracy and fairness of subsequent product pricing; at the same time, the preset privacy platform model can realize the role of privacy calculation, that is, under the premise of protecting data privacy and data security, a comprehensive assessment of the risk characteristics corresponding to the user is achieved, and the scientificity and rationality of product pricing are improved; the company's operating costs are reduced, operational efficiency is improved, and at the same time, the corresponding users of the enterprise can enjoy more reasonable and more preferential product prices.

[0034] It should be understood that the order of execution of the steps in the above embodiment does not necessarily mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiment of the present invention.

[0035] In one embodiment, a joint pricing device is provided, which corresponds one-to-one to the joint pricing method in the above embodiment. Figure 3 As shown, the joint pricing device includes an input module 11, a privacy calculation module 12 and a feedback module 13. The functional modules are described in detail as follows: An input module 11 is used to input the acquired pricing query information triggered by the user terminal into the first operating terminal, and then send a joint pricing instruction to the second operating terminal through the first operating terminal to obtain a joint pricing privacy calculation instruction; A privacy computing module 12 is used to obtain a joint pricing query feature after performing privacy computing on the joint pricing privacy computing instruction through a preset privacy platform model; the preset privacy platform model is pre-trained by jointly training the data fed back by the first operating terminal and the second operating terminal; The feedback module 13 is used to feed back the calculation result of the joint pricing query feature to the user terminal through the preset privacy platform model.

[0036] Furthermore, the feedback module includes: A first feedback submodule, configured to feed back the joint pricing query feature to the second operating end through the preset privacy platform model; A second feedback submodule, configured to calculate the combined pricing query feature and the calculation factor through the second operation end user end, and feed back the calculation result to the first operation end; The third feedback submodule is used to feed back the operation result to the user end through the first operation end.

[0037] Furthermore, the joint pricing device also includes: The diversion module is used to split the pricing query information triggered by the user terminal according to the requested data type through the first operation terminal, and divert the split joint pricing instructions to the second operation terminal.

[0038] Furthermore, the joint pricing device also includes: The acquisition module is used to obtain business data through the preset middleware; An extraction module is used to obtain the business data transmitted by the preset middleware through a preset distributed system, and after analyzing and processing the business data, extract target training data that can be used as the preset privacy platform model training.

[0039] Furthermore, the joint pricing device also includes: A disassembly module, used for disassembling the same associated target feature data to obtain the first target feature data and the second target feature data; A first encryption module, configured to perform encryption processing after training the first target feature data using a first preset privacy computing method in a corresponding first processing model in the first operating terminal; The second encryption module is used to complete encryption processing after the second target feature data is trained by the second preset privacy computing method in the corresponding second processing model in the second operating end.

[0040] Furthermore, the joint pricing device also includes: An adding module is used to add noise parameters to the model parameters in the first initial processing model and the second initial processing model respectively through a preset differential privacy method to obtain the first processing model and the second processing model.

[0041] Furthermore, the joint pricing device also includes: A data exploration module, used to import target training data into the input module of the model through a preset database, and perform data exploration through the preset database to obtain exploration data; A data preprocessing module, used to perform data preprocessing on the exploration data to obtain preprocessed data; A screening module, used to perform feature screening and extraction on the pre-processed data by a preset data screening method to obtain screening data; A modeling training module is used to select a preset model algorithm to perform data modeling training based on the screened data to obtain a preliminary model structure; The evaluation and optimization module is used to evaluate and optimize the preliminary model structure through a preset model evaluation and optimization method to obtain a first processing model and / or a second processing model.

[0042] For the specific definition of the joint pricing device, please refer to the definition of the joint pricing method above, which will not be repeated here. Each module in the above-mentioned joint pricing device can be implemented in whole or in part by software, hardware and a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0043] In one embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as follows: Figure 4As shown. The computer device includes a processor, a memory, a network interface and a database connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store data involved in the joint pricing method. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a joint pricing method is implemented.

[0044] In one embodiment, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of the joint pricing method in the above embodiment are implemented, such as Figure 2 Steps S10 to S30 shown include the following steps: After inputting the acquired pricing query information triggered by the user terminal into the first operating terminal, a joint pricing instruction is issued to the second operating terminal through the first operating terminal to obtain a joint pricing privacy calculation instruction; After performing privacy calculation on the joint pricing privacy calculation instruction through a preset privacy platform model, a joint pricing query feature is obtained; the preset privacy platform model is pre-trained by jointly training the data fed back by the first operation terminal and the second operation terminal; The calculation result of the joint pricing query feature response is fed back to the user end through the preset privacy platform model.

[0045] Alternatively, when the processor executes the computer program, the functions of each module / unit of the joint pricing device in the above embodiment are realized, for example Figure 3 The functions of modules 11 to 14 are shown in Figure 1. To avoid repetition, they will not be described here.

