Cross-platform data processing methods, devices, equipment and media

Abstract

This application relates to a cross-platform data processing method, belonging to the field of data security technology. The method includes: displaying data resource description information on a data resource selection interface at the data user end, describing data resources published by the data provider end to a converged computing platform compatible with at least two privacy computing components; each privacy computing component has a corresponding computing component interface; determining target data resource description information from the data resource description information in response to a data resource selection operation; the target data resource pointed to by the target data resource description information being provided by the target data provider end; triggering authorization for the target data resource in response to an authorization request operation; after the target data resource is authorized, instructing the user end to jointly train a business model based on the user end's sample data and the target data resource, during which the data user end and the target data provider end communicate with the converged computing platform based on their respective matching computing component interfaces. This method can meet the needs of more business scenarios.

Description

Technical Field

[0001] This application relates to data security technology, and in particular to a cross-platform data processing method, apparatus, device, and medium. Background Technology

[0002] In the era of big data, the cross-computation of massive amounts of data and the development of artificial intelligence have provided better support for businesses across various industries. However, this data often includes private information, such as internal data of enterprises and institutions. Due to data security considerations, this data is often not publicly available. Therefore, based on the privacy protection needs of enterprises or institutions, data from various parties has formed data silos, preventing data exchange and hindering the realization of data value.

[0003] To enable data flow across industries, traditional technologies primarily involve cross-computation between platforms built on the same privacy-preserving computing components. However, traditional data processing methods do not support data interaction between heterogeneous platforms; that is, they do not support cross-computation between platforms built on different privacy-preserving computing components. This limits the applicable business scenarios and cannot meet the needs of more diverse business scenarios. Summary of the Invention

[0004] Therefore, it is necessary to provide a cross-platform data processing method, apparatus, equipment, and medium that can meet the needs of more business scenarios in response to the above-mentioned technical problems.

[0005] In a first aspect, this application provides a cross-platform data processing method, the method comprising:

[0006] At least one data resource description is displayed on the data resource selection interface at the data user end; the data resource description is used to describe the data resource published by the data provider to the converged computing platform; the converged computing platform is a standardized platform compatible with at least two privacy computing components; each privacy computing component has a corresponding computing component interface;

[0007] In response to a data resource selection operation, a selected target data resource description is determined from the at least one data resource description information; the target data resource mapped to by the target data resource description information is provided by a target data provider.

[0008] In response to the authorization request operation, authorization is triggered for the target data resource; wherein, after the target data resource is authorized for use, it is used to instruct the joint training of the business model based on the user sample data of the data user and the target data resource. When communication interaction is involved in the training process, the data user and the target data provider communicate with the fusion computing platform based on the computing component interface that matches them respectively.

[0009] Secondly, this application provides a cross-platform data processing apparatus, the apparatus comprising:

[0010] The first display module is used to display at least one data resource description information on the data resource selection interface of the data user terminal; the data resource description information is used to describe the data resources published by the data provider terminal to the converged computing platform; the converged computing platform is a standardized platform compatible with at least two privacy computing components; each privacy computing component has a corresponding computing component interface;

[0011] A determination module is configured to, in response to a data resource selection operation, determine a selected target data resource description from the at least one data resource description information; the target data resource mapped to by the target data resource description information is provided by a target data provider.

[0012] The first triggering module is used to trigger authorization for the target data resource in response to the authorization request operation; wherein, after the target data resource is authorized for use, it is used to instruct the user terminal sample data and the target data resource to jointly train the business model. When communication interaction is involved in the training process, the data user terminal and the target data provider terminal communicate with the fusion computing platform based on the computing component interface that matches them respectively.

[0013] In one embodiment, after the target data resource is authorized for use, the first triggering module is further configured to respond to a model training triggering operation to trigger the data user to jointly train the business model to be trained based on the user's sample data and the target data provider's use of the target data resource.

[0014] In one embodiment, the data user end is matched with a first computing component interface; the target data provider end is matched with a second computing component interface; both the data user end and the target data provider end have the same business model to be trained locally; the first triggering module is further configured to respond to a model training triggering operation, train the business model in the data user end based on the sample data of the user end, encrypt the intermediate results in the model training using the first public key issued by the fusion computing platform based on the first computing component interface to obtain a first intermediate feature; send the first intermediate feature to the fusion computing platform based on the first computing component interface, so that the fusion computing platform forwards the first intermediate feature to the target data provider end based on the second computing component interface, thereby instructing the target data provider end to train the target business model based on the first intermediate feature and the second intermediate feature. The data provider has a business model; the second intermediate feature is obtained by encrypting the intermediate results generated when training the business model using the target data resources based on the second public key issued by the converged computing platform; the data provider receives the second intermediate feature sent by the converged computing platform based on the first computing component interface, and continues to train the business model in the data user based on the second intermediate feature and the first intermediate feature to obtain encrypted adjustment parameters; the data provider uploads the encrypted adjustment parameters to the converged computing platform based on the first computing component interface and obtains the adjustment parameters after decryption; the data user updates the business model in the data user based on the adjustment parameters, and obtains the first business model of the data user after training is completed; the first business model and the second business model trained by the target data provider constitute a jointly trained virtual shared business model.

[0015] In one embodiment, the first display module is further configured to display component identifiers of at least two privacy computing components on a privacy computing component selection interface; in response to a component selection operation, determine the selected target component identifier; the first trigger module is further configured to, in response to an authorization request operation, trigger authorization for the target privacy computing component corresponding to the target component identifier and the target data resource; wherein, after the target privacy computing component is authorized for use, it is configured to instruct the data user to communicate with the converged computing platform through the corresponding computing component interface of the target privacy computing component.

[0016] In one embodiment, the target component identifier and the target data resource are specified in the data usage contract to be created; the authorization application operation is a contract creation review operation for the data usage contract; the first triggering module is further configured to, in response to the contract creation review operation, trigger the converged computing platform and the data provider to perform creation review processing on the data usage contract respectively; the target privacy computing component and the target data resource specified in the approved data usage contract are also authorized through use.

[0017] In one embodiment, the target data resource includes provider sample data provided by the target data provider; the first triggering module is further configured to, after the target data resource is authorized for use, in response to a model configuration operation for a business model to be configured, perform model training configuration on the business model to be configured to obtain the configured business model to be trained and the configured sample-related information of the business model to be trained; in response to a model training triggering operation for the business model to be trained, trigger joint training of the business model to be trained based on the user-side sample data and the provider-side sample data corresponding to the sample-related information.

[0018] In one embodiment, the first display module is further configured to display the model call configuration interface of the data user terminal after the business model training is completed and a virtual shared business model is obtained; in response to the model call configuration operation on the model call configuration interface, obtain model call constraint information configured for the virtual shared business model; the model call constraint information is used to constrain the data user terminal's call to the virtual shared business model.

[0019] In one embodiment, the first display module is further configured to display the model deployment configuration interface of the data user terminal after the business model training is completed and a virtual shared business model is obtained; in response to the model deployment trigger operation on the model deployment configuration interface, trigger model deployment configuration for the virtual shared business model; the deployed and configured virtual shared business model is used to perform joint data prediction on the online data of the data user terminal and the target online data corresponding to the target data resource to obtain online prediction results for use by the data user terminal in business processing; the target online data is online data provided by the target data provider.

[0020] The aforementioned cross-platform data processing method displays at least one data resource description on the data resource selection interface of the data user. This description describes the data resource published by the data provider to the converged computing platform. The converged computing platform is a standardized platform compatible with at least two privacy computing components, each with a corresponding computing component interface. In response to a data resource selection operation, a selected target data resource description can be determined from the at least one description. The target data resource mapped to by the target data resource description is provided by the target data provider. In response to an authorization request operation, authorization for the target data resource can be triggered. Once authorized, the target data resource can be used to instruct the joint training of a business model based on the user's sample data and the target data resource. During training, if communication interaction is involved, the data user and the target data provider can communicate with the converged computing platform respectively based on their respective matching computing component interfaces. Compared to traditional data processing methods, this application designs a converged computing platform compatible with at least two privacy computing components. During data interaction, the data user and the target data provider can communicate with the converged computing platform based on their respective matching computing component interfaces. This enables cross-computation of data between platforms built on the same privacy computing component, as well as cross-computation of data between platforms built on different privacy computing components, thus meeting the needs of more business scenarios.

[0021] Thirdly, this application provides a cross-platform data processing method, the method comprising:

[0022] The data provider receives a data resource authorization request sent by the data user; the data resource authorization request carries target data resource description information; the target data resource description information is selected from at least one data resource description information displayed on the data user; the at least one data resource description information is used to describe the data resources published by the data provider to the converged computing platform; the converged computing platform is a standardized platform compatible with at least two privacy computing components; each privacy computing component has a corresponding computing component interface;

[0023] In response to the data resource authorization request, an authorization review interface for the target data resource is displayed; the target data resource is the data resource described in the target data resource description information.

[0024] In response to the authorization operation for the target data resource in the authorization review interface, a resource configuration interface for the target data resource is displayed;

[0025] In response to the mapping configuration operation for the target data resource in the resource configuration interface, a mapping relationship is established between the target data resource description information and the target data resource locally on the data provider end; wherein, after the mapping relationship is established, it is used to instruct the joint training of the business model based on the user sample data of the data user end and the target data resource. When communication interaction is involved in the training process, the data provider end and the data user end communicate with the fusion computing platform based on the computing component interface that matches them respectively.

[0026] Fourthly, this application provides a cross-platform data processing apparatus, the apparatus comprising:

[0027] A receiving module is used for the data provider to receive a data resource authorization request sent by the data user; the data resource authorization request carries target data resource description information; the target data resource description information is selected from at least one data resource description information displayed on the data user; the at least one data resource description information is used to describe the data resources published by the data provider to the converged computing platform; the converged computing platform is a standardized platform compatible with at least two privacy computing components; each privacy computing component has a corresponding computing component interface;

[0028] The second display module is used to display an authorization review interface for the target data resource in response to the data resource authorization request; the target data resource is the data resource described by the target data resource description information; and to display a resource configuration interface for the target data resource in response to the authorization operation for the target data resource in the authorization review interface.

[0029] A module is established to respond to a mapping configuration operation for the target data resource in the resource configuration interface, establishing a mapping relationship between the target data resource description information and the target data resource locally on the data provider; wherein, after the mapping relationship is established, it is used to instruct the joint training of a business model based on the user sample data of the data user and the target data resource. When communication interaction is involved during training, the data provider and the data user communicate with the fusion computing platform based on their respective matching computing component interfaces. Fourthly, this application provides a cross-platform data processing apparatus, the apparatus comprising:

[0030] In one embodiment, the apparatus further includes:

[0031] The generation module is used to generate data resource description information of the data resource to be published in response to the basic information configuration operation of the data resource to be published in the data provider; in response to the resource publishing operation, it triggers the publishing of the data resource to be published to the fusion computing platform; the data resource description information of the published data resource is used to be displayed on the data resource selection interface of the data user.

[0032] In one embodiment, the second display module is further configured to display a call configuration review interface in response to a model call configuration review request sent by a data user for a virtual shared business model; the call configuration review interface displays model call constraint information that is pending review and configured for the virtual shared business model; in response to a call configuration approval operation, the module authorizes the model call constraint information; the authorized model call constraint information is used to constrain the data user's call to the virtual shared business model.

[0033] In the aforementioned cross-platform data processing method, the data provider receives a data resource authorization request from the data user. This request carries target data resource description information, selected from at least one data resource description displayed on the data user. This description describes the data resources published by the data provider to the converged computing platform, a standardized platform compatible with at least two privacy-preserving computing components, each with its own interface. Responding to the authorization request, an authorization review interface for the target data resource is displayed. The target data resource is the one described in the data resource description. Responding to the authorization operation on the review interface, a resource configuration interface for the target data resource is displayed. Responding to the mapping configuration operation on the resource configuration interface, a mapping relationship is established between the target data resource description and the target data resource locally on the data provider. Once the mapping relationship is established, it is used to instruct the joint training of the business model based on the user's sample data and the target data resources. During training, when communication interaction is involved, the data provider and the data user can communicate with the fusion computing platform based on their respective matching computing component interfaces. Compared to traditional data processing methods, this application designs a fusion computing platform compatible with at least two privacy computing components. During data interaction, the data user and the target data provider can communicate with the fusion computing platform based on their respective matching computing component interfaces. This enables cross-computation of data between platforms built on the same privacy computing component, as well as cross-computation between platforms built on different privacy computing components, thus meeting the needs of more business scenarios.

[0034] Fifthly, this application provides a cross-platform data processing method, the method comprising:

[0035] In response to a data resource publishing request sent by the data provider, the data resource description information of the data resource provided by the data provider is displayed on the resource publishing review interface of the platform operation side.

[0036] Based on the resource approval operation applied to the displayed data resource description information, the approved data resource is triggered to be published to the converged computing platform; the converged computing platform is a standardized platform compatible with at least two privacy computing components; each privacy computing component has a corresponding computing component interface;

[0037] The data resource description information of the published data resources is displayed on the data resource selection interface of the data user terminal, so that the data user terminal can select the target data resource description information from the displayed data resource description information, and instruct that after the target data resource mapped to by the target data resource description information is authorized for use, the user terminal sample data of the data user terminal and the target data resource are used to jointly train the business model. When communication interaction is involved in the training process, the data provider and the data user terminal communicate with the fusion computing platform based on the computing component interface that matches them respectively.

