Intelligent data modeling method, device, equipment and medium

Abstract

The present application relates to the field of data analysis, and specifically discloses an intelligent data modeling method, apparatus, device, and storage medium. The method includes: obtaining modeling data and user portraits based on a preset data modeling interface; generating one or more data modeling scenarios based on the user portraits; obtaining the user's intended data modeling scenario from the modeling scenario, generating multiple model nodes based on the intended data modeling scenario, and the multiple model nodes including configuration relationships; modifying the configuration relationships of the model nodes in response to user operations; obtaining the calculation parameters corresponding to the model nodes, modeling according to the calculation parameters and configuration relationships, and generating a data model of the modeled data. Through the above method, users can be intelligently assisted, providing a simple and efficient data modeling tool.

Description

Technical Field

[0001] The present application relates to the field of data analysis, and in particular to an intelligent data modeling method, apparatus, device and storage medium. Background Art

[0002] Currently, artificial intelligence and big data technologies are booming, creating enormous value and benefits for society. Of these two technologies, data modeling is the hottest and most cutting-edge. As the pace of digitalization accelerates, the existing capabilities of the data modeling industry are no longer able to meet growing business demands. This is due to high barriers to entry and a shortage of specialized data professionals. Statistical analysis shows that those with data modeling needs can be roughly divided into: professional AI modelers (10%); data analysts (20%); and business professionals with data modeling needs (70%). Existing data modeling tools have complex operational logic and require extremely high mathematical skills, making them suitable only for professional AI modelers and data analysts, but not for a wide range of business professionals. Business professionals, such as risk managers, fraud detection specialists, and investment experts, have experience that is crucial for data analysis and modeling.

[0003] Therefore, it is necessary to provide an intelligent data modeling method to provide intelligent assistance to users during the data modeling process, reduce the demand for users' mathematical ability, and meet the data modeling needs of many business expert users. Summary of the Invention

[0004] The present application provides an intelligent data modeling method, apparatus, device and storage medium for dynamically detecting and summarizing information of the front-end technology stack to assist enterprises in adjusting the layout of the front-end technology stack.

[0005] In a first aspect, the present application provides an intelligent data modeling method, the method comprising:

[0006] Based on the preset data modeling interface, obtain modeling data and user portraits;

[0007] generating one or more data modeling scenarios based on the user profile;

[0008] Acquire a user's intended data modeling scenario from the modeling scenario, and generate a plurality of model nodes according to the intended data modeling scenario, wherein the plurality of model nodes include configuration relationships;

[0009] In response to a user operation, modifying the configuration relationship of the model nodes;

[0010] The calculation parameters corresponding to the model nodes are obtained, modeling is performed according to the calculation parameters and the configuration relationship, and a data model of the modeling data is generated.

[0011] In a second aspect, the present application further provides an intelligent data modeling device, the intelligent data modeling device comprising: a data receiving module, a scenario building module, a node generating module, a node modifying module and a model generating module;

[0012] The data receiving module is used to obtain modeling data and user portraits based on the preset data modeling interface;

[0013] A scenario construction module, configured to generate one or more data modeling scenarios based on the user profile;

[0014] A node generation model is used to obtain the user's intention data modeling scenario from the modeling scenario, and generate a plurality of model nodes according to the intention data modeling scenario, wherein the plurality of model nodes include configuration relationships;

[0015] A node modification module, configured to modify the configuration relationship of the model nodes in response to a user operation;

[0016] The model generation module is used to obtain the calculation parameters corresponding to the model nodes, perform modeling according to the calculation parameters and the configuration relationship, and generate a data model of the modeling data.

[0017] In a third aspect, the present application also provides a computer device, comprising a memory and a processor; the memory is used to store a computer program; the processor is used to execute the computer program and implement any one of the intelligent data modeling methods provided in the embodiments of the present application when executing the computer program.

[0018] In a fourth aspect, the present application further provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, the processor implements any one of the intelligent data modeling methods provided in the embodiments of the present application.

