Data processing method and device, equipment and storage medium
By retrieving the target model and configuration parameters based on the business type when receiving a business request, the problem of poor processing effect and difficulty in upgrading caused by a single algorithm model is solved. This enables comprehensive processing of multiple models, improving the analysis effect and the flexibility and convenience of model deployment.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- CHINA CONSTRUCTION BANK
- Filing Date
- 2022-09-28
- Publication Date
- 2026-06-05
AI Technical Summary
In existing technologies, incorporating a single algorithm model into business services results in poor model processing performance and makes upgrades difficult, thereby affecting user asset security.
Upon receiving a business request to be analyzed, the business type is determined, the corresponding target model and its configuration parameters are retrieved, including the source of the request input parameters and the input interface for the output data. Based on these parameters, the input data is determined and model analysis is performed. Finally, the analysis results are output in the rule model.
It enables integrated processing of multiple models, improves the effectiveness and flexibility of model analysis, simplifies the model deployment process, and protects users' property security.
Smart Images
Figure CN115827084B_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of big data data analysis technology, and in particular to a data processing method, apparatus, device and storage medium. Background Technology
[0002] With the rapid development of big data technology, various algorithm models based on big data are also gradually increasing. Generally, a large number of algorithm models developed by technical personnel are applied to services to solve practical business problems. For example, classification prediction algorithms (such as random forest algorithms) are applied to the financial field to monitor user transaction information to ensure user transaction security. Therefore, how to effectively and quickly apply algorithm models to services has become an urgent technical problem to be solved.
[0003] In existing technologies, the common approach to applying algorithm models to services is to hard-code a specific algorithm model into the corresponding business scenario. When data from that business scenario is received, the algorithm model is invoked to process the data. However, in practical applications, due to the wide variety of algorithm models and the increasing number of user business scenarios, this method of writing a single algorithm model into the service suffers from poor data monitoring. Furthermore, when the coefficients in the algorithm model change, the modified algorithm model needs to be redeployed into the service, which is not only inefficient and cumbersome but may also lead to financial losses for users. Summary of the Invention
[0004] This application provides a data processing method, apparatus, device, and storage medium to improve the flexibility, speed, and convenience of model deployment while enhancing the model analysis and processing effect, thereby achieving the technical effect of protecting users' property security.
[0005] In a first aspect, this application provides a data processing method, comprising: upon receiving a business request to be analyzed, determining the business type corresponding to the business request; retrieving a target model corresponding to the business type and retrieving pre-configured configuration parameters corresponding to the target model; based on the configuration parameters, determining target input data to be input into the target model, and inputting the target input data into the target model to obtain target output data; and based on the configuration parameters, inputting the target output data into a corresponding rule model to obtain a target analysis result corresponding to the business request to be analyzed.
[0006] Secondly, this application provides a data processing apparatus, comprising: a business type determination module, configured to determine the business type corresponding to the business request to be analyzed upon receiving the business request to be analyzed; a configuration parameter retrieval module, configured to retrieve a target model corresponding to the business type and retrieve pre-configured configuration parameters corresponding to the target model; a target output data determination module, configured to determine target input data to be input into the target model based on the configuration parameters, and input the target input data into the target model to obtain target output data; and a target analysis result determination module, configured to input the target output data into a corresponding rule model based on the configuration parameters to obtain a target analysis result corresponding to the business request to be analyzed.
[0007] Thirdly, embodiments of the present invention also provide an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the data processing method as described in any of the embodiments of the present invention.
[0008] Fourthly, embodiments of the present invention also provide a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the data processing method as described in any of the embodiments of the present invention.
[0009] Fifthly, embodiments of the present invention also provide a computer program product, including a computer program that, when executed by a processor, implements the data processing method as described in any of the embodiments of the present invention.
[0010] The data processing method, apparatus, device, and storage medium provided in this application, upon receiving a business request to be analyzed, determines the business type corresponding to the business request; retrieves the target model corresponding to the business type and pre-configured configuration parameters corresponding to the target model; wherein the configuration parameters include the source of request input parameters and the input interface corresponding to the output data of the target model; based on the configuration parameters, determines the target input data to be input into the target model and inputs the target input data into the target model to obtain the target output data; based on the configuration parameters, inputs the target output data into the corresponding rule model to obtain the target analysis result corresponding to the business request to be analyzed. This solves the problem in the prior art of writing a single algorithm model into the business service, resulting in poor model processing effect and difficulty in model upgrade, and realizes the solution of pre-configuring the input parameters of the target model to obtain the target analysis result. This approach, which involves determining the input parameter source and configuring corresponding input interfaces for output data, upon receiving a business request to be analyzed, retrieves the target model corresponding to the business type based on the request's business type. It then retrieves the target model's input parameter source and output data input interface configuration parameters. These parameters determine the target input data to the target model, ensuring that the target output data is output after the target input data is obtained. Furthermore, based on the configuration parameters, it determines the rule model to which the target output data is input, enabling the output of target analysis results based on the rule model. This multi-model integrated approach to business data processing leads to more accurate target analysis results and improves the effectiveness of model analysis. Additionally, by changing the target model's configuration parameters, it meets the upgrade requirements of the target model, thereby enhancing the flexibility, speed, and convenience of model deployment. Attached Figure Description
[0011] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.