[0038] Sixthly, this application provides a cross-platform data processing apparatus, the apparatus comprising:

[0039] The third display module is used to respond to the data resource publishing request sent by the data provider and display the data resource description information of the data resource provided by the data provider on the resource publishing review interface of the platform operation end.

[0040] The second triggering module is used to trigger the publication of the approved data resources to the converged computing platform based on the resource approval operation applied to the displayed data resource description information; the converged computing platform is a standardized platform compatible with at least two privacy computing components; each privacy computing component has a corresponding computing component interface;

[0041] The data resource description information of the published data resources is displayed on the data resource selection interface of the data user terminal, so that the data user terminal can select the target data resource description information from the displayed data resource description information, and instruct that after the target data resource mapped to by the target data resource description information is authorized for use, the user terminal sample data of the data user terminal and the target data resource are used to jointly train the business model. When communication interaction is involved in the training process, the data provider and the data user terminal communicate with the fusion computing platform based on the computing component interface that matches them respectively.

[0042] In the aforementioned cross-platform data processing method, in response to a data resource publishing request sent by the data provider, the data resource description information of the data resource provided by the data provider is displayed on the resource publishing review interface of the platform operator. Based on the resource review approval operation applied to the displayed data resource description information, the approved data resource is triggered to be published to the converged computing platform. The converged computing platform is a standardized platform compatible with at least two privacy computing components; each privacy computing component has a corresponding computing component interface. The published data resource description information is displayed on the data resource selection interface of the data user, allowing the data user to select a target data resource description from the displayed information. It also instructs the user to use the user's sample data and the target data resource to jointly train a business model after the target data resource description is authorized for use. During training, when communication interaction is involved, the data provider and the data user can communicate with the converged computing platform based on their respective matching computing component interfaces. Compared to traditional data processing methods, this application designs a converged computing platform compatible with at least two privacy computing components. During data interaction, the data user and the target data provider can communicate with the converged computing platform based on their respective matching computing component interfaces. This enables cross-computation of data between platforms built on the same privacy computing component, as well as cross-computation of data between platforms built on different privacy computing components, thus meeting the needs of more business scenarios.

[0043] In a seventh aspect, this application provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the various method embodiments of this application.

[0044] Eighthly, this application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps in the various method embodiments of this application.

[0045] Ninthly, this application provides a computer program product, including a computer program that, when executed by a processor, implements the steps in the various method embodiments of this application. Attached Figure Description

[0046] Figure 1 This is an application environment diagram of a cross-platform data processing method in one embodiment;

[0047] Figure 2 This is a flowchart illustrating a cross-platform data processing method executed by a data user in one embodiment.

[0048] Figure 3 This is a schematic diagram illustrating the principle of training a business model in one embodiment;

[0049] Figure 4 This is a schematic diagram of the homepage interface of the data user terminal in one embodiment;

[0050] Figure 5 This is a schematic diagram of the contract creation process in one embodiment;

[0051] Figure 6 A schematic diagram of the privacy computing component selection interface at the data user end in one embodiment;

[0052] Figure 7 This is a schematic diagram of the data resource selection interface on the data user side in one embodiment;

[0053] Figure 8 This is a schematic diagram of the contract information confirmation interface on the data user side in one embodiment;

[0054] Figure 9 This is a schematic diagram of the contract information confirmation interface on the data user side in another embodiment;

[0055] Figure 10 This is a schematic diagram of the model training configuration interface on the data user side in one embodiment;

[0056] Figure 11 This is a schematic diagram of the model training configuration interface for the data user end in another embodiment;

[0057] Figure 12 This is a schematic diagram of the model call configuration interface for the data user end in one embodiment;

[0058] Figure 13 This is a schematic diagram of the model deployment configuration interface for the data user end in one embodiment;

[0059] Figure 14 A schematic diagram of the model deployment configuration interface for the data user end in another embodiment;

[0060] Figure 15 This is a flowchart illustrating a cross-platform data processing method executed by a data provider in one embodiment.

[0061] Figure 16 This is a schematic diagram of the authorization and review interface of the data provider in one embodiment;

[0062] Figure 17 This is a schematic diagram of the resource configuration interface of the data provider in one embodiment;

[0063] Figure 18 A schematic diagram of the basic information configuration interface for data resources to be published in one embodiment;

[0064] Figure 19 This is a schematic diagram of the call configuration review interface of the data provider in one embodiment;

[0065] Figure 20 This is a flowchart illustrating a cross-platform data processing method executed by the platform operator in one embodiment.

[0066] Figure 21 This is a schematic diagram of the resource publishing and review interface on the platform operation side in one embodiment;

[0067] Figure 22 This is a flowchart illustrating a cross-platform data processing method executed by a data user in another embodiment;

[0068] Figure 23 This is a schematic diagram of data flow and interaction between various platforms in one embodiment;

[0069] Figure 24 This is an overall architecture diagram of the converged computing platform in one embodiment;

[0070] Figure 25 This is an architecture diagram of the management and control center of the converged computing platform in one embodiment;

[0071] Figure 26 This is a schematic diagram of a cross-platform data processing flow in yet another embodiment;

[0072] Figure 27 This is a schematic diagram of the data resource description information management process of a converged computing platform in one embodiment;

[0073] Figure 28 This is a data architecture diagram of a converged computing platform in one embodiment;

[0074] Figure 29 This is a deployment architecture diagram of a converged computing platform in one embodiment;

[0075] Figure 30 This is a structural block diagram of a cross-platform data processing device in one embodiment;

[0076] Figure 31 This is a structural block diagram of a cross-platform data processing device in another embodiment;

[0077] Figure 32 This is a structural block diagram of a cross-platform data processing device in yet another embodiment;

[0078] Figure 33 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0079] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0080] The cross-platform data processing method provided in this application can be applied to, for example... Figure 1 In the application environment shown, data user 102 and data provider 104 communicate with the converged computing platform 106 via a network. Data user 102 and data provider 104 can be, but are not limited to, various desktop computers, laptops, smartphones, tablets, IoT devices, and portable wearable devices. IoT devices can include smart speakers, smart TVs, smart air conditioners, and smart in-vehicle devices. Portable wearable devices can include smartwatches, smart bracelets, and head-mounted devices. The converged computing platform 106 can be deployed on a server, which can be a standalone physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms.

[0081] Data user terminal 102 can display at least one data resource description on the data resource selection interface. This description describes the data resource published by data provider terminal 104 to converged computing platform 106. Converged computing platform 106 is a standardized platform compatible with at least two privacy computing components; each privacy computing component has a corresponding computing component interface. In response to a data resource selection operation, data user terminal 102 can determine the selected target data resource description from at least one data resource description. The target data resource mapped by the target data resource description is provided by target data provider terminal 104. In response to an authorization request operation, data user terminal 102 can trigger authorization for the target data resource. Once authorized, the target data resource is used to instruct the joint training of a business model based on user sample data from data user terminal 102 and the target data resource. During training, if communication interaction is involved, data user terminal 102 and target data provider terminal 104 communicate with converged computing platform 106 based on their respective matching computing component interfaces.

[0082] It should be noted that some embodiments of this application employ artificial intelligence (AI) technology in their cross-platform data processing methods. For example, jointly training a business model based on user-side sample data and target data resources, and jointly predicting online data from the user-side and the target online data corresponding to the target data resources to obtain online prediction results for use by the user-side in business processing, also fall under the category of online prediction results obtained using AI technology.

[0083] Furthermore, some embodiments of this application utilize blockchain technology for cross-platform data processing methods. For instance, user-side sample data, target data resources provided by the target data provider, and intermediate data generated during the training of business models can all be stored on the blockchain to prevent data tampering.

[0084] Blockchain is a novel application model of computer technologies such as distributed data storage, peer-to-peer transmission, consensus mechanisms, and cryptographic algorithms. Essentially, a blockchain is a decentralized database, a chain of data blocks linked together using cryptographic methods. Each data block contains information about a batch of network transactions, used to verify the validity of the information (anti-counterfeiting) and generate the next block. A blockchain can include an underlying platform, a platform product service layer, and an application service layer.

[0085] In one embodiment, such as Figure 2 As shown, a cross-platform data processing method is provided. This embodiment applies this method to... Figure 1 Taking terminal 102 as an example, the following steps are included:

[0086] Step 202: Display at least one data resource description on the data resource selection interface of the data user; the data resource description is used to describe the data resources published by the data provider to the converged computing platform; the converged computing platform is a standardized platform compatible with at least two privacy computing components; each privacy computing component has a corresponding computing component interface.

[0087] In this context, "data user" refers to a terminal that uses data from other terminals for corresponding business processing. "Data provider" refers to a terminal that provides data to other terminals for corresponding business processing. The data resource selection interface is the interface on the data user's end used to select data resources. It can be understood that the converged computing platform can connect the platform corresponding to the data user's end and the platform corresponding to the data provider's end. The platforms corresponding to the data user's end and the data provider's end can be homogeneous or heterogeneous. A homogeneous platform means that the platforms corresponding to the data user's end and the data provider's end are built based on the same privacy computing component. A heterogeneous platform means that the platforms corresponding to the data user's end and the data provider's end are built based on different types of privacy computing components. A privacy computing component is an algorithmic component used for privacy computing. The computing component interface is a standardized interface designed within the converged computing platform for connecting to privacy computing components.

[0088] Specifically, the data user can display a data resource selection interface and render at least one data resource description on the displayed data resource selection interface.

[0089] In one embodiment, the data resource description information may include at least one of the following: the name of the data resource, the domain to which the data resource belongs, the format of the data resource, the provider information of the data resource, the status of the data resource, the update time of the data resource, and the time when the data resource was published to the converged computing platform.

[0090] In one embodiment, the privacy computing component may specifically include at least one of a federated learning algorithm component and a multi-party secure computing component.

[0091] In one embodiment, there may be multiple data providers, and multiple data resource descriptions. It is understood that multiple data resource descriptions can be published to the converged computing platform by different data providers. For example, the data resource selection interface on the data user displays data resource descriptions A, B, and C. Data resource description A is published to the converged computing platform by data provider 1, data resource description B is published to the converged computing platform by data provider 2, and data resource description C is published to the converged computing platform by data provider 3.

[0092] Step 204: In response to the data resource selection operation, determine the selected target data resource description information from at least one data resource description information; the target data resource mapped to by the target data resource description information is provided by the target data provider.

[0093] The data resource selection operation is used to trigger the selection of a data resource. The target data resource description information is the data resource description information selected as the target from at least one data resource description information. The target data provider is the data provider that publishes the target data resource mapped by the target data resource description information to the converged computing platform. The target data resource is the data resource mapped by the target data resource description information. The target data resource description information displayed on the data resource selection interface of the data user only describes the target data resource published by the target data provider to the converged computing platform; it does not represent the target data resource itself. In reality, the target data provider does not actually send its local target data resource to the converged computing platform; the target data resource never leaves the target data provider's local area. There is a mapping relationship between the target data resource description information and the target data resource; the target data resource can be determined based on the target data resource description information.

[0094] Specifically, the data user can trigger a data resource selection operation, and the data user can respond to the data resource selection operation by determining the selected target data resource description information from at least one data resource description information displayed on the data resource selection interface.

[0095] In one embodiment, the target data resource description information may include multiple entries. It is understood that the data user can select multiple data resource description information as the target data resource description information from at least one data resource description information displayed on the data resource selection interface.

[0096] Step 206: In response to the authorization request operation, authorization is triggered for the target data resource; wherein, after the target data resource is authorized for use, it is used to instruct the user's sample data and the target data resource to jointly train the business model. In the case of communication interaction during the training process, the data user and the target data provider communicate with the fusion computing platform based on their respective matching computing component interfaces.

[0097] The authorization request operation is used to trigger the authorization of usage rights for the target data resource. The user-side sample data is offline data stored locally on the data user's end, used as sample data for training the business model. Both the data user and the target data provider communicate with the fusion computing platform based on their respective matching computing component interfaces. It can be understood that the fusion computing platform is designed with computing component interfaces that match their respective privacy computing components. During training, when communication interaction is involved, the data user can communicate with the fusion computing platform based on its matching computing component interface, and simultaneously, the data provider can also communicate with the fusion computing platform based on its matching computing component interface.

[0098] Specifically, the data user can trigger an authorization request operation, and the data user can respond to the authorization request operation to trigger the authorization of usage rights for the target data resource.