[0019] The embodiments of the present application provide an intelligent data modeling method, apparatus, device and storage medium. The specific steps of the intelligent data modeling method include: obtaining modeling data and user portraits based on a preset data modeling interface; generating one or more data modeling scenarios based on the user portraits; obtaining the user's intended data modeling scenario from the modeling scenario, generating multiple model nodes based on the intended data modeling scenario, and the multiple model nodes including configuration relationships; modifying the configuration relationships of the model nodes in response to user operations; obtaining the calculation parameters corresponding to the model nodes, modeling according to the calculation parameters and configuration relationships, and generating a data model of the modeled data. According to the intelligent data modeling method provided in the embodiments of the present application, data modeling scenarios are recommended through user portraits, and the calculation parameters and configuration relationships of the model nodes are corrected based on business experience, which can realize the intelligent establishment of data modeling logic and the acquisition of mathematical parameters, and provide a simple and efficient data modeling tool for business experts. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following is a brief introduction to the drawings required for use in the description of the embodiments. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0021] Figure 1 This is a schematic diagram of a data modeling scenario provided by an embodiment of the present application;

[0022] Figure 2 is a schematic diagram of a data modeling interface provided in an embodiment of the present application;

[0023] Figure 3 is a schematic flow chart of an intelligent data modeling method provided in an embodiment of the present application;

[0024] Figure 4 This is a schematic diagram of a model node configuration interface provided in an embodiment of the present application;

[0025] Figure 5 This is a schematic flowchart of an intelligent parameter adjustment provided in an embodiment of the present application;

[0026] Figure 6 is a schematic block diagram of an intelligent data modeling device provided in an embodiment of the present application;

[0027] Figure 7 This is a schematic block diagram of the structure of a computer device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0028] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are part of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0029] The flowcharts shown in the accompanying drawings are for illustrative purposes only and do not necessarily include all contents and operations / steps, nor must they be executed in the order described. For example, some operations / steps may be decomposed, combined, or partially merged, so the actual execution order may vary depending on the actual situation.

[0030] It should be understood that the terms used in this specification are only for the purpose of describing specific embodiments and are not intended to limit the present application. As used in this specification and the appended claims, the singular forms "a", "an", and "the" are intended to include the plural forms unless the context clearly indicates otherwise.

[0031] It will also be understood that the term "and / or" used in this specification and the appended claims refers to and includes any and all possible combinations of one or more of the associated listed items.

[0032] The following describes some embodiments of the present application in detail with reference to the accompanying drawings. In the absence of conflict, the following embodiments and features therein may be combined with each other.

[0033] In response to the above technical problems, the present application provides an intelligent data modeling method, apparatus, computing device and storage medium for intelligently assisting users in the data modeling process to meet the data modeling needs of many business expert users.

[0034] See also Figure 1 , Figure 1 A schematic diagram of a data modeling scenario is shown. Figure 1 As shown, the data modeling process typically includes a terminal and a server. The terminal is used to respond to the user's data modeling operations and send data modeling requests to the server. The server is used to receive and respond to data modeling requests, call software and hardware resources to implement data modeling, obtain data modeling results, and send the data modeling results to the terminal. The terminal is also used to display the data modeling results to the user.

[0035] The server can be a standalone server or a server cluster, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery networks (CDNs), and big data and artificial intelligence platforms. The terminal can be an electronic device such as a mobile phone, tablet computer, laptop computer, desktop computer, personal digital assistant, or wearable device.

[0036] See also Figure 2 , Figure 2 A schematic diagram of a data modeling interface is shown. Figure 2 As shown in the figure, the data modeling interface consists of a standard component area, a shared component area, an operation document area, a user login area, and a modeling area. The standard component area includes three primary input windows: the modeling data import window, the user profile acquisition window, and the common toolbar window. The common toolbar acquisition window includes five secondary windows: the statistical analysis window, the feature engineering window, the model training window, the model prediction window, and the data export window. The shared component area is used to store data modeling components shared by project team members. The operation document area is used to introduce the functions included in the data modeling interface. The user login area is used to obtain user authentication information.