[0012] Figure 1 Flowchart of the data processing method provided in the embodiments of this application Figure 1 ;
[0013] Figure 2 Flowchart of the data processing method provided in the embodiments of this application Figure 2 ;
[0014] Figure 3 Example diagrams of the data processing method provided in the embodiments of this application;
[0015] Figure 4 This is a schematic diagram of the structure of the data processing apparatus provided in the embodiments of this application;
[0016] Figure 5This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application.
[0017] The accompanying drawings illustrate specific embodiments of this application, which will be described in more detail below. These drawings and descriptions are not intended to limit the scope of the concept in any way, but rather to illustrate the concept of this application to those skilled in the art through reference to particular embodiments. Detailed Implementation
[0018] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.
[0019] The technical solution of this application and how it solves the above-mentioned technical problems are described in detail below with specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments. The embodiments of this application will be described below with reference to the accompanying drawings. The acquisition, storage, use, and processing of data in the technical solution of this application all comply with the relevant provisions of national laws and regulations.
[0020] This application provides a data processing method. Figure 1 Flowchart of the data processing method provided in the embodiments of this application Figure 1 .like Figure 1 As shown, the data processing method includes:
[0021] S101. Upon receiving a service request to be analyzed, determine the service type corresponding to the service request to be analyzed.
[0022] The business request to be analyzed can be understood as the business request that needs to be analyzed. The data format of the business request can be JSON or XML. The business type can be used to represent the service type of the business, such as transaction, order, equity transfer out, equity transfer in, and authentication.
[0023] In this embodiment, a request for analysis from the client can be considered received when a user triggers a request control on a page in the client's business system. Alternatively, a request can be considered received when the server receives uploaded data to be analyzed. Furthermore, an algorithm can be used to parse the request to obtain the business type carried in it.
[0024] For example, during the authentication process of user A in the authentication system, when user A triggers the authentication request control on the authentication system page, the server receives user A's authentication request (i.e., the business request to be analyzed), parses the authentication request to obtain the business type of the authentication request, and calls the corresponding algorithm model to analyze the authentication request information based on the business type.
[0025] S102. Retrieve the target model corresponding to the business type, and retrieve the pre-configured configuration parameters corresponding to the target model.
[0026] The target model can be understood as an algorithm model, which can be any PMML (Predictive Model Markup Language) model, such as logistic regression, decision trees, support vector machines, randomized forests, neural network models, etc. Configuration parameters may include, but are not limited to, the source of the request input parameters and the input interface corresponding to the target model's output data. The source of the request input parameters can be understood as the origin of the input parameters; for example, if the input parameter is transaction information, its source could be the user's account information. The input interface can be understood as the input location of the output data.
[0027] In practical applications, after determining the business type corresponding to the business request to be analyzed, mapping techniques can be used to retrieve the corresponding algorithm model as the target model. There can be one or multiple target models. If there is only one target model, configuration parameters such as the request input parameter source and the input interface for output data can be retrieved. If there are multiple target models, configuration parameters such as the request input parameter source and the input interface for output data for each target model can be retrieved. This allows for determining the data required for the input parameters of the target model based on the corresponding request input parameter source, obtaining the input parameter values, quickly outputting the target model's output data, and determining the input positions required for the target model's output data based on the input interfaces corresponding to the output data for use by subsequent models.
[0028] It should be noted that, to improve the speed of data processing and the accuracy of data analysis results, corresponding algorithm models can be configured based on the business type of the business data. This allows the system to determine which corresponding algorithm models should analyze and process the data when data of a certain business type is detected. Optionally, retrieving the target model corresponding to the business type includes: determining and invoking the target model corresponding to the business type based on a pre-created correspondence between business types and models to be invoked.
[0029] The models to be invoked can be pre-stored in the database for later use. Each business type can correspond to one or more models to be invoked. For example, a mapping relationship between each business type and its corresponding model to be invoked can be established in advance using mapping technology and stored in a mapping table, so that the corresponding model to be invoked can be determined from the mapping table based on the business type as the target model.
[0030] Specifically, the model to be invoked can be selected as the target model, and the target model can be invoked. For example, for business type A, algorithm models (such as model B, model C, and model D) that have a corresponding relationship with business type A can be invoked, and these models can be used as the target models.
[0031] S103. Based on the configuration parameters, determine the target input data to be input into the target model, and input the target input data into the target model to obtain the target output data.
[0032] In this embodiment, the required input data for the target model can be obtained using the request input parameter source in the configuration parameters, thus obtaining the target input data. For example, if the input parameter is transaction information and its input parameter source is the user's account information, the transaction information in the account information can be used as the target input data. Furthermore, the target input data can be input into the target model. After analysis and processing by the target model, the target model can output at least one output parameter data, which serves as the target output data.
[0033] S104. Based on the configuration parameters, the target output data is input into the corresponding rule model to obtain the target analysis result corresponding to the business request to be analyzed.
[0034] It should be noted that the rule model can be a pre-set rule algorithm. The rule model includes input parameter information and output parameter information. The required data corresponding to the input parameter information can be the output parameter result of the target model. When the output parameter result of the target model is obtained, the output parameter result is input to the corresponding input parameter of the rule model.