[0099] In the aforementioned cross-platform data processing method, at least one data resource description is displayed on the data resource selection interface of the data user. This description describes the data resource published by the data provider to the converged computing platform. The converged computing platform is a standardized platform compatible with at least two privacy computing components, each with a corresponding computing component interface. In response to a data resource selection operation, a selected target data resource description can be determined from the at least one description. The target data resource mapped to by the target data resource description is provided by the target data provider. In response to an authorization request operation, authorization for the target data resource can be triggered. Once authorized, the target data resource can be used to instruct the joint training of a business model based on the user's sample data and the target data resource. During training, if communication interaction is involved, the data user and the target data provider can communicate with the converged computing platform respectively based on their respective matching computing component interfaces. Compared to traditional data processing methods, this application designs a converged computing platform compatible with at least two privacy computing components. During data interaction, the data user and the target data provider can communicate with the converged computing platform based on their respective matching computing component interfaces. This enables cross-computation of data between platforms built on the same privacy computing component, as well as cross-computation of data between platforms built on different privacy computing components, thus meeting the needs of more business scenarios.

[0100] In one embodiment, after the target data resource is authorized for use, the method further includes: responding to a model training trigger operation to trigger the data user to jointly train the business model to be trained based on the user's sample data and the target data provider's use of the target data resource.

[0101] Among them, the model training trigger operation is used to trigger the model training operation of the business model.

[0102] Specifically, the user can trigger the model training trigger operation, and the data user can respond to the model training trigger operation to trigger the data user to jointly train the business model to be trained based on the user's sample data and the target data provider based on the target data resources. It can be understood that during the model training process, the user's sample data does not leave the local data user, and the target data resources of the target data provider also do not leave the local target data provider.

[0103] In the above embodiments, by responding to the model training trigger operation, the data user can be triggered to jointly train the business model to be trained based on the user's sample data and the target data provider's target data resources, which can make the trained virtual shared business model have higher prediction accuracy.

[0104] In one embodiment, the data user end is matched with the interface of the first computing component; the target data provider end is matched with the interface of the second computing component; both the data user end and the target data provider end have the same business model to be trained locally; in response to a model training trigger operation, the data user end is triggered to jointly train the business model to be trained based on the user end sample data and the target data provider end using target data resources, including: in response to the model training trigger operation, training the business model in the data user end based on the user end sample data; encrypting the intermediate results in the model training using the first public key issued by the fusion computing platform based on the first computing component interface to obtain a first intermediate feature; sending the first intermediate feature to the fusion computing platform based on the first computing component interface, so that the fusion computing platform forwards the first intermediate feature to the target data provider end based on the second computing component interface to indicate the target data... The data provider trains a business model in the target data provider based on a first intermediate feature and a second intermediate feature. The second intermediate feature is obtained by encrypting the intermediate results generated when training the business model using the target data resources based on a second public key issued by the converged computing platform. The data provider receives the second intermediate feature sent by the converged computing platform through the first computing component interface and continues to train the business model in the data user based on the second intermediate feature and the first intermediate feature to obtain encrypted adjustment parameters. The data provider uploads the encrypted adjustment parameters to the converged computing platform through the first computing component interface and obtains the adjustment parameters after decryption. The data user updates the business model in the data user based on the adjustment parameters, and the first business model of the data user is obtained after training is completed. The first business model and the second business model trained by the target data provider constitute a jointly trained virtual shared business model.

[0105] The first computing component interface is the computing component interface in the converged computing platform that matches the privacy computing component on the data user's end. The second computing component interface is the computing component interface in the converged computing platform that matches the privacy computing component on the data provider's end. The first public key is the public key issued by the converged computing platform based on the first computing component interface. The second public key is the public key issued by the converged computing platform based on the second computing component interface. The first intermediate feature is the intermediate feature obtained by encrypting the intermediate results generated when training a business model based on user-side sample data using the first public key issued by the converged computing platform. The encryption adjustment parameters are the parameters obtained by continuing to train the business model on the data user's end based on the encrypted second intermediate feature and the encrypted first intermediate feature. The adjustment parameters are the parameters obtained by the converged computing platform after decrypting the encryption adjustment parameters, used to adjust the model parameters of the business model during model training. The first business model is the business model obtained by training the business model to be trained locally on the data user's end. The second business model is the business model obtained by training the business model to be trained locally on the data provider's end. The virtual shared business model is a virtual business model jointly formed by the first and second business models, and shared by both the data user and the data provider. It can be understood that the virtual shared business model is not a real business model; some of its model parameters are stored on the data user's end, and the other part is stored on the data provider's end.

[0106] Specifically, a user can trigger a model training operation, and the data user can respond to the model training operation and train the business model in the data user based on the user's sample data, obtaining the intermediate results of the model training in the data user. The converged computing platform can issue a first public key to the data user via a first computing component interface, and a second public key to the data provider via a second computing component interface. The data user can encrypt the intermediate results generated when training the business model based on the user's sample data using the first public key issued by the converged computing platform, obtaining a first intermediate feature. The target data provider can encrypt the intermediate results generated when training the business model using the target data resources using the second public key issued by the converged computing platform, obtaining a second intermediate feature. The data user can send the first intermediate feature to the converged computing platform via the first computing component interface, and the converged computing platform can forward the received first intermediate feature to the target data provider via the second computing component interface. The target data provider can train the business model in the target data provider based on the first and second intermediate features, obtaining the corresponding encryption adjustment parameters for the target data provider. The target data provider can upload its corresponding encrypted adjustment parameters to the converged computing platform via the second computing component interface. The converged computing platform can decrypt the encrypted adjustment parameters to obtain the decrypted adjustment parameters and return them to the target data provider via the second computing component interface. The target data provider can update its business model based on these adjustment parameters, resulting in a second business model after training. The data provider can send a second intermediate feature to the converged computing platform via the second computing component interface. The converged computing platform can forward the received second intermediate feature to the data user via the first computing component interface. The data user can train its business model based on the first and second intermediate features to obtain the corresponding encrypted adjustment parameters. The data user can upload its corresponding encrypted adjustment parameters to the converged computing platform via the first computing component interface. The converged computing platform can decrypt the encrypted adjustment parameters to obtain the decrypted adjustment parameters and return them to the data user via the first computing component interface. The data user can update the business model in the data user based on the corresponding adjustment parameters, and obtain the first business model of the data user after training is completed.

[0107] For example, such as Figure 3As shown, platform A is the platform corresponding to data user A, and platform B is the platform corresponding to data provider B. The cross-platform data processing method of this application enables the joint training of a business model using data from both platform A and platform B. See also... Figure 3 In the example 'a', a firewall exists between platform A and platform B during model training to protect data privacy. There is no data interaction between platforms A and B during training. Both platforms A and B have the same business model to be trained locally. Before jointly training the business model, the user-side sample data from platform A and the provider-side sample data from platform B can be encrypted before sample alignment. After sample alignment, see [link to documentation]. Figure 3 In the context of 'b', the aligned data A corresponding to platform A and the aligned data B corresponding to platform B, after joint alignment, are used for encrypted training of the business model. For details regarding the encrypted model training process, please refer to... Figure 3In step b, data user A can train its business model based on aligned data A, obtaining intermediate training results. The fusion computing platform can issue a first public key to data user A via a first computing component interface and a second public key to data provider B via a second computing component interface. Data user A can encrypt the intermediate results generated during business model training based on aligned data A using the first public key issued by the fusion computing platform, obtaining a first intermediate feature. Target data provider B can encrypt the intermediate results generated during business model training using aligned data B using the second public key issued by the fusion computing platform, obtaining a second intermediate feature. Data user A can send the first intermediate feature to the fusion computing platform via the first computing component interface, and the fusion computing platform can forward the received first intermediate feature to target data provider B via the second computing component interface. Target data provider B can train its business model based on the first and second intermediate features, obtaining corresponding encryption adjustment parameters. Target data provider B can upload its corresponding encrypted adjustment parameters to the converged computing platform via the second computing component interface. The converged computing platform can decrypt the encrypted adjustment parameters of target data provider B, obtaining the decrypted adjustment parameters, and return them to target data provider B via the second computing component interface. Target data provider B can update its business model based on these adjustment parameters, obtaining a second business model after training. Target data provider B can send its second intermediate feature to the converged computing platform via the second computing component interface. The converged computing platform can forward the received second intermediate feature to data user A via the first computing component interface. Data user A can train its business model based on the first and second intermediate features, obtaining its corresponding encrypted adjustment parameters. Data user A can then upload its corresponding encrypted adjustment parameters to the converged computing platform via the first computing component interface. The converged computing platform can decrypt the encrypted adjustment parameters corresponding to data user A, obtaining the decrypted adjustment parameters for data user A, and then return these adjustment parameters to data user A via the first computing component interface. Data user A can then update its business model based on these adjustment parameters, resulting in its first business model after training. Further, see... Figure 3 In the model 'a', the first business model and the second business model constitute a jointly trained virtual shared business model.

[0108] In the above embodiments, during data interaction, the data user and the target data provider can communicate with the converged computing platform based on their respective matching computing component interfaces. This enables cross-computation of data between platforms built on different privacy computing components, meeting the needs of more business scenarios. Simultaneously, the data user and the target data provider do not directly interact with each other. Instead, they perform privacy computations on the local data of both the data user and the target data provider through the privacy computing component, and encrypt the data using a public key issued by the converged computing platform before interaction. This further ensures the data security of both the data user and the target data provider.

[0109] In one embodiment, the method further includes: displaying component identifiers of at least two privacy computing components on a privacy computing component selection interface; determining the selected target component identifier in response to a component selection operation; and triggering authorization for the target data resource in response to an authorization request operation, including: triggering authorization for the target privacy computing component and the target data resource corresponding to the target component identifier in response to the authorization request operation; wherein, after the target privacy computing component is authorized for use, it is used to instruct the data user to communicate with the converged computing platform through the corresponding computing component interface of the target privacy computing component.

[0110] Here, the component identifier is a string used to uniquely identify the privacy-preserving computation component. The string may include at least one of the following: text, numbers, letters, and special characters. The component selection operation is used to select the target component identifier. The target component identifier is the component identifier used as the target, i.e., the component identifier selected in response to the component selection operation. The target privacy-preserving computation component is the privacy-preserving computation component used as the target, i.e., the privacy-preserving computation component corresponding to the target component identifier.

[0111] Specifically, the data user can display component identifiers for at least two privacy computing components on the privacy computing component selection interface. The data user can trigger a component selection operation, and in response, the data user can determine the selected target component identifier. The data user can also trigger an authorization request operation, and in response, the data user can trigger the converged computing platform to authorize the target privacy computing component corresponding to the target component identifier, and trigger the target data provider to authorize the target data resource. After the target privacy computing component is authorized for use by the data user, it can be used to instruct the data user to communicate with the converged computing platform through the corresponding computing component interface of the target privacy computing component.

[0112] In the above embodiments, displaying component identifiers of at least two privacy computing components on the privacy computing component selection interface facilitates the selection of the type of privacy computing component and obtains the target component identifier. By responding to the authorization request operation, authorization is triggered for the target privacy computing component and target data resources corresponding to the target component identifier. This facilitates subsequent instructions to the user terminal to communicate with the converged computing platform through the corresponding computing component interface of the target privacy computing component, thereby achieving cross-platform and even cross-component data interaction.

[0113] In one embodiment, the target component identifier and the target data resource are specified in the data usage contract to be created; the authorization application operation is a contract creation review operation for the data usage contract; in response to the authorization application operation, authorization is triggered for the target privacy computing component and the target data resource corresponding to the target component identifier, including: in response to the contract creation review operation, the converged computing platform and the data provider are triggered to perform creation review processing on the data usage contract respectively; the target privacy computing component and the target data resource specified in the approved data usage contract are also authorized for use.

[0114] Among them, the data usage contract is a data usage agreement established between the data user, the converged computing platform, and the target data provider.

[0115] Specifically, the data user can trigger a contract creation and review process. In response, the data user can trigger the converged computing platform to review the data usage contract and the data provider to review the target data usage contract. Once the converged computing platform approves the data usage contract, the target privacy computing component specified in the approved contract is also authorized for use. Similarly, once the target data provider approves the data usage contract, the target data resources provided by the target data provider are also authorized for use.

[0116] In one embodiment, Figure 4 This is a schematic diagram of the homepage interface for data users. As seen on the homepage, users can view project contract information in the "Overview" section, including the total number of projects, the number of contracts awaiting authorization, the number of contracts awaiting publication, the number of published contracts, and detailed lists of individual project contracts, such as contract name, contract type, contract creation time, contract status, and publication status. The homepage also displays the current status of common data resources, such as the total data volume and the update time of data asset management. Users can also set up and manage their accounts in the "Account Center," manage data resources in the "Data Resources" section, manage individual projects in the "Project Contract Management" section, and manage pending messages in the "Message Center."

[0117] In one embodiment, such as Figure 5 As shown, the contract creation process mainly includes steps such as filling in the basic information of the contract, confirming the contract information, selecting data resources, waiting for the contract creation review, and completing the contract creation.

[0118] In one embodiment, such as Figure 6 As shown, the data user can select the target component identifier from the component identifiers displayed in the "Contract Type" section of the privacy computing component selection interface to determine the target privacy computing component. This allows the data user to subsequently communicate with the converged computing platform based on the computing component interface that matches the target privacy computing component. In the privacy computing component selection interface, the data user can also complete the basic information of the contract, such as editing the contract name, the project to which the contract belongs, the expected contract duration, the expected number of contract calls, and providing a brief description of the contract's application scenario.