[0037] It should be noted that embodiments of the present application can acquire and process relevant data based on artificial intelligence technology, such as accessing software and hardware resources to implement data modeling and obtain data modeling results. Artificial Intelligence (AI) is the theory, method, technology, and application system that uses digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use knowledge to achieve optimal results.

[0038] Fundamental AI technologies generally include sensors, dedicated AI chips, cloud computing, distributed storage, big data processing, operating / interaction systems, and mechatronics. AI software technologies primarily encompass computer vision, robotics, biometrics, speech processing, natural language processing, and machine learning / deep learning.

[0039] See also Figure 3 , Figure 3 This is a schematic flow chart of an intelligent data modeling method provided by an embodiment of the present application. The intelligent data modeling method can dynamically detect and summarize change information of the front-end technology stack, assisting enterprises in adjusting the layout of the front-end technology stack.

[0040] like Figure 3As shown, the intelligent data modeling method specifically includes: steps S101 to S105.

[0041] S101. Provide a data modeling interface, and obtain modeling data and user portraits based on the data modeling interface.

[0042] Specifically, based on a preset data modeling interface, modeling data is imported; a historical user portrait of the user is obtained, or an input window is displayed to obtain a target user portrait input by the user; and the historical user portrait or the target user portrait is set as the current user portrait.

[0043] For example, a user logs in to a computer and uses a browser installed on the computer to access a web interface (i.e., a data modeling interface) provided by the cloud server. The user can perform an operation to obtain modeling data on the data modeling interface. The computer responds to the operation and obtains the corresponding modeling data from the cloud server and displays it on the interface, thereby obtaining the data used for modeling.

[0044] In some embodiments, when a user logs into the data modeling interface, user identity authentication information is obtained and the user is authenticated. User identity authentication information includes user identity information and user level information. Access history information and usage history information can be obtained based on the user identity information. User level information includes: beginner user, advanced user, and senior user.

[0045] In some embodiments, based on the user identity authentication information, it is determined whether the user is a primary user logging in for the first time. If the user is a primary user, a user portrait acquisition window is displayed on the data modeling page, and the user portrait is obtained through the user portrait acquisition window. In addition to manual input by the user, the user portrait acquisition window can also be manually selected. Click each user portrait acquisition window to obtain the corresponding drop-down interface for the user to manually select. Exemplarily, the window can be a selection window with a tree diagram structure. The user can organize the user portrait by clicking the options in the input box according to his or her own modeling needs. For example, the first tree selection chain of "Modeling Field-Modeling Business-Modeling Product" can also be the second tree selection chain of "Team Organization-Common Project-Personal Project".

[0046] In some embodiments, if the user is an advanced user or a senior user, the user portrait is determined based on the user identity authentication information. Specifically, if the user is an advanced user or a senior user, the user's usage record information is obtained, the historical user portrait is obtained from the usage record information, and the current user portrait is generated based on the historical user portrait.

[0047] For example, the user's usage history is obtained, and based on this history, information such as the user's "last login time," "last run task," "whether there is a valid model," "whether there is an unfinished model," and "community contribution status" is obtained. Based on this information, a historical user profile is obtained, and this historical profile is set as the current profile.

[0048] In some embodiments, when a novice user logs in to the data modeling page, a data modeling interface function introduction and process guidance are provided; when an advanced user or a senior user logs in to the data modeling page, community contribution status information prompts and new function information prompts are provided.

[0049] S102: Generate one or more data modeling scenarios based on the user portrait.

[0050] Specifically, when a user creates a data modeling task in the data modeling interface, one or more corresponding data modeling scenarios are generated according to the user portrait, and the modeling scenarios are displayed in the data modeling interface to facilitate user selection.