[0046] In one embodiment, a computer-readable storage medium is provided on which a computer program is stored. When the computer program is executed by a processor, the steps of the joint pricing method in the above embodiment are implemented, such as Figure 2 Steps S10 to S30 shown include the following steps: After inputting the acquired pricing query information triggered by the user terminal into the first operating terminal, a joint pricing instruction is issued to the second operating terminal through the first operating terminal to obtain a joint pricing privacy calculation instruction; After performing privacy calculation on the joint pricing privacy calculation instruction through a preset privacy platform model, a joint pricing query feature is obtained; the preset privacy platform model is pre-trained by jointly training the data fed back by the first operation terminal and the second operation terminal; The calculation result of the joint pricing query feature response is fed back to the user end through the preset privacy platform model.

[0047] Alternatively, when the computer program is executed by the processor, the functions of each module / unit of the joint pricing device in the above embodiment are realized, for example Figure 3 The functions of modules 11 to 14 are shown in Figure 1. To avoid repetition, they will not be described here.

[0048] A person of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment method can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided by the present invention can include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0049] Those skilled in the art can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example. In actual applications, the above-mentioned functions can be distributed and completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above.

[0050] The embodiments described above are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included in the protection scope of the present invention.

Claims

1. A joint pricing method, characterized in that: include: After inputting the acquired pricing query information triggered by the user terminal into the first operating terminal, a joint pricing instruction is issued to the second operating terminal through the first operating terminal to obtain a joint pricing privacy computing instruction; After performing privacy calculation on the joint pricing privacy calculation instruction through a preset privacy platform model, a joint pricing query feature is obtained; The preset privacy platform model is pre-trained by jointly training the data fed back by the first operation terminal and the second operation terminal; The calculation result of the joint pricing query feature response is fed back to the user end through the preset privacy platform model.

2. The joint pricing method according to claim 1, characterized in that: Feeding back the operation result corresponding to the joint pricing query feature to the user terminal through the preset privacy platform model includes: Feedback the joint pricing query feature to the second operating end through the preset privacy platform model; After the second operation end user terminal calculates the joint pricing query feature and the calculation factor, the calculation result is fed back to the first operation end; The operation result is fed back to the user end through the first operation end.

3. The joint pricing method according to claim 1, characterized in that: Before the first operating terminal issues a joint pricing instruction to the second operating terminal, the method further includes: The pricing query information triggered by the user terminal is split by the first operating terminal according to the data type requested, and the split joint pricing instructions are diverted to the second operating terminal.

4. The joint pricing method according to claim 1, characterized in that: Before performing privacy calculation on the joint pricing privacy calculation instruction through a preset privacy platform model, it also includes: Obtain business data through preset middleware; The business data transmitted by the preset middleware is obtained through a preset distributed system, and after analyzing and processing the business data, target training data that can be used as training for the preset privacy platform model is extracted.

5. The joint pricing method according to claim 1, characterized in that: Before performing privacy calculation on the joint pricing privacy calculation instruction through the preset privacy platform model, the method further includes: Decomposing the same associated target feature data to obtain first target feature data and second target feature data; The first processing model corresponding to the first operating terminal completes encryption processing after training the first target feature data through a first preset privacy computing method; The second processing model corresponding to the second operating end completes encryption processing after training the second target feature data through a second preset privacy computing method.

6. The joint pricing method according to claim 5, characterized in that: The first processing model corresponding to the first operation end trains the first target feature data by a first preset privacy computing method and completes encryption processing, and the second processing model corresponding to the second operation end trains the second target feature data by a second preset privacy computing method and completes encryption processing, further comprising: Noise parameters are respectively added to the model parameters in the first initial processing model and the second initial processing model in a preset differential privacy manner to obtain the first processing model and the second processing model.

7. The joint pricing method according to claim 5 or 6, characterized in that: The first processing model corresponding to the first operation end trains the first target feature data by a first preset privacy computing method and completes encryption processing, and the second processing model corresponding to the second operation end trains the second target feature data by a second preset privacy computing method and completes encryption processing, further comprising: Importing target training data into an input module of the model through a preset database, and performing data exploration through the preset database to obtain exploration data; Performing data preprocessing on the exploration data to obtain preprocessed data; Performing feature screening and extraction on the pre-processed data by a preset data screening method to obtain screening data; According to the screened data, a preset model algorithm is selected to perform data modeling training to obtain a preliminary model structure; After the preliminary model structure is evaluated and optimized by a preset model evaluation and optimization method, a first processing model and / or a second processing model is obtained.

8. A joint pricing device, characterized in that: include: An input module, configured to input the acquired pricing query information triggered by the user terminal into the first operation terminal, and then send a joint pricing instruction to the second operation terminal through the first operation terminal to obtain a joint pricing privacy calculation instruction; A privacy computing module, used to perform privacy computing on the joint pricing privacy computing instruction through a preset privacy platform model to obtain joint pricing query features; The preset privacy platform model is pre-trained by jointly training the data fed back by the first operation terminal and the second operation terminal; A feedback module is used to feed back the calculation result of the joint pricing query feature to the user end through the preset privacy platform model.

9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the joint pricing method according to any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the joint pricing method according to any one of claims 1 to 7 is implemented.

Images