[0119] In one embodiment, such as Figure 7 As shown, the data resource selection interface on the data user terminal can display data resource description information (i.e., data resource name, data provider, resource status, update time, release time, etc.). The data user terminal can determine the data resource to use based on the data resource description information displayed on the data resource selection interface.

[0120] In one embodiment, such as Figure 8 As shown, the contract information confirmation interface on the data user's end displays basic contract information, including contract name, contract type, project, estimated contract usage time, estimated number of contract calls, and contract application scenario notes. In addition, the contract information confirmation interface also displays the contract creation review status.

[0121] In one embodiment, such as Figure 9 As shown, the contract information confirmation interface on the data user's end displays not only the basic information of the contract and the review status of the contract creation, but also information about the data resources selected by the data user, including the data resource name and data provider information.

[0122] In the above embodiments, by introducing data usage contracts, the authorization of usage rights for the target privacy computing components and target data resources specified in the data usage contract is granted during the creation and review process of the data usage contract, making the authorization application more convenient.

[0123] In one embodiment, the target data resource includes provider sample data provided by the target data provider; the method further includes: after the target data resource is authorized for use, in response to a model configuration operation for a business model to be configured, performing model training configuration on the business model to be configured to obtain the configured business model to be trained and the sample-related information of the configured business model to be trained; in response to a model training trigger operation for the business model to be trained, triggering joint training of the business model to be trained based on the user-side sample data and provider-side sample data corresponding to the sample-related information.

[0124] Among them, the provider-side sample data is offline data stored locally on the data provider's end, used as sample data for training the business model. Model configuration operations are used to configure the business model to be configured for training. Model training configuration includes configuration for the business model to be configured and configuration for the user-side sample data. Sample-related information is information obtained from the sample configuration of the user-side sample data used to train the business model, and is related to the user-side sample data.

[0125] Specifically, after the target data resource is authorized for use, the data user can trigger a model configuration operation for the business model to be configured. In response to this operation, the data user can configure the model training process for the business model to be configured, configuring the business model to be trained and the user's sample data accordingly, thus obtaining the configured business model to be trained and its sample-related information. The data user can then trigger a model training trigger operation for the business model to be trained. In response, the data user can trigger joint training of the business model to be trained based on the user's sample data and the provider's sample data corresponding to the sample-related information.

[0126] In one embodiment, the target data resource is specified in the data usage contract to be created; the authorization request operation is a contract creation and review operation for the data usage contract. The model configuration operation is a contract configuration operation for the data usage contract.

[0127] In one embodiment, such as Figure 10 As shown, after the target data resource is authorized for use, the model training configuration interface on the data user's end can display the authorization status of the associated data resource, the data access status, and the review status of the associated data resource by the data provider and the platform operator. After the target data resource is authorized for use, the data user can click "Edit Configuration" on the model training configuration interface on the data user's end to trigger the model configuration operation for the business model, so as to configure the business model and obtain the business model to be trained.

[0128] In one embodiment, such as Figure 11 As shown, after the data user clicks "Edit Configuration" on the model training configuration interface of the data user terminal, the data user terminal can display the model training configuration interface, which can display information such as contract name, contract type, contract creation and review status, contract release status, and authorization time. Clicking "Train" in the model training configuration interface will trigger the model training operation, initiating joint training of the business model to be trained based on the user's sample data and the provider's sample data. Clicking "Predict" in the model training configuration interface will trigger model prediction for the business model. Clicking "Release" in the model training configuration interface will trigger the release configuration of the business model.

[0129] In the above embodiments, by configuring the business model for training, and by jointly training the business model to be trained using the user-side sample data and provider-side sample data corresponding to the sample-related information obtained from the configuration, the model training effect can be improved, thereby improving the prediction accuracy of the trained virtual shared business model.

[0130] In one embodiment, the method further includes: after the business model training is completed and a virtual shared business model is obtained, displaying the model call configuration interface of the data user terminal; in response to the model call configuration operation on the model call configuration interface, obtaining the model call constraint information configured for the virtual shared business model; the model call constraint information is used to constrain the data user terminal's call to the virtual shared business model.

[0131] The model call configuration interface is the interface used by the data user to configure the model call constraint information of the virtual shared business model. The model call configuration operation is the operation used to trigger the configuration of the model call constraint information of the virtual shared business model. The model call constraint information is the constraint information for calling the virtual shared business model.

[0132] Specifically, after the business model training is completed and a virtual shared business model is obtained, the data user can display the model call configuration interface. The data user can trigger the model call configuration operation on the model call configuration interface, and the data user can respond to the model call configuration operation on the model call configuration interface to obtain the model call constraint information configured for the virtual shared business model.

[0133] In one embodiment, the model invocation constraint information configured for the virtual shared service model may specifically include at least one of the following: the validity period for invoking the virtual shared service model and the number of times the virtual shared service model can be invoked.

[0134] In one embodiment, the target data resource is specified in the data usage contract to be created; the authorization request operation is a contract creation and review operation for the data usage contract. The model invocation configuration operation is a contract publication configuration operation for the data usage contract.

[0135] In one embodiment, such as Figure 12 As shown, on the model call configuration interface of the data user, the data user can edit the model call constraint information configured for the virtual shared business model, such as the contract usage period and the number of contract calls. Furthermore, the data user can obtain the model call constraint information configured for the virtual shared business model.

[0136] In the above embodiments, by performing model call configuration operations, model call constraint information configured for the virtual shared business model can be obtained, thereby constraining the data user's call to the virtual shared business model and further improving the security of local data on the data provider side.

[0137] In one embodiment, the method further includes: after the business model training is completed and a virtual shared business model is obtained, displaying the model deployment configuration interface of the data user; in response to the model deployment trigger operation on the model deployment configuration interface, triggering model deployment configuration for the virtual shared business model; the deployed and configured virtual shared business model is used to perform joint data prediction on the online data of the data user and the target online data corresponding to the target data resource, to obtain online prediction results for use by the data user in business processing; the target online data is the online data provided by the target data provider.

[0138] The model deployment configuration interface is used to configure the deployment of virtual shared service models. The model deployment trigger operation is used to trigger the model deployment configuration operation for the virtual shared service models.

[0139] Specifically, after the business model training is completed and a virtual shared business model is obtained, the data user can display the model deployment configuration interface. The data user can trigger a model deployment operation on the model deployment configuration interface, and the data user can respond to this operation by triggering model deployment configuration for the virtual shared business model. The deployed and configured virtual shared business model can be used to perform joint prediction on the online data of the data user and the target online data corresponding to the target data resource, obtaining online prediction results. The data user can then use these online prediction results during business processing.

[0140] In one embodiment, the virtual shared service model includes a first service model deployed at the data user end and a second service model deployed at the target data provider end. By using the first service model deployed at the data user end to predict the online data at the data user end and the target online data corresponding to the target data resources, the online prediction result at the data user end can be obtained. By using the second service model deployed at the target data provider end to predict the online data at the data user end and the target online data corresponding to the target data resources, the online prediction result at the target data provider end can be obtained. Furthermore, the fusion computing platform can determine the final online prediction result of the joint data prediction based on the online prediction results at the data user end and the online prediction results at the target data provider end.

[0141] In one embodiment, the target data resource is specified in the data usage contract to be created; the authorization request operation is a contract creation and review operation for the data usage contract. The model deployment trigger operation is a contract deployment trigger operation for the data usage contract.

[0142] In one embodiment, such as Figure 13 As shown, after the business model training is completed, resulting in a virtual shared business model, and the model is published, the data user can display the model deployment configuration interface. This interface displays information such as contract name, project to which the contract belongs, contract type, remaining call count, contract expiration time, and contract creation time. The data user can click "Deploy" in the model deployment configuration interface to trigger the deployment configuration of the virtual shared business model. The deployed and configured virtual shared business model is used to perform joint prediction on the online data from the data user and the target online data corresponding to the target data resources, obtaining online prediction results for use by the data user in business processing.

[0143] In one embodiment, such as Figure 14 As shown, after the model deployment configuration is completed, the data user can display another model deployment configuration interface. From this interface, the data user can obtain a key and a data interface document. Based on the key and the data interface document, the data user can call the virtual shared business model to perform joint data prediction on the online data of the data user and the target online data corresponding to the target data resource, and obtain online prediction results for use by the data user in business processing.

[0144] In the above embodiments, the model deployment trigger operation triggers the model deployment configuration for the virtual shared business model, so that the virtual shared business model can be applied online to perform joint data prediction on the online data of the data user and the target online data corresponding to the target data resource, and obtain online prediction results for the data user to use in business processing. This solves the problems of data silos and platform silos and improves the accuracy of business processing at the data user.

[0145] In one embodiment, such as Figure 15 As shown, a cross-platform data processing method is provided. This embodiment applies this method to... Figure 1 Taking data provider 104 as an example, the following steps are included:

[0146] Step 1502: The data provider receives a data resource authorization request sent by the data user; the data resource authorization request carries target data resource description information; the target data resource description information is selected from at least one data resource description information displayed on the data user; at least one data resource description information is used to describe the data resources published by the data provider to the converged computing platform; the converged computing platform is a standardized platform compatible with at least two privacy computing components; each privacy computing component has a corresponding computing component interface.

[0147] Among them, the data resource authorization request is a request used to apply for authorization to use data resources.

[0148] Specifically, the data user can generate a data resource authorization request carrying the target data resource description information. The data user can then send the data resource authorization request to the converged computing platform using a computing component interface that matches the data user's local privacy computing component. The converged computing platform can then send the data resource authorization request to the data provider using a computing component interface that matches the data provider's local privacy computing component. The data provider can then receive the data resource authorization request.

[0149] Step 1504: In response to the data resource authorization request, display the authorization review interface for the target data resource; the target data resource is the data resource described in the target data resource description information.

[0150] The authorization review interface is located on the data provider's side and is used to review the usage permissions for the target data resources.

[0151] Specifically, the data provider can trigger a data resource authorization request, and the data provider can respond to the data resource authorization request by displaying an authorization review interface for the target data resource.

[0152] In one embodiment, the target data resource is specified in the data usage contract to be created; the authorization application operation corresponding to the data resource authorization request is a contract creation and review operation for the data usage contract.

[0153] In one embodiment, such as Figure 16 As shown, the authorization review interface on the data provider's end displays basic information about the contract to be reviewed, the contract creation review status, and contract review comments. Basic contract information may include the contract name, contract type, contract creation time, expected contract usage period, expected number of contract calls, and remarks regarding the contract application scenario. The data provider can review contract creation based on the authorization review interface. After approving the contract creation application, clicking "Configure Data Resources" on the authorization review interface triggers the mapping configuration of relevant data resources.

[0154] Step 1506: In response to the authorization operation for the target data resource in the authorization review interface, display the resource configuration interface for the target data resource.

[0155] The authorization operation is used to grant usage rights to the target data resource. The resource configuration interface is the interface available to the data user for configuring mapping operations on the target data resource. The mapping configuration operation establishes a mapping relationship between the target data resource description information and the target data resource on the local machine of the data provider.

[0156] Specifically, the data provider can trigger an authorization operation for the target data resource in the authorization review interface, and the data provider can respond to the authorization operation for the target data resource in the authorization review interface by displaying the resource configuration interface for the target data resource.

[0157] Step 1508: In response to the mapping configuration operation for the target data resource in the resource configuration interface, establish a mapping relationship between the target data resource description information and the target data resource on the local side of the data provider; wherein, after the mapping relationship is established, it is used to instruct the joint training of the business model based on the user sample data and the target data resource on the data user side. When communication interaction is involved in the training process, the data provider and the data user side communicate with the fusion computing platform based on the computing component interface that matches them respectively.

[0158] Specifically, the data provider can configure mapping for target data resources in the resource configuration interface. In response to this configuration, the data provider establishes a mapping relationship between the target data resource description information and the target data resource on its local machine. This mapping relationship signifies that the data provider has agreed to allow the corresponding data user to use the data provider's data resources for joint data computation.

[0159] In one embodiment, such as Figure 17 As shown, after agreeing to the contract creation request, click... Figure 16 The "Configure Data Resources" section of the authorization and review interface can trigger the display of the resource configuration interface on the data provider's end, and perform mapping configuration on the relevant data resources based on the resource configuration interface on the data provider's end. Specifically, the resource configuration interface can display the name of the data resource description information, the domain to which the data resource belongs, and a detailed description of the data resource. The mapping configuration can include limiting the way the data resource is accessed to the converged computing platform, establishing a mapping relationship between the target data resource description information and the target data resource on the data provider's local end, editing the data resource's introduction information, and configuring the fields of the data resource.

[0160] In the aforementioned cross-platform data processing method, the data provider receives a data resource authorization request from the data user. This request carries target data resource description information, selected from at least one data resource description displayed on the data user. This description describes the data resources published by the data provider to the converged computing platform, a standardized platform compatible with at least two privacy-preserving computing components, each with its own interface. Responding to the authorization request, an authorization review interface for the target data resource is displayed. The target data resource is the one described in the data resource description. Responding to the authorization operation on the review interface, a resource configuration interface for the target data resource is displayed. Responding to the mapping configuration operation on the resource configuration interface, a mapping relationship is established between the target data resource description and the target data resource locally on the data provider. Once the mapping relationship is established, it is used to instruct the joint training of the business model based on the user's sample data and the target data resources. During training, when communication interaction is involved, the data provider and the data user can communicate with the fusion computing platform based on their respective matching computing component interfaces. Compared to traditional data processing methods, this application designs a fusion computing platform compatible with at least two privacy computing components. During data interaction, the data user and the target data provider can communicate with the fusion computing platform based on their respective matching computing component interfaces. This enables cross-computation of data between platforms built on the same privacy computing component, as well as cross-computation between platforms built on different privacy computing components, thus meeting the needs of more business scenarios.