[0051] In some embodiments, data modeling scenarios include: insurance policy scenarios, reporting scenarios, quotation scenarios, survey scenarios, pricing scenarios, loss assessment scenarios, renewal scenarios, underwriting scenarios, payment scenarios, actuarial scenarios, correction scenarios, payment scenarios, marketing scenarios, underwriting risk control scenarios, claims risk control scenarios, operational risk control scenarios, customer operation scenarios, full claims process scenarios, and scenarios not involving specific business.

[0052] In some embodiments, if the primary user organizes the user portrait through the second tree-like selection chain and generates a data modeling scenario based on the user portrait, it also includes providing an authorization authentication window to obtain user authorization authentication information, and unlocking the corresponding team data based on the user authorization authentication information. The team data includes: data modeling scenarios shared by team members, data modeling scenarios dedicated to each project team, and data modeling scenarios for personal projects.

[0053] In some embodiments, data modeling scenarios such as "reference model", "custom modeling" and "code modeling" are provided to the advanced user based on the user profile of the advanced user.

[0054] S103: Acquire the user's intended data modeling scenario from the modeling scenario, and generate a plurality of model nodes according to the intended data modeling scenario, wherein the plurality of model nodes include configuration relationships.

[0055] Specifically, in response to a user's selection operation, the data modeling scenario selected by the user is set as the intended data modeling scenario, schematic labels of a plurality of model nodes are generated according to the intended data modeling scenario, and schematic connectors of the configuration relationships are generated; a model node configuration interface is displayed, and the schematic labels and schematic connectors are displayed on the model node configuration interface. The schematic connectors include: a data output symbol, a data input symbol, a data flow schematic line, a data sharing symbol, and a sub-model data symbol; the data sharing symbol is used to indicate that different model nodes are allowed to share parameters; the sub-model data symbol is used to indicate data generated by a machine learning model; the data input symbol and the corresponding data output symbol are connected via the data flow schematic line.

[0056] Exemplarily, a user selects a data modeling scenario through a terminal. In response to the user's operation, the terminal retrieves the schematic labels and schematic connectors corresponding to the data modeling scenario from a server. The schematic labels represent multiple preset model nodes, and the schematic connectors represent the configuration relationships between the model nodes. These relationships are displayed through a model node configuration interface, which is a sub-interface of the data modeling interface. If a data interaction relationship exists between any two model nodes, their corresponding schematic labels are connected using a data flow schematic line. The data flow schematic line can be a solid line with an arrow, where the arrow points in the direction of data flow.

[0057] See also Figure 4 , Figure 4 A schematic diagram of a model node configuration interface is shown, Figure 4 As shown, the model node configuration interface includes various schematic labels and schematic connectors. Each schematic label represents a model node, including: feature binning, discrete feature encoding, machine learning model, model prediction and evaluation, parameter sharing, and model prediction. Schematic connectors include: D symbol, P symbol, M symbol, and solid lines with arrows. The D symbol includes data input and data output symbols. The D symbol pointed to by the solid arrow is the data input symbol, and the end connected by the solid line is the data output symbol. The P symbol is the data sharing symbol, and the M symbol is the sub-model data symbol.

[0058] In some embodiments, if it is detected that the user is a premium user and it is detected that the user's most recent usage record information includes an unfinished model node configuration interface, the user is preferentially recommended to jump to the unfinished model node configuration interface.

[0059] In some embodiments, before generating multiple model nodes according to the intended data modeling scenario, it also includes: obtaining the user's usage record information, obtaining all the user's model node configuration schemes and corresponding data modeling scenarios from the usage record information, if it is detected that the user uses the same model node configuration scheme for a data modeling scenario more than once, saving the model node configuration scheme corresponding to the data modeling scenario to the server, and when the user selects the data modeling scenario again, retrieving the model node configuration scheme based on the user's usage record from the server.

[0060] It should be noted that in the embodiment of the present application, the model node is a process node for processing data. Commonly used model nodes include: data import node, statistical analysis node, target setting node, data verification node, data preprocessing node, feature engineering node, model training node, model evaluation node, model prediction node and data export node.