[0035] In this embodiment, the target output data can be input to the input interface corresponding to the output data in the configuration parameters. The input interface can be the input parameter of a rule model, so that when the rule model receives the input parameter data, it outputs the target analysis result corresponding to the business request to be analyzed.
[0036] Specifically, in the process of inputting target output data into the corresponding rule model based on configuration parameters, the target output data can be input into the rule model corresponding to the input interface based on the input interface in the configuration parameters.
[0037] In practical applications, the input interface corresponding to the target output data can be determined based on the input interface in the configuration parameters. That is, the input parameter of which rule model corresponds to the target output data can be determined. Then, the target output data is used as the input parameter data of the rule model and input into the rule model so that the rule model outputs the target analysis result. Subsequently, the target analysis result can be used to determine whether the business request to be analyzed is abnormal (such as transaction abnormality, authentication abnormality, etc.). If the target analysis result is abnormal, the user can be prompted to ensure user security.
[0038] For example, assuming that the input interface corresponding to the target output data A is the input parameter B of rule model 1, the target output data A can be used as the data of the input parameter B of rule model 1 and input so that rule model 1 outputs the result.
[0039] This embodiment solves the problem in existing technologies where writing a single algorithm model into a business service leads to poor model processing and difficulty in model upgrades. It achieves this by pre-configuring the input parameter source for the target model's input parameters and the corresponding input interface for the output data, based on the business type of the business request being received. Upon receiving a business request, it determines the business type corresponding to the request, retrieves the target model corresponding to the business type, and pre-configures the pre-configured configuration parameters and pre-sets the corresponding configuration parameters for the output data. The system retrieves the target model corresponding to the business type, along with its configuration parameters such as the source of input parameters and the input interface for output data. These parameters are then used to determine the target input data to the target model, ensuring that the target output data is output after the target input data is received. Based on the configuration parameters, the system determines the rule model to which the target output data is input, enabling the output of target analysis results based on the rule model. This multi-model approach to business data processing leads to more accurate target analysis results and improves the effectiveness of model analysis. Furthermore, by modifying the target model's configuration parameters, upgrade requirements can be met, enhancing the flexibility, speed, and convenience of model deployment.
[0040] Based on the above embodiments, when retrieving pre-configured configuration parameters corresponding to the target model, a pre-created target configuration table corresponding to the target model can be retrieved. The target configuration table contains configuration information of the target model, from which configuration parameters can be obtained. Furthermore, when determining the target input data to the target model based on the configuration parameters, the request input parameter source in the configuration parameters can determine the location for obtaining the data required for the request input parameters, and obtain the required input data, i.e., the data to be used, so as to input the data to be used into the target model to obtain the output data of the target model. Accordingly, this application proposes the following embodiments:
[0041] Figure 2 Flowchart of the data processing method provided in the embodiments of this application Figure 2 .like Figure 2 As shown, this data processing method includes the following steps:
[0042] S201. Determine the target configuration table corresponding to the target model.
[0043] The target configuration table may contain configuration information corresponding to the target model. The configuration information may include, but is not limited to, the corresponding business type, model name, model number, model type, input parameters and their sources, input parameter processing order, output parameters and their input interfaces.
[0044] Specifically, the target configuration table corresponding to the target model can be read from the database to obtain the corresponding configuration parameters.
[0045] Optionally, the creation of the target configuration table can be achieved by: determining the configuration parameters corresponding to each model to be called; creating the target configuration table based on the configuration parameters and the identifier corresponding to the model to be called, so as to determine the configuration parameters corresponding to the target model based on the target configuration table.
[0046] The identifier can be used to represent the uniqueness of the model. For example, for model 1, 1101 can be used as the identifier of model 1, and for model 2, 1102 can be used as the identifier of model 2.
[0047] In practical applications, various attribute information from the model to be invoked can be used as input parameters, and a corresponding input source can be configured for each input parameter. Similarly, corresponding input interfaces can be configured for the output data of the model to be invoked. After configuring the input sources for the input parameters and the input interfaces for the output data of the model to be invoked, the configuration parameters of the model to be invoked can be considered determined. Furthermore, the identifier corresponding to the model to be invoked can be mapped to the corresponding configuration parameters to create a configuration table corresponding to that model. It should be noted that the configuration table corresponding to each model to be invoked can be used as the target configuration table, or multiple configuration tables corresponding to models to be invoked can be combined into a single target configuration table. This allows the model configuration parameters matching the identifier of the target model to be retrieved from the target configuration table after the target model is determined.
[0048] To improve the convenience and speed of model configuration, models can be imported through the model import interface on the system page. The model can then be parsed to obtain the input and output parameters, allowing for the configuration of corresponding input parameter sources for the input parameters and corresponding input interfaces for the output parameters. Based on the input parameter sources and input interfaces, the model's configuration parameters can be determined.
[0049] Optionally, the configuration parameters corresponding to each model to be called are determined, including: when the uploaded document to be parsed corresponding to each model to be called is received, the document to be parsed is parsed to obtain the parameters to be used corresponding to each model to be called; the parameters to be used include at least one model input parameter and at least one model output parameter; for each model to be called, the configuration parameters corresponding to the current model to be called are determined based on the input parameter source corresponding to each model input parameter in the current model to be called and the input interface corresponding to each model output parameter.