[0161] In one embodiment, before the data provider receives the data resource authorization request sent by the data user, the method further includes: generating data resource description information of the data resource to be published in response to a basic information configuration operation for the data resource to be published in the data provider; triggering the publication of the data resource to be published to the converged computing platform in response to a resource publication operation; and displaying the data resource description information of the published data resource on the data resource selection interface of the data user.

[0162] The basic information configuration operation is used to configure the basic information of the data resources to be published. The resource publishing operation is used to publish the data resources to be published to the converged computing platform.

[0163] Specifically, the data provider can trigger a basic information configuration operation for the data resources to be published on the data provider's end. In response to this configuration operation, the data provider can generate data resource description information for the data resources to be published. The data provider can also trigger a resource publishing operation, which in turn triggers the publication of the data resources to be published to the converged computing platform.

[0164] In one embodiment, in response to the configuration operation of basic information of data resources to be published in the data provider, at least one of the following information can be configured: information of the data provider of the data resource, information of the domain to which the data resource belongs, information of the open conditions of the data resource, information of the dataset description of the data resource, information of the format of the data resource, information of the type of the data resource, and information of the update frequency of the data resource.

[0165] In one embodiment, such as Figure 18 As shown, data providers can edit the basic information of the data resources to be published on the basic information configuration interface. This includes basic information such as data provider information, data resource description name, industry classification of the data resource, data resource update frequency, and the field to which the data resource belongs. After the basic information is edited, the data provider can generate the data resource description information of the data resource to be published and publish the data resource description information to the converged computing platform.

[0166] In the above embodiments, by obtaining the data resource description information of the data resource to be published and triggering the publication of the data resource to be published to the converged computing platform, the data user can select the corresponding data resource for joint data processing, which solves the problems of data silos and platform silos and improves the accuracy of business processing of the data user.

[0167] In one embodiment, the method further includes: in response to a model call configuration review request sent by a data user for a virtual shared business model, displaying a call configuration review interface; the call configuration review interface displays model call constraint information that is pending review and configured for the virtual shared business model; in response to a call configuration approval operation, authorizing the model call constraint information; the authorized model call constraint information is used to constrain the data user's calls to the virtual shared business model.

[0168] The "Model Call Configuration Review Request" is used to review the model call constraint information. The "Call Configuration Review Interface" is the interface on the data provider's side used to review the model call constraint information. The "Call Configuration Approve Operation" is used to authorize the model call constraint information.

[0169] Specifically, the data user can generate a model call configuration review request for the virtual shared business model and send it to the converged computing platform based on a computing component interface that matches the data user's local privacy computing component. The converged computing platform can then forward the model call configuration review request to the data provider based on a computing component interface that matches the data provider's local privacy computing component. The data provider can respond to the model call configuration review request by displaying a call configuration review interface. This interface displays model call constraint information configured for the virtual shared business model that is pending review. The data provider can trigger a call configuration consent operation, and the data provider can authorize the model call constraint information in response. The authorized model call constraint information is used to restrict the data user's calls to the virtual shared business model.

[0170] In one embodiment, the target data resource is specified in the data usage contract to be created; the authorization request operation corresponding to the data resource authorization request is a contract creation review operation for the data usage contract. The model call configuration review request corresponding to the model call configuration review operation is a contract release configuration review request for the data usage contract to be published.

[0171] In one embodiment, such as Figure 19 As shown, the call configuration review interface on the data provider side can display basic information about the contract to be published, data resource information, description information, and related algorithm information. The basic information of the contract to be published specifically includes the contract name, contract type, project name, and description. The data provider side can edit the call configuration review interface to review the contract to be published and to edit the review reasons for the contract to be published.

[0172] In the above embodiments, by invoking the configuration consent operation to authorize the model call constraint information, the data user's call to the virtual shared business model can be restricted, thereby further improving the security of the local data of the data provider.

[0173] In one embodiment, such as Figure 20 As shown, a cross-platform data processing method is provided. This embodiment applies this method to management. Figure 1 Taking the platform operation side of the Zhongrong Computing Platform 106 as an example, the following steps are included:

[0174] Step 2002: In response to the data resource publishing request sent by the data provider, the data resource description information provided by the data provider is displayed on the resource publishing review interface of the platform operation side.

[0175] The data resource publishing request is used to publish the data resources of the data provider to the converged computing platform. The platform operation terminal is the terminal responsible for operating the converged computing platform. The resource publishing review interface is used to review the data resources provided by the data provider.

[0176] Specifically, the data provider can generate a data resource publishing request and send it to the converged computing platform through a computing component interface that matches the data provider's local privacy computing component. The platform operator responsible for operating the converged computing platform can respond to the data resource publishing request sent by the data provider and display the data resource description information provided by the data provider on the resource publishing review interface.

[0177] Step 2004: Based on the resource approval operation applied to the displayed data resource description information, the approved data resource is triggered to be published to the converged computing platform. The converged computing platform is a standardized platform compatible with at least two privacy computing components. Each privacy computing component has a corresponding computing component interface. The published data resource description information is displayed on the data resource selection interface of the data user, so that the data user can select the target data resource description information from the displayed data resource description information. It is also indicated that after the target data resource mapped by the target data resource description information is authorized for use, the user's sample data and the target data resource are used to jointly train the business model. In the case of communication interaction during the training process, the data provider and the data user communicate with the converged computing platform based on their respective matching computing component interfaces.

[0178] The resource approval process is used to authorize the release of data resources from the data provider.

[0179] Specifically, the platform operator can trigger a resource approval operation on the displayed data resource description information. Based on this operation, the platform operator can then publish the approved data resource to the converged computing platform. It can be understood that the data resource description information corresponding to the data resource published to the converged computing platform can be displayed on the data resource selection interface belonging to the data user, allowing the data user to select the data resource.

[0180] In one embodiment, the target data resource is specified in the data usage contract to be created; the authorization request operation for the target data resource is a contract creation and review operation for the data usage contract.

[0181] In one embodiment, such as Figure 21As shown, the resource release review interface on the platform's operation side displays the number of data resources awaiting review, the number of published data resources, the total number of data resources, the number of published data, and a detailed list of release review information for each data resource. The detailed list of data resource release review information may include information such as the data resource name, the data resource's domain, release status, review status, release time, and data resource update time. After the data provider releases the data resources to the converged computing platform, the platform can review the relevant data resources based on the resource release review interface. Once approved, the data resource description information can be displayed on the data resource selection interface on the data user's end.

[0182] In the aforementioned cross-platform data processing method, in response to a data resource publishing request sent by the data provider, the data resource description information of the data resource provided by the data provider is displayed on the resource publishing review interface of the platform operator. Based on the resource review approval operation applied to the displayed data resource description information, the approved data resource is triggered to be published to the converged computing platform. The converged computing platform is a standardized platform compatible with at least two privacy computing components; each privacy computing component has a corresponding computing component interface. The published data resource description information is displayed on the data resource selection interface of the data user, allowing the data user to select a target data resource description from the displayed information. It also instructs the user to use the user's sample data and the target data resource to jointly train a business model after the target data resource description is authorized for use. During training, when communication interaction is involved, the data provider and the data user can communicate with the converged computing platform based on their respective matching computing component interfaces. Compared to traditional data processing methods, this application designs a converged computing platform compatible with at least two privacy computing components. During data interaction, the data user and the target data provider can communicate with the converged computing platform based on their respective matching computing component interfaces. This enables cross-computation of data between platforms built on the same privacy computing component, as well as cross-computation of data between platforms built on different privacy computing components, thus meeting the needs of more business scenarios.

[0183] like Figure 22 As shown, in one embodiment, a cross-platform data processing method is provided. This embodiment applies the method to... Figure 1 Taking terminal 102 as an example, this method specifically includes the following steps:

[0184] Step 2202: Display at least one data resource description on the data resource selection interface of the data user; the data resource description is used to describe the data resource published by the data provider to the converged computing platform; the converged computing platform is a standardized platform compatible with at least two privacy computing components; each privacy computing component has a corresponding computing component interface; the data user matches the first computing component interface; the target data provider matches the second computing component interface.

[0185] Step 2204: In response to the data resource selection operation, determine the selected target data resource description information from at least one data resource description information; the target data resource mapped to by the target data resource description information is provided by the target data provider; the target data resource includes the provider sample data provided by the target data provider; the same business model to be trained is present in the local storage of both the data user and the target data provider.

[0186] Step 2206: In response to the authorization request operation, authorization is triggered for the target data resource.

[0187] Step 2208: After the target data resource is authorized for use, in response to the model configuration operation for the business model to be configured, the model training configuration is performed on the business model to be configured, and the configured business model to be trained and the sample-related information of the configured business model to be trained are obtained.

[0188] Step 2210: Respond to the model training trigger operation, train the business model in the user terminal based on the user terminal sample data corresponding to the sample information, and use the first public key issued by the fusion computing platform based on the first computing component interface to encrypt the intermediate results in the model training to obtain the first intermediate feature.

[0189] Step 2212: The first intermediate feature is sent to the fusion computing platform based on the first computing component interface, so that the fusion computing platform forwards the first intermediate feature to the target data provider based on the second computing component interface, so as to instruct the target data provider to train the business model in the target data provider based on the first intermediate feature and the second intermediate feature; the second intermediate feature is obtained by the target data provider encrypting the intermediate result generated when training the business model using the target data resources corresponding to the sample information based on the second public key issued by the fusion computing platform.

[0190] Step 2214: Receive the second intermediate feature sent by the fusion computing platform based on the first computing component interface, and continue to train the business model in the data user terminal based on the second intermediate feature and the first intermediate feature to obtain the encryption adjustment parameters.

[0191] Step 2216: Upload the encrypted adjustment parameters to the fused computing platform based on the interface of the first computing component and obtain the adjustment parameters after decryption.

[0192] Step 2218: Update the business model in the data user end based on the adjusted parameters, and obtain the first business model of the data user end after training is completed; the first business model and the second business model trained by the target data provider end constitute a jointly trained virtual shared business model.

[0193] Step 2220: After the business model training is completed and the virtual shared business model is obtained, the model call configuration interface of the data user end is displayed.

[0194] Step 2222: In response to the model call configuration operation on the model call configuration interface, obtain the model call constraint information configured for the virtual shared business model; the model call constraint information is used to constrain the data user's call to the virtual shared business model.

[0195] Step 2224 displays the model deployment configuration interface on the data user end.

[0196] Step 2226: In response to the model deployment trigger operation on the model deployment configuration interface, trigger the model deployment configuration for the virtual shared business model; the deployed and configured virtual shared business model is used to perform joint data prediction on the online data of the data user and the target online data corresponding to the target data resource to obtain the online prediction result used by the data user in business processing; the target online data is the online data provided by the target data provider.

[0197] In one embodiment, such as Figure 23 As shown, the converged computing platform of this application can support access to various platforms built on various privacy computing components. From Figure 23 As can be seen from this, the data resources accessed by the converged computing platform can include the data resources of various platforms (i.e., Platform 1 to Platform 6) participating in cross-data computation. After the data resources of each platform undergo corresponding privacy and public computation through the converged computing platform, they can be used by various services (i.e., Services 1 to Services 7) to meet various application scenarios (i.e., Applications 1 to Applications 4).

[0198] In one embodiment, such as Figure 24As shown, the overall architecture of the converged computing platform can be divided into five layers: cloud and infrastructure layer, data management layer, trusted computing and evidence storage layer, operations management layer, and application layer. The cloud and infrastructure layer mainly includes various cloud infrastructures, such as cloud servers, networks, storage, cluster management, and automated operations and maintenance. The data management layer mainly includes data access management and data resource management. The trusted computing and evidence storage layer mainly includes a management control center, privacy computing components, a blockchain control platform, and a blockchain engine. The management control center is mainly responsible for task management, logic configuration, process scheduling, policy management, and performance control. The privacy computing components mainly include multi-party secure computing components and federated learning algorithm components. The blockchain control platform is mainly responsible for consortium management, unified interfaces, contract management, chain management, and network management. The blockchain engine mainly includes various popular blockchain engines. The operations management layer can include a reporting center, comprehensive management, contract management, algorithm pool management, log evidence storage, and operations and maintenance management. The reporting center mainly includes billing rules, a transaction center, and a billing center. Comprehensive management mainly includes qualification management, audit management, and routing management. Algorithm pool management mainly includes algorithm demonstration, algorithm release, and algorithm operation. Log storage mainly includes storage recording, authentication, and log management. Operation and maintenance management mainly includes system management, user management, system monitoring, and interface engine. The application layer mainly includes applications from three roles: data users, data providers, and platform operators.