[0061] In some embodiments, the commonly used model nodes are encapsulated, arranged and stored in sequence, and visualized to reside in the toolbar area of ​​the data modeling interface for easy access.

[0062] In some embodiments, while obtaining the model node configuration scheme, the introduction information of the model node configuration scheme is also obtained, and the introduction information is displayed in the information display window of the data modeling interface, so that the user can understand the modeling background, modeling goals and modeling ideas, thereby improving the user's modeling speed.

[0063] S104: In response to the user's operation, modify the configuration relationship of the model nodes.

[0064] Before responding to the user's operation, the configuration relationship of each model node is checked, and when at least one of the preceding or succeeding model nodes and the connection relationship between the model nodes is missing or wrong, an invalid configuration prompt is output.

[0065] Specifically, in combination with the above-mentioned invalid configuration prompt, optimization prompt information of the configuration relationship is output; in response to the user's operation, the newly added model node is obtained, or the model node is deleted, or the data connection relationship between the model nodes is adjusted.

[0066] For example, a user can use a computer cursor to move the schematic labels of model nodes in the model node configuration interface, for example, to delete or add schematic labels, and can also adjust schematic connectors, for example, to add data output symbols, data input symbols, and data flow schematic lines between two schematic labels. In response to the user's modification operations, the computer displays the modified model nodes and their configuration relationships through the model node configuration interface. The server uses the hardware and software resources required to verify the modified model nodes and their configuration relationships, and outputs optimization prompts for any invalid configuration prompts.

[0067] In some embodiments, the prompt information includes graphic prompts and text prompts. For example, if the current model node lacks a preceding or succeeding model node, the schematic label of the current model node can be marked with a warning color. The warning color can be red, indicating that the current model node will inevitably have errors during operation. At the same time, it also includes generating a text prompt. When the user places the computer cursor on the model node with an error, a text box is displayed, and the text prompt is displayed on the text box. The text prompt can be "Model node A depends on model node B" and "Model node C has not yet configured a predecessor node."

[0068] In some embodiments, when the preceding node is not effectively set / connected, the system intelligently recommends the necessary preceding nodes or possible causes of the error (the recommended content is derived from business modeling precipitation and is recommended based on rules).

[0069] S105: Obtain calculation parameters corresponding to the model nodes, perform modeling according to the calculation parameters and the configuration relationship, and generate a data model of the modeling data.

[0070] Before obtaining the calculation parameters corresponding to the model nodes, the preferred calculation parameters of each model node are recommended based on the available calculation resources and the modeling data; a parameter recommendation interface is displayed, and the preferred calculation parameters are displayed on the parameter recommendation interface.

[0071] Exemplarily, a model framework of the current project is generated based on the model nodes and their configuration relationships, and the model framework is stress-tested according to the computing resources and data size that can be called by the current project. Optionally, during the stress test, the computing load of each model node is set to the maximum. If the test process causes the server's running memory to overflow, the test is terminated and a warning prompt is generated, and a first set of recommended computing parameters is generated based on business experience and test results. The user can modify the first set of recommended computing parameters, then input the modified computing parameters into the model framework, and retest until the running memory required by the model framework is less than the preset running memory, for example, the preset running memory is set to 90% of the running memory of the callable computing resources.

[0072] In some embodiments, when the running memory of the model architecture is less than the preset running memory, it is also necessary to estimate the computing time of the model architecture. When the computing time is greater than the preset value, for example, greater than 12 hours, a prompt message is generated and displayed in a prompt box to remind the user that the running time of the data model will exceed the preset duration, and to prompt the user to optimize the calculation parameters.