[0050] The document to be parsed refers to the model file, which can be in XML, TXT, or DOC format. The method for determining the configuration parameters corresponding to each model to be invoked is the same; any one of these models can be used as the current model to be invoked.
[0051] In this embodiment, the user can import the model files corresponding to each model to be called from the model import interface. When the system receives the imported model file, it can consider that it has received the uploaded document to be parsed. Furthermore, the document to be parsed can be parsed and stored in the model table. At this time, the model table can include the basic data of the model to be called (e.g., model name, model code, model function, etc.). The model code includes multiple model parameter information, which serves as the parameters to be used. For the current model to be called, various attribute information of the current model to be called can be extracted as model input parameters, and corresponding input parameter sources can be configured for the model input parameters. Corresponding output interfaces can be configured for the various model output parameters of the current model to be called. The input parameter sources and output interfaces can be used as configuration parameters for the current model to be called. For example, for model 1 to be called, the input parameter source of its model input parameter 1 can be an account, and the output interface of its model output parameter 2 can be the input parameter 3 of rule model 1.
[0052] It should be noted that, to further improve the flexibility of model configuration, the configuration information of the model to be called can be configured in multiple tables. Each data table can correspond to different configuration content, thereby constructing a target configuration table for the model to be called based on each data table. Optionally, the target configuration table includes an input parameter configuration table corresponding to the input parameters of the model to be called, an input parameter source configuration table corresponding to the input parameters of the model to be called, and a data input interface configuration table corresponding to the output parameters of the model to be called.
[0053] The input parameter configuration table includes the mapping relationship between the input parameters of the model to be called and the identifier of the model to be called; the input parameter source configuration table includes the mapping relationship between the input parameters of the model to be called and the input parameter source of the input parameters of the model to be called; and the data input interface configuration table includes the mapping relationship between the output parameters of the model to be called and the input interface of the output parameters.
[0054] In this embodiment, after determining the input parameters of the model to be called, a mapping relationship can be established between the input parameters of the model to be called and the identifier of the model to be called, generating an input parameter configuration table. For example, for model 1 to be called, its model identifier can be represented by PMML_XX1, and the input parameters in model 1 to be called can be represented by input parameter number 1, input parameter name Arg1, and / or input parameter identifier Deftetdse1. PMML_XX1, 1, Arg1, and Deftetdse1 can be used as a record in the input parameter configuration table. Accordingly, an input parameter configuration table containing the mapping relationship between multiple input parameters of the models to be called and their corresponding model identifiers can be obtained. For example, the input parameter configuration table is shown in Table 1 below:
[0055] Table 1
[0056] Model Identifier Input parameter number Input parameter name Input parameter identifier PMML_XX1 1 Arg 1 Deftetdse 1 … … … … PMML_XX1 n Arg n Deftetdse n
[0057] Based on the above scheme, corresponding input parameter sources can be configured for the input parameters of the model to be called, thereby establishing a mapping relationship between the input parameters of the model to be called and the corresponding input parameter sources, and generating an input parameter source configuration table. For example, assuming that the input parameter source of input parameter 1 of the model to be called is an account, input parameter 1 and account can be used as a record in the input parameter source configuration table. Optionally, this record can also include input parameter identifier, input parameter source type, input parameter processing order, input parameter source number, processing method, processing parameters, etc. As an example, the input parameter source configuration table is shown in Table 2 below:
[0058] Table 2
[0059]
[0060] In this embodiment, corresponding input interfaces can be configured for the output parameters of the model to be called, thereby establishing a mapping relationship between the output parameters of the model to be called and the corresponding input interfaces, and generating a data input interface configuration table. For example, assuming that the input interface of output parameter 1 of model 1 to be called is input parameter a of rule model 1, input parameter 1 and input parameter a of rule model 1 can be used as a record in the data input interface configuration table. Optionally, the record may also include model identifier, output parameter number, output parameter name, output parameter identifier, etc. For example, the data input interface configuration table is shown in Table 3 below:
[0061] Table 3
[0062] Model Identifier Parameter Number Output Parameter Name Output indicator Input Interface PMML_XX1 1 brg 1 beftetdse 1 a … … … … … PMML_XX1 n brg n beftetdse n n
[0063] S202. Based on the target configuration table, determine the configuration parameters corresponding to the target model.
[0064] In practical applications, the source of the request input parameters corresponding to the input parameters of the target model and the input interface corresponding to the output data of the target model can be obtained from the target configuration table, that is, the configuration parameters of the target model are obtained.
[0065] For example, assuming the target model is identified as PMML_XX1, the input parameter identifier Deftetdse 1 can be obtained from the input parameter configuration table based on PMML_XX1, and the input parameter source number 1 corresponding to Deftetdse 1 can be obtained from the input parameter source configuration table. The input interface a corresponding to the output parameter identifier beftetdse 1 of PMML_XX1 can be obtained from the data input interface configuration table. The input parameter source number 1 and the input interface a can be used as configuration parameters to determine from which location the required data for the input parameter is obtained based on the input parameter source number 1, and to determine from which location the corresponding output parameter is input based on the input interface a.