[0199] In one embodiment, such as Figure 25 As shown, the architecture of the management and control center of the converged computing platform mainly includes model application scheduling and privacy computing control. Model application scheduling mainly includes application management, application configuration logic, processing logic configuration, conditional logic configuration, scheduling node configuration, process scheduling configuration, multi-policy management configuration, time templates, and system configuration. Privacy computing control mainly includes performance monitoring, statistical analysis, and privacy computing component integration. Among them, performance monitoring mainly includes engine alerts, routing alerts, policy alerts, alert queries, visualization, and monitoring configuration. Statistical analysis mainly includes statistical configuration, viewing statistical results, querying statistical results, and exporting statistical results. Privacy computing component integration mainly includes task integration, data resource integration, computing resource integration, task structure integration, business monitoring integration, and statistical report integration.

[0200] In one embodiment, such as Figure 26As shown, data providers can publish data resources to the converged computing platform based on data resource description information and the data access configuration function of the privacy computing component. After the data resources are published to the converged computing platform, data users can see the data resource description information on their data user terminals. Then, data users can edit contract information and select data resources to create a data usage contract. After a data user initiates a contract creation application, the data provider and the platform operator review the application. Once the data user contract is approved by both parties, the data is considered integrated, and the data user has the right to use the relevant data resources provided by the data provider. Furthermore, the data user can configure the data usage contract based on the relevant data resources. During the configuration process, the training of the business model can be redirected to the corresponding privacy computing components on the data provider and data user terminals, such as federated learning algorithm components and multi-party secure computing components, to perform joint model training and obtain a trained business model, i.e., a virtual shared business model. Furthermore, data users can configure the data usage contract for publication, including configuring model call constraints such as the number of times the contract can be called and its usage period. After configuring the model call constraints, data users can initiate a request to publish the data usage contract. Data providers and platform operators can respectively review the published contracts. If both the data provider and platform operator approve the submission, the data user can manage the deployment of the published contract. Contract deployment can be implemented through the algorithmic logic of the platform management control center. If the contract configuration is complete, the data user can invoke the deployed contract externally, allowing the platform management control center to execute the data usage contract based on the corresponding privacy computing components. This involves calling the virtual shared business model to perform joint prediction on the online data on the data user's local end and the online data on the data provider's local end, obtaining online prediction results for the data user to use in business processing. If the contract call no longer meets the model call constraints—that is, the number of contract calls is exhausted or the contract call period has expired—the platform management control center of the converged computing platform will reject the execution of the contract.

[0201] In one embodiment, such as Figure 27As shown, the data resource description information management process of the converged computing platform mainly includes data resource publishing, data resource modification, and data resource delisting. For data resource publishing, data providers can register on the converged computing platform using their data provider portal. After successful registration and login, the data provider can submit review materials, which the platform operator can then review. If the review is approved, the data user can become a partner of the converged computing platform. After becoming a partner, the data provider can send a data access application to the platform operator, providing sample data and data interface documentation based on the data resource description information. The platform operator can review the data access application; after approval, the data resource corresponding to the data resource description information can be published on the converged computing platform for data users to view and use. For data resource modification, the data provider can send a data modification application to the platform operator, providing revised sample data and data interface documentation based on the revised data resource description information. The platform operator can review data change requests. Once approved, the data resources corresponding to the updated data resource description information can be published on the converged computing platform to remind data users to retrain their models based on the updated data resource description information. For data resource removal, data providers can send a data removal request to the platform operator. The platform operator can review the data removal request. Once approved, the converged computing platform can delete the published data resource description information and the corresponding computing component interface. After the computing component interface is deleted, the platform will notify the data user that the data resource has been updated to remind them to retrain their models based on the updated data resource description information.

[0202] In one embodiment, such as Figure 28 As shown, the data architecture of the converged computing platform includes resource data from various platforms (i.e., Platform 1 to Platform 5) connected to the converged computing platform. After the data resources of each platform undergo corresponding privacy and public computation through the converged computing platform, they can be used by various types of services (i.e., Business Category 1 to Business Category 3) (i.e., Business 1 to Business 4), thereby meeting various business scenarios.

[0203] In one embodiment, the converged computing platform can support resource data access from platforms built on various privacy-preserving computing components. The converged computing platform can be deployed to each accessing platform through corresponding business service clients. For example, ... Figure 29As shown, the converged computing platform can be deployed on the accessed platform A. Platform A's business network is separated from the internet by a secure isolation zone established by a firewall, which includes a bootstrap server and an access server. Business service clients can access Platform A's business network via the internet, through the bootstrap server, and connect to the converged computing platform and Platform A's business platform via a business gateway. Platform A's business platform includes Platform A's business network and Platform A's core intranet. Platform A's business network includes a privacy computing service cluster, an operations management center, a management control center, a blockchain service platform, privacy computing access, email and middleware services, and a business gateway. The privacy computing service cluster includes a federated learning computing service cluster and a multi-party secure computing service cluster. Platform A's core intranet includes a business gateway and intranet storage services.

[0204] This application also provides an application scenario in which the aforementioned cross-platform data processing method is applied. Specifically, this cross-platform data processing method can be applied to scenarios involving cross-computation of data between heterogeneous platforms, i.e., the privacy computing component of the platform corresponding to the data user is different from the privacy computing component of the platform corresponding to the data provider. The data user can display at least one data resource description on the data resource selection interface; the data resource description is used to describe the data resource published by the data provider to the converged computing platform; the converged computing platform is a standardized platform compatible with at least two privacy computing components; each privacy computing component has a corresponding computing component interface; the data user matches the first computing component interface of the first privacy computing component; the target data provider matches the second computing component interface of the second privacy computing component, wherein the first privacy computing component and the second privacy computing component are two different privacy computing components. In response to a data resource selection operation, a target data resource description is determined from at least one data resource description. The target data resource mapped to by the target data resource description is provided by a target data provider. The target data resource includes sample data provided by the target data provider. Both the data user and the target data provider have the same business model to be trained locally. In response to an authorization request operation, authorization is triggered for the target data resource.

[0205] After the target data resource is authorized for use, the data user can respond to the model configuration operation for the business model to be configured, perform model training configuration for the business model to be configured, and obtain the configured business model to be trained and the sample-related information of the configured business model to be trained. Responding to the model training trigger operation, the data user trains the business model in the data user based on the user's sample data corresponding to the sample-related information. Using the first public key issued by the fusion computing platform based on the first computing component interface, the intermediate results in the model training are encrypted to obtain the first intermediate feature. The first intermediate feature is sent to the fusion computing platform based on the first computing component interface, causing the fusion computing platform to forward the first intermediate feature to the target data provider based on the second computing component interface, instructing the target data provider to train the business model in the target data provider based on the first and second intermediate features. The second intermediate feature is obtained by the target data provider encrypting the intermediate results generated when training the business model using the target data resource corresponding to the sample-related information based on the second public key issued by the fusion computing platform. The data user receives the second intermediate feature sent by the fusion computing platform based on the first computing component interface, and continues to train the business model in the data user based on the second and first intermediate features to obtain encrypted adjustment parameters. The encrypted adjustment parameters are uploaded to the converged computing platform via the interface of the first computing component, and the decrypted adjustment parameters are obtained. Based on the adjustment parameters, the business model in the data user end is updated, and after training, the first business model of the data user end is obtained. The first business model and the second business model trained by the target data provider end constitute a jointly trained virtual shared business model.

[0206] After the business model training is completed and a virtual shared business model is obtained, the data user terminal can display the model call configuration interface. In response to the model call configuration operation on the interface, the model call constraint information configured for the virtual shared business model is obtained; this constraint information is used to constrain the data user terminal's calls to the virtual shared business model.

[0207] The data user terminal can display its model deployment configuration interface. Responding to a model deployment trigger operation on the interface, it initiates model deployment configuration for the virtual shared business model. The deployed virtual shared business model is used to perform joint prediction on the online data of the data user terminal and the target online data corresponding to the target data resource, obtaining online prediction results for use by the data user terminal in business processing. The target online data is the online data provided by the target data provider. This enables cross-computation of data between various platforms built on different privacy computing components, meeting the needs of more business scenarios.

[0208] This application also provides another application scenario where the aforementioned cross-platform data processing method is applied. Specifically, this cross-platform data processing method can be applied to scenarios involving cross-computation of data between homogeneous platforms, i.e., the privacy computing component of the platform corresponding to the data user is the same as the privacy computing component of the platform corresponding to the data provider. Using the cross-platform data processing method of this application, cross-computation of data is achieved between various platforms that support the construction of the same privacy computing component by integrating the interfaces of the same computing component in the computing platform that match the privacy computing components corresponding to both the data user and the data provider.

[0209] It should be understood that although the steps in the flowcharts of the above embodiments are shown sequentially, these steps are not necessarily executed in that order. Unless otherwise expressly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some of the steps in the above embodiments may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least a portion of the sub-steps or stages of other steps.

[0210] In one embodiment, such as Figure 30 As shown, a cross-platform data processing device 3000 is provided. This device can be a software module, a hardware module, or a combination of both, integrated into a computer device. Specifically, the device includes:

[0211] The first display module 3002 is used to display at least one data resource description information on the data resource selection interface of the data user end; the data resource description information is used to describe the data resources published by the data provider end to the converged computing platform; the converged computing platform is a standardized platform compatible with at least two privacy computing components; each privacy computing component has a corresponding computing component interface;

[0212] The determination module 3004 is used to determine the selected target data resource description information from at least one data resource description information in response to the data resource selection operation; the target data resource mapped to by the target data resource description information is provided by the target data provider.

[0213] The first trigger module 3004 is used to trigger authorization for the target data resource in response to the authorization request operation. After the target data resource is authorized for use, it is used to instruct the user's sample data and the target data resource to jointly train the business model. When communication interaction is involved in the training process, the data user and the target data provider communicate with the fusion computing platform based on the computing component interface that matches them respectively.

[0214] In one embodiment, after the target data resource is authorized for use, the first triggering module 3004 is further configured to respond to a model training triggering operation to trigger the data user to jointly train the business model to be trained based on the user's sample data and the target data provider's use of the target data resource.

[0215] In one embodiment, the data user end is matched with the first computing component interface; the target data provider end is matched with the second computing component interface; both the data user end and the target data provider end have the same business model to be trained locally; the first trigger module 3004 is further configured to respond to the model training trigger operation, train the business model in the data user end based on the sample data of the user end, encrypt the intermediate results in the model training using the first public key issued by the fusion computing platform based on the first computing component interface, and obtain the first intermediate feature; send the first intermediate feature to the fusion computing platform based on the first computing component interface, so that the fusion computing platform forwards the first intermediate feature to the target data provider end based on the second computing component interface, so as to instruct the target data provider end to train the target data based on the first intermediate feature and the second intermediate feature. The data provider provides a business model; the second intermediate feature is obtained by encrypting the intermediate results generated when training the business model using the target data resources based on the second public key issued by the converged computing platform; the data provider receives the second intermediate feature sent by the converged computing platform through the first computing component interface, and continues to train the business model in the data user end based on the second intermediate feature and the first intermediate feature to obtain encrypted adjustment parameters; the data provider uploads the encrypted adjustment parameters to the converged computing platform through the first computing component interface and obtains the adjustment parameters after decryption; the data user end updates the business model based on the adjustment parameters, and obtains the first business model of the data user end after training is completed; the first business model and the second business model trained by the target data provider constitute a jointly trained virtual shared business model.

[0216] In one embodiment, the first display module 3002 is further configured to display component identifiers of at least two privacy computing components on the privacy computing component selection interface; in response to the component selection operation, determine the selected target component identifier; the first trigger module 3004 is further configured to trigger authorization for the target privacy computing component and target data resource corresponding to the target component identifier in response to the authorization request operation; wherein, after the target privacy computing component is authorized for use, it is used to instruct the data user to communicate with the converged computing platform through the corresponding computing component interface of the target privacy computing component.

[0217] In one embodiment, the target component identifier and the target data resource are specified in the data usage contract to be created; the authorization application operation is a contract creation review operation for the data usage contract; the first trigger module 3004 is also used to trigger the converged computing platform and the data provider to perform creation review processing on the data usage contract respectively in response to the contract creation review operation; the target privacy computing component and the target data resource specified in the approved data usage contract are also authorized through use.

[0218] In one embodiment, the target data resource includes provider sample data provided by the target data provider; the first triggering module 3004 is further configured to, after the target data resource is authorized for use, in response to a model configuration operation for the business model to be configured, perform model training configuration on the business model to be configured to obtain the configured business model to be trained and the sample-related information of the configured business model to be trained; in response to a model training triggering operation for the business model to be trained, trigger joint training of the business model to be trained based on the user-side sample data and provider-side sample data corresponding to the sample-related information.

[0219] In one embodiment, the first display module 3002 is further configured to display the model call configuration interface of the data user terminal after the business model training is completed and the virtual shared business model is obtained; in response to the model call configuration operation on the model call configuration interface, obtain the model call constraint information configured for the virtual shared business model; the model call constraint information is used to constrain the data user terminal's call to the virtual shared business model.