[0073] See also Figure 5 , Figure 5 A schematic flow chart of intelligent parameter adjustment is shown. Figure 5 As shown, when a user modifies the configuration relationship of a model node, the server obtains the modified configuration relationship and verifies whether the modified configuration relationship will inevitably cause a calculation error. If it will inevitably cause a calculation error, the server terminates the user operation and provides a prompt. If it will not inevitably cause a calculation error, the server verifies whether the modified configuration relationship is reasonable or dangerous during the calculation process. If the result is negative, the operation is executed normally; if the result is positive, the user manually confirms the configuration relationship, and based on the manual confirmation result, it determines whether the configuration relationship can be executed.

[0074] In some embodiments, model nodes can be divided into data processing nodes and model parameter adjustment nodes based on their functions. The computational parameters for data processing nodes include: optimal processing algorithms and data feature construction schemes; the computational parameters for model parameter adjustment nodes include: optimal computation sub-models, sub-model parameter search schemes, and sub-model optimal parameters.

[0075] In some embodiments, data analysis can encounter numerous missing values. These can arise for various reasons, including intentional omission for privacy reasons, the lack of numerical values ​​in variables, and improper data merging that results in missing values. Therefore, before inputting into the model, the modeling data must be processed for missing values ​​to generate an optimized solution.

[0076] Exemplarily, the optimization scheme for missing value processing includes: data verification, data analysis, data preprocessing and feature engineering.

[0077] Data verification is used to intelligently determine operational anomalies and select the optimal processing method by default.

[0078] Data analysis is used to display features in order of significance, highlighting high-value data.

[0079] Data preprocessing is used to intelligently recommend the optimal solution, including but not limited to field screening, sampling, deduplication, replacement, grouping, and partitioning.

[0080] Feature engineering is used to recommend common feature construction solutions and provide real-time reminders for exception handling.

[0081] In some embodiments, data standardization is required to improve the processing speed of the data model. The purpose of data standardization is to index data of different properties and magnitudes and adjust them to a comparable range. For example, when building a logistic regression model, gender values ​​are 0 or 1, but income values ​​may range from 0 to 1 million, a large span that requires standardization. Generally, the best / maximum standardization (Min-Max standardization) method can be used to set the values ​​between 0 and 1 for ease of calculation.

[0082] The intelligent data modeling method provided in the above embodiment recommends data modeling scenarios through user portraits, corrects the calculation parameters and configuration relationships of model nodes based on business experience, realizes the intelligent establishment of data modeling logic and acquisition of mathematical parameters, and provides business experts with a simple and efficient modeling tool.

[0083] See also Figure 6 , Figure 6 The embodiment of the present application further provides a schematic block diagram of an intelligent data modeling device, wherein the intelligent data modeling device 300 is used to execute the aforementioned intelligent data modeling method. The intelligent data modeling device can be configured in a server or a terminal.

[0084] The server can be a standalone server or a server cluster, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery networks (CDNs), and big data and artificial intelligence platforms. The terminal can be an electronic device such as a mobile phone, tablet computer, laptop computer, desktop computer, personal digital assistant, or wearable device.

[0085] like Figure 6 As shown, the intelligent data modeling device 300 includes: a data receiving module 301, a scenario construction module 302, a node generation model 303, a node modification module 304 and a model generation module 305.

[0086] The data receiving module 301 is used to obtain modeling data and user portraits based on a preset data modeling interface.

[0087] In some embodiments, the data receiving module 301 is specifically used to: import the modeling data based on a preset data modeling interface; obtain the user's historical user portrait, or display an input window to obtain the target user portrait input by the user; set the historical user portrait or the target user portrait as the current user portrait.

[0088] The scenario construction module 302 is used to generate one or more data modeling scenarios based on the user portrait.

[0089] The node generation model 303 is used to obtain the user's intention data modeling scenario from the modeling scenario, and generate multiple model nodes according to the intention data modeling scenario, wherein the multiple model nodes include configuration relationships.

[0090] In some embodiments, the node generation model 303 is specifically used to: generate schematic labels for multiple model nodes, and generate schematic connectors for the configuration relationships; display a model node configuration interface, and display the schematic labels and the schematic connectors on the model node configuration interface.