[0066] The technical solution of this embodiment, by determining the target configuration table corresponding to the target model, and then determining the configuration parameters corresponding to the target model based on the target configuration table, has the advantage that: the target input data input to the target model and the position where the target output data of the target model should be input can be obtained based on the configuration parameters, thereby improving the speed of data processing and ensuring the accuracy of data analysis.
[0067] S203. Based on the request input parameter source in the configuration parameters, retrieve at least one piece of data to be used.
[0068] In practical applications, the required data for the request parameters is determined from the location specified in the configuration parameters, and the data is retrieved from the location corresponding to the request parameter source as the data to be used. It should be noted that each model input parameter can correspond to one request parameter source or multiple request parameter sources. Accordingly, at least one piece of data to be used can be obtained for each input parameter in the target model.
[0069] For example, assuming that the source of the request input parameter a is an account, the account information can be obtained from the user account corresponding to the business request to be analyzed as the data to be used in the request input parameter a, with the user's knowledge and permission.
[0070] S204. Based on the at least one data to be used and the calculation rules in the configuration parameters, determine the target input data.
[0071] The calculation rule can be understood as the processing method of the input parameters, which can be pre-configured in the input parameter source configuration table. For example, assuming that input parameter 1 represents the user's average monthly driving distance, its corresponding calculation rule can be the total driving distance in the current month divided by the number of days in the current month.
[0072] In practical applications, after obtaining at least one piece of data to be used corresponding to the model input parameters, the calculation rules corresponding to each model input parameter can be retrieved from the input parameter source configuration table. Furthermore, for the current model input parameter, at least one piece of data to be used corresponding to the current model input parameter can be input into the calculation rule to obtain the data value corresponding to the current model input parameter, which serves as the target input data.
[0073] For example, if the calculation rule for the model input parameter 'a' is a = b + c + d, the data to be used (1) corresponding to b, the data to be used (2) corresponding to c, and the data to be used (3) corresponding to d can be input into the calculation rule to obtain a = 1 + 2 + 3 = 6, which can be used as the target output data. This allows the target input data to be input into the target model to obtain the target output data.
[0074] In this embodiment, by retrieving at least one piece of data to be used based on the request input parameter source in the configuration parameters, the speed and accuracy of obtaining the required data can be improved. Then, based on at least one piece of data to be used and the calculation rules in the configuration parameters, the target input data can be determined, which can realize the rapid output of the input data of the subsequent rule model and improve the efficiency of multi-model fusion processing data.
[0075] Based on the above embodiments, in order to enable those skilled in the art to clearly understand the technical solutions of the embodiments of the present invention, the data processing method will be described in detail below through specific examples:
[0076] In practical applications, a model file (i.e., the document to be parsed) containing at least one algorithm model can be imported into the system using web page functionality. The model file format can be XML, TXT, or DOC, etc. Furthermore, the model file can be parsed, such as parsing tags (e.g., algorithm name, algorithm type, algorithm code, etc.) and extracting each tag and its content, storing these tags and their content in a table. Further, based on the data in the table, the source of input parameters (i.e., input parameter source) and the input location of output parameters (i.e., input interface) for each algorithm model can be configured on the system's web page. For example, input parameters can come from raw messages, supplementary information (such as IP address attribution), basic variables, etc., and the input interface for output parameters can be input parameter A of rule model 1. Furthermore, when data to be analyzed is obtained from the actual business system, at least one algorithm model corresponding to the business type (e.g., transaction type) of the data to be analyzed can be obtained as the target model. Furthermore, the required data can be obtained and processed based on the source of each input parameter in the target model to obtain multiple output parameter data. Further, based on the input interface of each output parameter, the output parameter data is output to the corresponding rule model position of the input interface. Further, based on the rule model, the result corresponding to the data to be analyzed is output as the target analysis result.
[0077] This embodiment pre-configures the input parameter source for the target model and the corresponding input interface for the output data. Upon receiving a business request to be analyzed, it retrieves the target model corresponding to the business type based on the business type of the request, and retrieves the target model's configuration parameters, such as the input parameter source and the input interface for the output data. Then, through these configuration parameters, it determines the target input data to be input into the target model, so that after obtaining the target input data, it outputs target output data. Based on the configuration parameters, it determines the rule model to which the target output data is input, so that the target analysis results are output based on the rule model. This achieves multi-model integrated processing of business data, resulting in more accurate target analysis results and improved model analysis and processing effectiveness. Furthermore, by changing the configuration parameters of the target model, it meets the upgrade requirements of the target model, achieving the technical effect of improving the flexibility, speed, and convenience of model deployment.