[0220] In one embodiment, the first display module 3002 is further configured to display the model deployment configuration interface of the data user end after the business model training is completed and a virtual shared business model is obtained; in response to the model deployment trigger operation on the model deployment configuration interface, trigger the model deployment configuration for the virtual shared business model; the deployed and configured virtual shared business model is used to perform joint data prediction on the online data of the data user end and the target online data corresponding to the target data resource to obtain the online prediction result used by the data user end in business processing; the target online data is the online data provided by the target data provider end.

[0221] The aforementioned cross-platform data processing device displays at least one data resource description on the data resource selection interface of the data user. This description describes the data resource published by the data provider to the converged computing platform, a standardized platform compatible with at least two privacy-preserving computing components, each with a corresponding computing component interface. In response to a data resource selection operation, a selected target data resource description can be determined from the at least one description. The target data resource mapped to by the target data resource description is provided by the target data provider. In response to an authorization request operation, authorization for the target data resource can be triggered. Once authorized, the target data resource can be used to instruct the joint training of a business model based on the user's sample data and the target data resource. During training, if communication interaction is involved, the data user and the target data provider can communicate with the converged computing platform respectively based on their respective matching computing component interfaces. Compared to traditional data processing methods, this application designs a converged computing platform compatible with at least two privacy computing components. During data interaction, the data user and the target data provider can communicate with the converged computing platform based on their respective matching computing component interfaces. This enables cross-computation of data between platforms built on the same privacy computing component, as well as cross-computation of data between platforms built on different privacy computing components, thus meeting the needs of more business scenarios.

[0222] In one embodiment, such as Figure 31 As shown, a cross-platform data processing device 3100 is provided. This device can be a software module, a hardware module, or a combination of both, integrated into a computer device. Specifically, the device includes:

[0223] The receiving module 3102 is used for the data provider to receive a data resource authorization request sent by the data user; the data resource authorization request carries target data resource description information; the target data resource description information is selected from at least one data resource description information displayed by the data user; the at least one data resource description information is used to describe the data resources published by the data provider to the converged computing platform; the converged computing platform is a standardized platform compatible with at least two privacy computing components; each privacy computing component has a corresponding computing component interface;

[0224] The second display module 3104 is used to display an authorization review interface for a target data resource in response to a data resource authorization request; the target data resource is the data resource described by the target data resource description information; and to display a resource configuration interface for the target data resource in response to an authorization operation for the target data resource in the authorization review interface.

[0225] A module 3106 is established to respond to a mapping configuration operation for a target data resource in the resource configuration interface, establishing a mapping relationship between the target data resource description information and the target data resource locally on the data provider end. After the mapping relationship is established, it instructs the user-end sample data and the target data resource to jointly train the business model. When communication interaction is involved during training, the data provider end and the data user end communicate with the fusion computing platform based on their respective matching computing component interfaces. Fourthly, this application provides a cross-platform data processing apparatus, comprising:

[0226] In one embodiment, the data processing device 3100 further includes: a generation module, configured to generate data resource description information of the data resource to be published in response to a basic information configuration operation for the data resource to be published in the data provider; to trigger the publication of the data resource to be published to the converged computing platform in response to a resource publication operation; and to display the data resource description information of the published data resource on the data resource selection interface of the data user.

[0227] In one embodiment, the second display module 3104 is further configured to display a call configuration review interface in response to a model call configuration review request sent by the data user terminal for the virtual shared business model; the call configuration review interface displays model call constraint information that is pending review and configured for the virtual shared business model; in response to the call configuration approval operation, authorize the model call constraint information; the authorized model call constraint information is used to constrain the data user terminal's call to the virtual shared business model.

[0228] The aforementioned cross-platform data processing device receives a data resource authorization request from a data user via a data provider. This request carries a target data resource description, selected from at least one data resource description displayed on the data user. This description describes the data resource published by the data provider to the converged computing platform, a standardized platform compatible with at least two privacy-preserving computing components, each with its own interface. Responding to the authorization request, an authorization review interface for the target data resource (described in the data resource description) can be displayed. Responding to the authorization operation on the review interface, a resource configuration interface for the target data resource can be displayed. Responding to the mapping configuration operation on the resource configuration interface, a mapping relationship between the target data resource description and the target data resource locally on the data provider can be established. Once the mapping relationship is established, it is used to instruct the joint training of the business model based on the user's sample data and the target data resources. During training, when communication interaction is involved, the data provider and the data user can communicate with the fusion computing platform based on their respective matching computing component interfaces. Compared to traditional data processing methods, this application designs a fusion computing platform compatible with at least two privacy computing components. During data interaction, the data user and the target data provider can communicate with the fusion computing platform based on their respective matching computing component interfaces. This enables cross-computation of data between platforms built on the same privacy computing component, as well as cross-computation between platforms built on different privacy computing components, thus meeting the needs of more business scenarios.

[0229] In one embodiment, such as Figure 32 As shown, a cross-platform data processing device 3200 is provided. This device can be a software module, a hardware module, or a combination of both, integrated into a computer device. Specifically, the device includes:

[0230] The third display module 3202 is used to respond to the data resource publishing request sent by the data provider and display the data resource description information provided by the data provider on the resource publishing review interface of the platform operation end.

[0231] The second trigger module 3204 is used to trigger the publication of the approved data resources to the converged computing platform based on the resource approval operation applied to the displayed data resource description information; the converged computing platform is a standardized platform compatible with at least two privacy computing components; each privacy computing component has a corresponding computing component interface;

[0232] The data resource description information of the published data resources is displayed on the data resource selection interface of the data user terminal, so that the data user terminal can select the target data resource description information from the displayed data resource description information. It also instructs that after the target data resource mapped to by the target data resource description information is authorized for use, the user terminal sample data and the target data resource are used to jointly train the business model. When communication interaction is involved in the training process, the data provider and the data user terminal communicate with the fusion computing platform based on the computing component interface that matches them respectively.

[0233] The aforementioned cross-platform data processing device, in response to a data resource publishing request sent by the data provider, displays the data resource description information of the data resource provided by the data provider on the resource publishing review interface of the platform operator. Based on the resource review approval operation applied to the displayed data resource description information, the approved data resource is triggered to be published to the converged computing platform. The converged computing platform is a standardized platform compatible with at least two privacy computing components; each privacy computing component has a corresponding computing component interface. The published data resource description information is displayed on the data resource selection interface of the data user, allowing the data user to select a target data resource description from the displayed information. It also instructs the user to use the user's sample data and the target data resource to jointly train a business model after the target data resource description is authorized for use. During training, when communication interaction is involved, the data provider and the data user can communicate with the converged computing platform based on their respective matching computing component interfaces. Compared to traditional data processing methods, this application designs a converged computing platform compatible with at least two privacy computing components. During data interaction, the data user and the target data provider can communicate with the converged computing platform based on their respective matching computing component interfaces. This enables cross-computation of data between platforms built on the same privacy computing component, as well as cross-computation of data between platforms built on different privacy computing components, thus meeting the needs of more business scenarios.

[0234] Each module in the aforementioned cross-platform data processing device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the memory of a computer device as software, so that the processor can call and execute the operations corresponding to each module.

[0235] In one embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as follows: Figure 33 As shown, the computer device includes a processor, memory, input / output interfaces, a communication interface, a display unit, and an input device. The processor, memory, and input / output interfaces are connected via a system bus, and the communication interface, display unit, and input device are also connected to the system bus via the input / output interfaces. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The input / output interfaces are used for exchanging information between the processor and external devices. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, mobile cellular networks, NFC (Near Field Communication), or other technologies. When the computer program is executed by the processor, it implements a cross-platform data processing method. The display unit of the computer device is used to form a visually visible image. It can be a display screen, a projection device, or a virtual reality imaging device. The display screen can be an LCD screen or an e-ink screen. The input device of the computer device can be a touch layer covering the display screen, or buttons, trackballs, or touchpads set on the casing of the computer device, or external keyboards, touchpads, or mice, etc.

[0236] Those skilled in the art will understand that Figure 33 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0237] In one embodiment, a computer device is also provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above method embodiments.

[0238] In one embodiment, a computer-readable storage medium is provided storing a computer program that, when executed by a processor, implements the steps in the above method embodiments.

[0239] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above method embodiments.

[0240] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data shall comply with the relevant laws, regulations and standards of the relevant countries and regions.

[0241] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the methods described above. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, or optical storage, etc. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc.

[0242] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0243] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the invention patent. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this patent application should be determined by the appended claims.

Claims

1. A cross-platform data processing method, characterized in that, The method includes: At least one data resource description is displayed on the data resource selection interface at the data user end; the data resource description is used to describe the data resource published by the data provider to the converged computing platform; the converged computing platform is a standardized platform compatible with at least two privacy computing components; each privacy computing component has a corresponding computing component interface; In response to a data resource selection operation, a selected target data resource description is determined from the at least one data resource description information; the target data resource mapped to by the target data resource description information is provided by a target data provider. In response to an authorization request operation, authorization is triggered for the target data resource; wherein, after the target data resource is authorized for use, it is used to instruct the training of a business model in the data user based on the user's sample data, using the first public key issued by the fusion computing platform based on the first computing component interface matched with the data user to encrypt the intermediate results in the model training, obtaining a first intermediate feature; the first intermediate feature is sent to the fusion computing platform based on the first computing component interface, so that the fusion computing platform forwards the first intermediate feature to the target data provider based on the second computing component interface matched with the target data provider, to instruct the target data provider to train the business model in the target data provider based on the first intermediate feature and the second intermediate feature; the second intermediate feature, The target data provider encrypts the intermediate results generated when training the business model using the target data resources based on the second public key issued by the converged computing platform. It receives the second intermediate feature sent by the converged computing platform through the first computing component interface, and continues to train the business model in the data user based on the second and first intermediate features to obtain encrypted adjustment parameters. It uploads the encrypted adjustment parameters to the converged computing platform through the first computing component interface and obtains the decrypted adjustment parameters. It updates the business model in the data user based on the adjustment parameters, and obtains the first business model of the data user after training is complete. The first business model and the second business model trained by the target data provider constitute a jointly trained virtual shared business model.

2. The method according to claim 1, characterized in that, Once the target data resource is authorized for use, it is used to indicate a response to a model training trigger operation to trigger the training of a business model in the data user terminal based on the user terminal sample data. The intermediate results in the model training are encrypted using the first public key issued by the fusion computing platform based on the first computing component interface matched with the data user terminal to obtain a first intermediate feature.

3. The method according to claim 1, characterized in that, The method further includes: Display component identifiers for at least two privacy computing components on the privacy computing component selection interface; In response to the component selection operation, determine the identifier of the selected target component; The response to the authorization request operation, triggering authorization for the target data resource, includes: In response to the authorization request operation, authorization is triggered for the target privacy computing component corresponding to the target component identifier and the target data resource; wherein, after the target privacy computing component is authorized for use, it is used to instruct the data user to communicate with the converged computing platform through the corresponding computing component interface of the target privacy computing component.

4. The method according to claim 3, characterized in that, The target component identifier and the target data resource are specified in the data usage contract to be created; the authorization application operation is a contract creation and review operation for the data usage contract; The response to the authorization request operation triggers authorization for the target privacy computing component corresponding to the target component identifier and the target data resource, including: In response to the contract creation and review operation, the converged computing platform and the data provider are triggered to perform contract creation and review processes on the data respectively. The approved data uses the target privacy computing component and the target data resources specified in the contract, which are also authorized for use.

5. The method according to claim 1, characterized in that, The target data resource includes the provider sample data provided by the target data provider; the method further includes: After the target data resource is authorized for use, in response to the model configuration operation for the business model to be configured, the business model to be configured is configured for model training to obtain the configured business model to be trained and the sample-related information of the configured business model to be trained. In response to a model training trigger operation for the business model to be trained, the user-side sample data and the provider-side sample data corresponding to the sample-related information are used to jointly train the business model to be trained.

6. The method according to claim 1, characterized in that, The method further includes: After the business model training is completed and a virtual shared business model is obtained, the model call configuration interface of the data user terminal is displayed; In response to the model call configuration operation on the model call configuration interface, model call constraint information configured for the virtual shared service model is obtained; the model call constraint information is used to constrain the data user terminal's call to the virtual shared service model.

7. The method according to any one of claims 1 to 6, characterized in that, The method further includes: After the business model training is completed and a virtual shared business model is obtained, the model deployment configuration interface of the data user terminal is displayed. In response to the model deployment trigger operation on the model deployment configuration interface, model deployment configuration is triggered for the virtual shared business model; the deployed and configured virtual shared business model is used to perform joint data prediction on the online data of the data user and the target online data corresponding to the target data resource to obtain online prediction results for the data user to use in business processing; the target online data is the online data provided by the target data provider.