[0091] The node modification module 304 is used to modify the configuration relationship of the model nodes in response to user operations.

[0092] In some embodiments, the node modification module 304 is further used to: check the configuration relationship of each of the model nodes, and output an invalid configuration prompt when at least one of the preceding or succeeding model nodes and the connection relationship between the model nodes is missing.

[0093] In some embodiments, the node modification module 304 is specifically used to: output optimization prompt information of the configuration relationship; obtain the newly added model node, or delete the model node, or adjust the data connection relationship between the model nodes in response to the user's operation.

[0094] The model generation module 305 is used to obtain the calculation parameters corresponding to the model nodes, perform modeling according to the calculation parameters and the configuration relationship, and generate a data model of the modeling data.

[0095] In some embodiments, the model generation module 305 is further used to: recommend preferred computing parameters for each of the model nodes based on the available computing resources and the modeling data; display a parameter recommendation interface, and display the preferred computing parameters on the parameter recommendation interface.

[0096] It should be noted that technical personnel in the relevant field can clearly understand that for the convenience and conciseness of description, the specific working processes of the intelligent data modeling device and each module described above can refer to the corresponding processes in the aforementioned intelligent data modeling method embodiment, and will not be repeated here.

[0097] The above-mentioned intelligent data modeling device can be implemented in the form of a computer program. The computer program can be used in Figure 7 Runs on the computer equipment shown.

[0098] See also Figure 7 , Figure 7This is a schematic block diagram of the structure of a computer device provided in an embodiment of the present application. The computer device may be a server or a terminal.

[0099] See Figure 7 The computer device includes a processor, a memory and a network interface connected through a system bus, wherein the memory may include a storage medium and an internal memory.

[0100] The storage medium may store an operating system and a computer program. The computer program includes program instructions, which, when executed, may cause a processor to perform any one of the intelligent data modeling methods.

[0101] The processor is used to provide computing and control capabilities and support the operation of the entire computer equipment.

[0102] The internal memory provides an environment for the execution of a computer program stored in a storage medium. When executed by a processor, the computer program enables the processor to perform any of the intelligent data modeling methods. The storage medium can be either non-volatile or volatile.

[0103] The network interface is used for network communication, such as sending assigned tasks, etc. Those skilled in the art will understand that Figure 7 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than shown in the figure, or combine certain components, or have a different component arrangement.

[0104] It should be understood that the processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor, etc.

[0105] Exemplarily, in one embodiment, the processor is configured to execute a computer program stored in the memory to implement the following steps:

[0106] Based on the preset data modeling interface, obtain modeling data and user portraits;

[0107] generating one or more data modeling scenarios based on the user profile;

[0108] Acquire a user's intended data modeling scenario from the modeling scenario, and generate a plurality of model nodes according to the intended data modeling scenario, wherein the plurality of model nodes include configuration relationships;

[0109] In response to a user operation, modifying the configuration relationship of the model nodes;

[0110] The calculation parameters corresponding to the model nodes are obtained, modeling is performed according to the calculation parameters and the configuration relationship, and a data model of the modeling data is generated.

[0111] In some embodiments, when the processor generates multiple model nodes according to the intended data modeling scenario, and the multiple model nodes include configuration relationships, the processor is further specifically configured to implement:

[0112] Generate schematic labels for a plurality of the model nodes and generate schematic connectors for the configuration relationships; display a model node configuration interface, and display the schematic labels and the schematic connectors on the model node configuration interface.

[0113] In some embodiments, before implementing the modification of the configuration relationship of the model nodes in response to the user's operation, the processor is further configured to implement:

[0114] The configuration relationship of each of the model nodes is checked, and when at least one of the preceding or succeeding model nodes is missing and the connection relationship between the model nodes is wrong, an invalid configuration prompt is output.

[0115] In some embodiments, when the processor implements modifying the configuration relationship of the model nodes in response to a user operation, the processor is further configured to implement:

[0116] Outputting optimization prompt information of the configuration relationship;

[0117] In response to the user's operation, the newly added model nodes are acquired, or the model nodes are deleted, or the data connection relationship between the model nodes is adjusted.