[0078] Based on the above embodiments, in order to make the technical solutions of the embodiments of the present invention clearer to those skilled in the art, the data processing method will be described in detail below through specific examples:
[0079] Figure 3 Example diagrams of the data processing method provided in the embodiments of this application, such as... Figure 3As shown, this technology allows importing the model file corresponding to the model to be called from the front-end configuration page. During import, the back-end parsing module automatically retrieves data such as the name, input parameters, and output parameters of the model to be called, and records this data in the model table. Furthermore, the front-end configuration page can configure the input parameter sources and output parameter input interfaces based on the data in the model table. For example, the attributes of the model to be called can be set as all the input parameters of the model, thereby establishing a mapping relationship between the identifier of the model to be called and the corresponding input parameters, and constructing an input parameter configuration table, as shown in Table 1 above. Further, corresponding input parameter sources can be configured for the input parameters of the model to be called, thereby establishing a mapping relationship between the input parameters of the model to be called and the corresponding input parameter sources, generating an input parameter source configuration table, as shown in Table 2 above. Corresponding input interfaces can also be configured for the output parameters of the model to be called, thereby establishing a mapping relationship between the output parameters of the model to be called and the corresponding input interfaces, generating a data input interface configuration table, as shown in Table 3 above. The input parameter configuration table, input parameter source configuration table, and data input interface configuration table can be used as the target configuration table corresponding to the configuration parameters. Furthermore, the target configuration table can be stored in the database of the storage interaction layer for later use. When the application module receives a business request to be analyzed, it reads the target model corresponding to the business type of the business request and the configuration parameters of the target model from the target configuration table. Then, based on the request input parameter source in the configuration parameters, it obtains at least one piece of data to be used corresponding to the input parameter, and inputs the data to be used into the calculation rule of the corresponding input parameter to obtain the target output data of the corresponding input parameter. Further, the target model can be called, and the target output data of each input parameter in the target model can be input accordingly. The target model outputs the target output data. The input interface corresponding to each target output data is determined from the target configuration table. The input interface is a certain input parameter of the rule model, and the target output data is input into the corresponding input parameter of the rule model so that the rule model outputs the target analysis result corresponding to the business request to be analyzed. This allows subsequent judgments based on the target analysis results to determine whether the business request to be analyzed is abnormal or whether the user has network risks, etc.
[0080] This technical solution stores the algorithm model's corresponding model file in a tabular format as input parameters. This allows the algorithm model to be combined with a rule model, using the algorithm model's output as the rule model's input. This improves the convenience and flexibility of model fusion processing. Furthermore, this solution supports modifying and updating the algorithm model at any time, enabling dynamic model deployment and enhancing the flexibility and speed of model deployment.
[0081] This embodiment solves the problem in existing technologies where writing a single algorithm model into a business service leads to poor model processing and difficulty in model upgrades. It achieves this by pre-configuring the input parameter source for the target model's input parameters and the input interface corresponding to the target model's output data upon receiving the business request. It then retrieves the target model corresponding to the business request, along with pre-configured configuration parameters for the target model's input parameters and output data. The configuration parameters include the request input parameter source and the input interface corresponding to the target model's output data. By configuring corresponding input interfaces, upon receiving a business request to be analyzed, the system retrieves the target model corresponding to the business type based on the business type of the request. It also retrieves the target model's input parameter source and output data input interface configuration parameters. These configuration parameters determine the target input data to the target model, ensuring that the target output data is output after the target input data is received. Furthermore, based on the configuration parameters, the system determines the rule model to which the target output data is input, enabling the output of target analysis results based on the rule model. This multi-model integrated approach to business data processing results more accurately and improves the effectiveness of model analysis. Additionally, by changing the target model's configuration parameters, upgrade requirements can be met, enhancing the flexibility, speed, and convenience of model deployment.
[0082] Figure 4 This is a schematic diagram of the structure of the data processing apparatus provided in an embodiment of this application. Figure 4 As shown, the data processing device includes: a business type determination module 501, a configuration parameter retrieval module 502, a target output data determination module 503, and a target analysis result determination module 504;
[0083] The system includes a business type determination module 501, which determines the business type corresponding to the business request to be analyzed upon receiving the request; a configuration parameter retrieval module 502, which retrieves the target model corresponding to the business type and retrieves pre-configured configuration parameters corresponding to the target model; a target output data determination module 503, which determines the target input data to be input into the target model based on the configuration parameters and inputs the target input data into the target model to obtain target output data; and a target analysis result determination module 504, which inputs the target output data into the corresponding rule model based on the configuration parameters to obtain the target analysis result corresponding to the business request to be analyzed.
[0084] This embodiment, upon receiving a business request to be analyzed, determines the business type corresponding to the request; retrieves the target model corresponding to the business type and pre-configured configuration parameters corresponding to the target model; the configuration parameters include the request input parameter source and the input interface corresponding to the target model's output data; based on the configuration parameters, determines the target input data to be input into the target model and inputs the target input data into the target model to obtain the target output data; based on the configuration parameters, inputs the target output data into the corresponding rule model to obtain the target analysis result corresponding to the business request to be analyzed. This solves the problem in the prior art where writing a single algorithm model into the business service leads to poor model processing performance and difficulty in model upgrades. It achieves the goal of pre-configuring the request input parameter source for the target model's input parameters and the output data... By configuring corresponding input interfaces, upon receiving a business request to be analyzed, the system retrieves the target model corresponding to the business type based on the business type of the request. It also retrieves the target model's input parameter source and output data input interface configuration parameters. These configuration parameters determine the target input data to the target model, ensuring that the target output data is output after the target input data is received. Furthermore, based on the configuration parameters, the system determines the rule model to which the target output data is input, enabling the output of target analysis results based on the rule model. This multi-model integrated approach to business data processing results more accurately and improves the effectiveness of model analysis. Additionally, by changing the target model's configuration parameters, upgrade requirements can be met, enhancing the flexibility, speed, and convenience of model deployment.