8. A cross-platform data processing method, characterized in that, The method includes: The data provider receives a data resource authorization request sent by the data user; the data resource authorization request carries target data resource description information; the target data resource description information is selected from at least one data resource description information displayed on the data user; the at least one data resource description information is used to describe the data resources published by the data provider to the converged computing platform; the converged computing platform is a standardized platform compatible with at least two privacy computing components; each privacy computing component has a corresponding computing component interface; In response to the data resource authorization request, an authorization review interface for the target data resource is displayed; the target data resource is the data resource described in the target data resource description information. In response to the authorization operation for the target data resource in the authorization review interface, a resource configuration interface for the target data resource is displayed; In response to the mapping configuration operation for the target data resource in the resource configuration interface, a mapping relationship is established between the target data resource description information and the target data resource locally on the data provider. After the mapping relationship is established, it is used to instruct the data user to train a business model based on user-side sample data. The intermediate results during model training are encrypted using a first public key issued by the fusion computing platform based on a first computing component interface matching the data user, yielding a first intermediate feature. This first intermediate feature is then sent to the fusion computing platform via the first computing component interface, causing the fusion computing platform to forward the first intermediate feature to the data provider via a second computing component interface matching the data provider, thus instructing the data provider to train the data provider based on the first and second intermediate features. The data provider obtains the business model in the first computing component interface. The second intermediate feature is obtained by encrypting the intermediate results generated during training the business model using the target data resources, based on the second public key issued by the converged computing platform. The data provider receives the second intermediate feature from the converged computing platform via the first computing component interface and continues training the business model in the data user based on the second and first intermediate features to obtain encrypted adjustment parameters. The data provider uploads the encrypted adjustment parameters to the converged computing platform via the first computing component interface and obtains the decrypted adjustment parameters. The data user updates the business model based on the adjustment parameters, resulting in a first business model for the data user after training. The first business model and the second business model trained by the data provider constitute a jointly trained virtual shared business model.

9. The method according to claim 8, characterized in that, Before the data provider receives the data resource authorization request sent by the data user, the method further includes: In response to the basic information configuration operation for the data resources to be published in the data provider, generate data resource description information for the data resources to be published. In response to the resource publishing operation, the data resource to be published is triggered to be published to the converged computing platform; the data resource description information of the published data resource is used to display on the data resource selection interface of the data user terminal.

10. The method according to claim 8, characterized in that, The method further includes: In response to a model call configuration review request sent by the data user for the virtual shared business model, a call configuration review interface is displayed; the call configuration review interface displays model call constraint information that is pending review and configured for the virtual shared business model; In response to the configuration agreement operation, the model call constraint information is authorized; the authorized model call constraint information is used to restrict the data user's call to the virtual shared service model.

11. A cross-platform data processing method, characterized in that, The method includes: In response to a data resource publishing request sent by the data provider, the data resource description information of the data resource provided by the data provider is displayed on the resource publishing review interface of the platform operation side. Based on the resource approval operation applied to the displayed data resource description information, the approved data resource is triggered to be published to the converged computing platform; the converged computing platform is a standardized platform compatible with at least two privacy computing components; each privacy computing component has a corresponding computing component interface; The published data resource description information is displayed on the data resource selection interface of the data user, allowing the data user to select the target data resource description information from the displayed information. It also instructs the user that after the target data resource mapped to by the target data resource description information is authorized for use, the user trains a business model on the data user based on sample data. The user then uses the first public key issued by the fusion computing platform based on the first computing component interface matched with the data user to encrypt the intermediate results during model training, obtaining a first intermediate feature. The user then sends the first intermediate feature to the fusion computing platform based on the first computing component interface, causing the fusion computing platform to forward the first intermediate feature to the target data provider based on the second computing component interface matched with the target data provider. This instructs the target data provider to use the first intermediate feature and the second public key to encrypt the intermediate results during model training, obtaining a first intermediate feature. The intermediate features are used to train the business model in the target data provider. The second intermediate feature is obtained by encrypting the intermediate results generated when training the business model using the target data resources based on the second public key issued by the converged computing platform. The second intermediate feature sent by the converged computing platform is received through the first computing component interface, and the business model in the data user is further trained based on the second intermediate feature and the first intermediate feature to obtain encrypted adjustment parameters. The encrypted adjustment parameters are uploaded to the converged computing platform through the first computing component interface, and the decrypted adjustment parameters are obtained. The business model in the data user is updated based on the adjustment parameters, and a first business model of the data user is obtained after training. The first business model and the second business model trained by the target data provider constitute a jointly trained virtual shared business model.

12. A cross-platform data processing device, characterized in that, The device includes: The first display module is used to display at least one data resource description information on the data resource selection interface of the data user terminal; the data resource description information is used to describe the data resources published by the data provider terminal to the converged computing platform; the converged computing platform is a standardized platform compatible with at least two privacy computing components; each privacy computing component has a corresponding computing component interface; A determination module is configured to, in response to a data resource selection operation, determine a selected target data resource description from the at least one data resource description information; the target data resource mapped to by the target data resource description information is provided by a target data provider. The first triggering module is used to trigger authorization for the target data resource in response to an authorization request operation; wherein, after the target data resource is authorized for use, it is used to instruct the training of a business model in the data user based on the user's sample data, and to encrypt the intermediate results in the model training using a first public key issued by the fusion computing platform based on a first computing component interface matching the data user, to obtain a first intermediate feature; the first intermediate feature is sent to the fusion computing platform based on the first computing component interface, so that the fusion computing platform forwards the first intermediate feature to the target data provider based on a second computing component interface matching the target data provider, thereby instructing the target data provider to train the business model in the target data provider based on the first intermediate feature and the second intermediate feature; the second The intermediate feature is obtained by encrypting the intermediate results generated when training the business model using the target data resources based on the second public key issued by the converged computing platform. The second intermediate feature sent by the converged computing platform is received by the first computing component interface, and the business model in the data user is further trained based on the second and first intermediate features to obtain encrypted adjustment parameters. The encrypted adjustment parameters are uploaded to the converged computing platform via the first computing component interface, and the decrypted adjustment parameters are obtained. The business model in the data user is updated based on the adjustment parameters, resulting in a first business model for the data user after training is complete. The first business model and the second business model trained by the target data provider constitute a jointly trained virtual shared business model.

13. The cross-platform data processing apparatus according to claim 12, characterized in that, After the target data resource is authorized for use, the first triggering module is also used to respond to the model training triggering operation to trigger the training of the business model in the data user terminal based on the user terminal sample data, and use the first public key issued by the fusion computing platform based on the first computing component interface matched with the data user terminal to encrypt the intermediate results in the model training to obtain the first intermediate feature.

14. The cross-platform data processing apparatus according to claim 12, characterized in that, The first display module is further configured to display component identifiers of at least two privacy computing components on the privacy computing component selection interface; and to determine the selected target component identifier in response to the component selection operation; The first triggering module is also used to respond to the authorization request operation by triggering authorization for the target privacy computing component corresponding to the target component identifier and the target data resource; wherein, after the target privacy computing component is authorized for use, it is used to instruct the data user to communicate with the converged computing platform through the corresponding computing component interface of the target privacy computing component.

15. The cross-platform data processing apparatus according to claim 14, characterized in that, The target component identifier and the target data resource are specified in the data usage contract to be created; the authorization application operation is a contract creation and review operation for the data usage contract; The first triggering module is also used to respond to the contract creation and review operation by triggering the fused computing platform and the data provider to perform contract creation and review processing on the data respectively. The approved data uses the target privacy computing component and the target data resources specified in the contract, which are also authorized for use.

16. The cross-platform data processing apparatus according to claim 12, characterized in that, The target data resource includes the provider sample data provided by the target data provider; The first triggering module is further configured to, after the target data resource is authorized for use, respond to the model configuration operation for the business model to be configured, perform model training configuration on the business model to be configured, and obtain the configured business model to be trained and the sample-related information of the configured business model to be trained; In response to a model training trigger operation for the business model to be trained, the user-side sample data and the provider-side sample data corresponding to the sample-related information are used to jointly train the business model to be trained.

17. The cross-platform data processing apparatus according to claim 12, characterized in that, The first display module is further configured to display the model call configuration interface of the data user terminal after the business model training is completed and a virtual shared business model is obtained; in response to the model call configuration operation on the model call configuration interface, obtain the model call constraint information configured for the virtual shared business model; the model call constraint information is used to constrain the data user terminal's call to the virtual shared business model.

18. The cross-platform data processing apparatus according to any one of claims 12 to 17, characterized in that, The first display module is further configured to display the model deployment configuration interface of the data user terminal after the business model training is completed and a virtual shared business model is obtained; in response to the model deployment trigger operation on the model deployment configuration interface, trigger the model deployment configuration for the virtual shared business model; the deployed and configured virtual shared business model is used to perform joint data prediction on the online data of the data user terminal and the target online data corresponding to the target data resource to obtain online prediction results for the data user terminal to use in business processing; The target online data is the online data provided by the target data provider.

19. A cross-platform data processing device, characterized in that, The device includes: A receiving module is used for the data provider to receive a data resource authorization request sent by the data user; the data resource authorization request carries target data resource description information; the target data resource description information is selected from at least one data resource description information displayed on the data user; the at least one data resource description information is used to describe the data resources published by the data provider to the converged computing platform; the converged computing platform is a standardized platform compatible with at least two privacy computing components; each privacy computing component has a corresponding computing component interface; The second display module is used to display an authorization review interface for the target data resource in response to the data resource authorization request; the target data resource is the data resource described by the target data resource description information; and to display a resource configuration interface for the target data resource in response to the authorization operation for the target data resource in the authorization review interface. A module is established to respond to a mapping configuration operation for the target data resource in the resource configuration interface, establishing a mapping relationship between the target data resource description information and the target data resource locally on the data provider. After the mapping relationship is established, it instructs the data user to train a business model based on user-side sample data, using a first public key issued by the fusion computing platform based on a first computing component interface matching the data user to encrypt intermediate results during model training, obtaining a first intermediate feature. The first intermediate feature is then sent to the fusion computing platform via the first computing component interface, causing the fusion computing platform to forward the first intermediate feature to the data provider via a second computing component interface matching the data provider, thus instructing the data provider to train the data user based on the first and second intermediate features. The data provider uses a business model; the second intermediate feature is obtained by encrypting the intermediate results generated when training the business model using the target data resources based on the second public key issued by the converged computing platform; the data provider receives the second intermediate feature sent by the converged computing platform based on the first computing component interface, and continues to train the business model in the data user based on the second intermediate feature and the first intermediate feature to obtain encrypted adjustment parameters; the data provider uploads the encrypted adjustment parameters to the converged computing platform based on the first computing component interface and obtains the adjustment parameters after decryption; the data user updates the business model in the data user based on the adjustment parameters, and obtains the first business model of the data user after training is completed; the first business model and the second business model trained by the data provider constitute a jointly trained virtual shared business model.

20. The cross-platform data processing apparatus according to claim 19, characterized in that, The device further includes a generation module, which is used to generate data resource description information of the data resource to be published in response to a basic information configuration operation for the data resource to be published in the data provider. In response to the resource publishing operation, the data resource to be published is triggered to be published to the converged computing platform; the data resource description information of the published data resource is used to display on the data resource selection interface of the data user terminal.

21. The cross-platform data processing apparatus according to claim 19, characterized in that, The second display module is also used to respond to a model call configuration review request sent by the data user terminal for the virtual shared business model, and display a call configuration review interface; the call configuration review interface displays model call constraint information that is pending review and configured for the virtual shared business model; In response to the configuration agreement operation, the model call constraint information is authorized; the authorized model call constraint information is used to restrict the data user's call to the virtual shared service model.

22. A cross-platform data processing device, characterized in that, The device includes: The third display module is used to respond to the data resource publishing request sent by the data provider and display the data resource description information of the data resource provided by the data provider on the resource publishing review interface of the platform operation end. The second triggering module is used to trigger the publication of the approved data resources to the converged computing platform based on the resource approval operation applied to the displayed data resource description information; the converged computing platform is a standardized platform compatible with at least two privacy computing components; each privacy computing component has a corresponding computing component interface; The published data resource description information is displayed on the data resource selection interface of the data user, allowing the data user to select the target data resource description information from the displayed information. It also instructs the user that after the target data resource mapped to by the target data resource description information is authorized for use, the user trains a business model on the data user based on sample data. The user then uses the first public key issued by the fusion computing platform based on the first computing component interface matched with the data user to encrypt the intermediate results during model training, obtaining a first intermediate feature. The user then sends the first intermediate feature to the fusion computing platform based on the first computing component interface, causing the fusion computing platform to forward the first intermediate feature to the target data provider based on the second computing component interface matched with the target data provider. This instructs the target data provider to use the first intermediate feature and the second public key to encrypt the intermediate results during model training, obtaining a first intermediate feature. The intermediate features are used to train the business model in the target data provider. The second intermediate feature is obtained by encrypting the intermediate results generated when training the business model using the target data resources based on the second public key issued by the converged computing platform. The second intermediate feature sent by the converged computing platform is received through the first computing component interface, and the business model in the data user is further trained based on the second intermediate feature and the first intermediate feature to obtain encrypted adjustment parameters. The encrypted adjustment parameters are uploaded to the converged computing platform through the first computing component interface, and the decrypted adjustment parameters are obtained. The business model in the data user is updated based on the adjustment parameters, and a first business model of the data user is obtained after training. The first business model and the second business model trained by the target data provider constitute a jointly trained virtual shared business model.

23. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the method of any one of claims 1 to 11.

24. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the method of any one of claims 1 to 11.

25. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the method of any one of claims 1 to 11.

Images