[0085] In some embodiments, the configuration parameters may optionally include the request input source and the input interface corresponding to the target model output data.
[0086] In some embodiments, optionally, the configuration parameter retrieval module 502 is further configured to determine and invoke the target model corresponding to the business type based on the pre-created correspondence between the business type and the model to be invoked.
[0087] In some embodiments, the configuration parameter retrieval module 502 may optionally include a target configuration table determination unit and a configuration parameter determination unit.
[0088] The target configuration table determination unit is used to determine the target configuration table corresponding to the target model.
[0089] The configuration parameter determination unit is used to determine the configuration parameters corresponding to the target model based on the target configuration table.
[0090] In some embodiments, the target output data determination module 503 may optionally include: a data to be used determination unit and a target input data determination unit.
[0091] The data to be used determination unit is used to retrieve at least one data to be used based on the request input parameter source in the configuration parameters;
[0092] The target input data determination unit is used to determine the target input data based on the at least one data to be used and the calculation rules in the configuration parameters.
[0093] In some embodiments, optionally, the target analysis result determination module 504 is further configured to input the target output data into a rule model corresponding to the input interface based on the input interface in the configuration parameters.
[0094] In some embodiments, the apparatus may optionally further include a target configuration table creation module, which includes a configuration parameter determination unit and a target configuration table determination unit.
[0095] The configuration parameter determination unit is used to determine the configuration parameters corresponding to each model to be called;
[0096] The target configuration table determination unit is used to create the target configuration table based on the configuration parameters and the identifier corresponding to the model to be called, so as to determine the configuration parameters corresponding to the target model based on the target configuration table.
[0097] In some embodiments, the configuration parameter determination unit may optionally include: a parameter to be used determination subunit and a configuration parameter determination subunit.
[0098] The parameter determination subunit is used to parse the uploaded document corresponding to each model to be called when it receives the document to be parsed, and obtain the parameters to be used corresponding to each model to be called; wherein the parameters to be used include at least one model input parameter and at least one model output parameter;
[0099] The configuration parameter determination subunit is used to determine the configuration parameters corresponding to each model to be called, based on the input parameter source corresponding to each model input parameter in the current model to be called and the input interface corresponding to each model output parameter.
[0100] In some embodiments, optionally, the target configuration table includes an input parameter configuration table corresponding to the input parameters of the model to be called, an input parameter source configuration table corresponding to the input parameters of the model to be called, and a data input interface configuration table corresponding to the output parameters of the model to be called; wherein,
[0101] The input parameter configuration table includes the mapping relationship between the input parameters of the model to be called and the identifier of the model to be called;
[0102] The input parameter source configuration table includes the mapping relationship between the input parameters of the model to be called and the input parameter sources of the model to be called;
[0103] The data input interface configuration table includes the mapping relationship between the output parameters of the model to be called and the input interface of the output parameters.
[0104] The data processing apparatus provided in this application embodiment can be used to execute the technical solution of the data processing method in the above embodiment. Its implementation principle and technical effect are similar, and will not be described again here.
[0105] It should be noted that the division of the various modules in the above device is merely a logical functional division. In actual implementation, they can be fully or partially integrated into a single physical entity, or they can be physically separated. Furthermore, these modules can be implemented entirely in software via processing element calls; they can be fully implemented in hardware; or some modules can be implemented by processing element calls to software, while others are implemented in hardware. For example, the service type determination module 501 can be a separate processing element, or it can be integrated into a chip in the above device. Alternatively, it can be stored as program code in the memory of the above device, and its function can be called and executed by a processing element of the above device. The implementation of other modules is similar. Moreover, these modules can be fully or partially integrated together, or they can be implemented independently. The processing element here can be an integrated circuit with signal processing capabilities. In the implementation process, each step of the above method or each of the above modules can be completed through integrated logic circuits in the hardware of the processor element or through software instructions.
[0106] Figure 5 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Figure 5 As shown, the electronic device may include: transceiver 121, processor 122, and memory 123.
[0107] Processor 122 executes computer execution instructions stored in memory, causing processor 122 to perform the scheme in the above embodiments. Processor 122 may be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it may also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.
[0108] The memory 123 is connected to the processor 122 via the system bus and completes communication between them. The memory 123 is used to store computer program instructions.
[0109] Transceiver 121 can be used to send the target processing result corresponding to the service request.
[0110] The system bus can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. The system bus can be divided into address bus, data bus, control bus, etc. For ease of representation, only one thick line is used in the diagram, but this does not indicate that there is only one bus or one type of bus. Transceivers are used to enable communication between database access devices and other computers (e.g., clients, read-write libraries, and read-only libraries). Memory may include random access memory (RAM) and may also include non-volatile memory.
[0111] The electronic device provided in this application embodiment can be the terminal device described in the above embodiments.
[0112] This application also provides a chip for executing instructions, which is used to execute the data processing method described in the above embodiments.
[0113] This application also provides a computer-readable storage medium storing computer instructions that, when executed on a computer, cause the computer to perform the data processing method described in the above embodiments.
[0114] This application also provides a computer program product, which includes a computer program stored in a computer-readable storage medium. At least one processor can read the computer program from the computer-readable storage medium, and when the at least one processor executes the computer program, it can implement the technical solution of the data processing method in the above embodiments.
[0115] Other embodiments of this application will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of this application that follow the general principles of this application and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of this application are indicated by the following claims.
[0116] It should be understood that this application is not limited to the precise structure described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this application is limited only by the appended claims.
Claims
1. A data processing method, characterized in that, include: Upon receiving a service request to be analyzed, determine the service type corresponding to the service request; The system retrieves the target model corresponding to the business type and the pre-configured configuration parameters corresponding to the target model; the configuration parameters include the source of the request input parameters and the input interface corresponding to the output data of the target model. Based on the configuration parameters, the target input data is determined and input into the target model to obtain the target output data; Based on the input interface in the configuration parameters, the target output data is input into the rule model corresponding to the input interface to obtain the target analysis result corresponding to the business request to be analyzed; When the uploaded document corresponding to each model to be called is received, the document is parsed to obtain the parameters to be used corresponding to each model to be called; wherein, the parameters to be used include at least one model input parameter and at least one model output parameter; Based on the input parameter sources corresponding to the input parameters of each model in the current model to be called, and the input interfaces corresponding to the output parameters of each model, determine the configuration parameters corresponding to the current model to be called. Based on the configuration parameters and the identifiers corresponding to the models to be invoked, a target configuration table is created to determine the configuration parameters corresponding to the target models.
2. The method according to claim 1, characterized in that, The retrieval of the target model corresponding to the business type includes: Based on the pre-created correspondence between business types and models to be invoked, the target model corresponding to the business type is determined and invoked.
3. The method according to claim 1, characterized in that, The process of retrieving pre-configured configuration parameters corresponding to the target model includes: Determine the target configuration table corresponding to the target model; Based on the target configuration table, the configuration parameters corresponding to the target model are determined.
4. The method according to claim 1, characterized in that, The step of determining the target input data to the target model based on the configuration parameters includes: Based on the request input parameters in the configuration parameters, retrieve at least one piece of data to be used; The target input data is determined based on the at least one piece of data to be used and the calculation rules in the configuration parameters.
5. The method according to claim 1, characterized in that, The target configuration table includes an input parameter configuration table corresponding to the input parameters of the model to be called, an input parameter source configuration table corresponding to the input parameters of the model to be called, and a data input interface configuration table corresponding to the output parameters of the model to be called; wherein, The input parameter configuration table includes the mapping relationship between the input parameters of the model to be called and the identifier of the model to be called; The input parameter source configuration table includes the mapping relationship between the input parameters of the model to be called and the input parameter sources of the model to be called; The data input interface configuration table includes the mapping relationship between the output parameters of the model to be called and the input interface of the output parameters.
6. A data processing apparatus, characterized in that, include: The business type determination module is used to determine the business type corresponding to the business request to be analyzed when a business request to be analyzed is received. The configuration parameter retrieval module is used to retrieve the target model corresponding to the business type and retrieve the pre-configured configuration parameters corresponding to the target model; the configuration parameters include the request input parameter source and the input interface corresponding to the output data of the target model; The target output data determination module is used to determine the target input data to be input to the target model based on the configuration parameters, and input the target input data into the target model to obtain the target output data; The target analysis result determination module is used to input the target output data into the rule model corresponding to the input interface based on the input interface in the configuration parameters, so as to obtain the target analysis result corresponding to the business request to be analyzed; The parameter determination subunit is used to parse the uploaded document corresponding to each model to be called when it receives the document to be parsed, and obtain the parameters to be used corresponding to each model to be called; wherein the parameters to be used include at least one model input parameter and at least one model output parameter; The configuration parameter determination subunit is used to determine the configuration parameters corresponding to the current model to be called, based on the input parameter source corresponding to the input parameter of each model in the current model to be called and the input interface corresponding to the output parameter of each model. The target configuration table determination unit is used to create a target configuration table based on the configuration parameters and the identifier corresponding to the model to be called, so as to determine the configuration parameters corresponding to the target model based on the target configuration table.
7. The apparatus according to claim 6, characterized in that, The configuration parameter retrieval module is also used to determine and invoke the target model corresponding to the business type based on the pre-created correspondence between the business type and the model to be invoked.
8. The apparatus according to claim 6, characterized in that, The configuration parameter retrieval module includes: The target configuration table determination unit is used to determine the target configuration table corresponding to the target model. The configuration parameter determination unit is used to determine the configuration parameters corresponding to the target model based on the target configuration table.
9. An electronic device, characterized in that, include: A processor, and a memory communicatively connected to the processor; The memory stores computer-executed instructions; The processor executes computer execution instructions stored in the memory to implement the data processing method as described in any one of claims 1-5.
10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions, which, when executed by a processor, are used to implement the data processing method as described in any one of claims 1-5.
11. A computer program product, characterized in that, It includes a computer program that, when executed by a processor, implements the data processing method according to any one of claims 1-5.