[0118] In some embodiments, before obtaining the calculation parameters corresponding to the model nodes, the processor is further configured to implement:

[0119] Recommending optimal computing parameters for each of the model nodes based on available computing resources and the modeling data;

[0120] A parameter recommendation interface is displayed, and the preferred calculation parameters are displayed on the parameter recommendation interface.

[0121] In some embodiments, when implementing a preset data modeling interface and obtaining modeling data and user portraits, the processor is further configured to implement:

[0122] Importing the modeling data based on a preset data modeling interface;

[0123] Get the user's historical user profile, or display an input window to get the target user profile entered by the user;

[0124] The historical user profile or the target user profile is set as the current user profile.

[0125] A computer-readable storage medium is also provided in an embodiment of the present application. The computer-readable storage medium stores a computer program, and the computer program includes program instructions. The processor executes the program instructions to implement any one of the intelligent data modeling methods provided in the embodiments of the present application.

[0126] The computer-readable storage medium may be an internal storage unit of the computer device described in the aforementioned embodiment, such as a hard disk or memory of the computer device. The computer-readable storage medium may also be an external storage device of the computer device, such as a plug-in hard disk, a SmartMedia Card (SMC), a Secure Digital (SD) card, a flash memory card, etc. equipped on the computer device.

[0127] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in the present application, and such modifications or substitutions should be included in the scope of protection of the present application. Therefore, the scope of protection of the present application should be based on the scope of protection of the claims.

Claims

1. An intelligent data modeling method, characterized in that: The method comprises: Based on the preset data modeling interface, obtain modeling data and user portraits; generating one or more data modeling scenarios based on the user profile; Acquire the user's intended data modeling scenario from the modeling scenario, generate a plurality of model nodes according to the intended data modeling scenario, generate a plurality of schematic labels for the model nodes, and generate schematic connectors for the configuration relationships of the model nodes, display a model node configuration interface, and display the schematic labels and the schematic connectors on the model node configuration interface, wherein the schematic connectors include: a data output symbol, a data input symbol, a data flow schematic line, a data sharing symbol, and a sub-model data symbol; the data sharing symbol is used to indicate that different model nodes are allowed to share parameters; the sub-model data symbol is used to indicate data generated by a machine learning model; the data input symbol and the corresponding data output symbol are connected via the data flow schematic line; In response to the user's operation, modifying the configuration relationship of the model nodes; The calculation parameters corresponding to the model nodes are obtained, modeling is performed according to the calculation parameters and the configuration relationship, and a data model of the modeling data is generated.

2. The method according to claim 1, characterized in that Before modifying the configuration relationship of the model nodes in response to the user's operation, the method includes: The configuration relationship of each of the model nodes is checked, and when at least one of the preceding or succeeding model nodes is missing and the connection relationship between the model nodes is wrong, an invalid configuration prompt is output.

3. The method according to claim 1, characterized in that The modifying the configuration relationship of the model nodes in response to the user's operation includes: Outputting optimization prompt information of the configuration relationship; In response to the user's operation, the newly added model nodes are acquired, or the model nodes are deleted, or the data connection relationship between the model nodes is adjusted.

4. The method according to claim 1, wherein Before obtaining the calculation parameters corresponding to the model nodes, the method includes: Recommending optimal computing parameters for each of the model nodes based on available computing resources and the modeling data; A parameter recommendation interface is displayed, and the preferred calculation parameters are displayed on the parameter recommendation interface.

5. The method according to claim 1, wherein The method of obtaining modeling data and user portraits based on a preset data modeling interface includes: Based on a preset data modeling interface, obtaining the modeling data imported by the user; Obtain a historical user portrait of the user, or obtain a target user portrait input by the user, wherein the historical user portrait or the target user portrait is used as the user portrait of the user.

6. An intelligent data modeling device, characterized in that: include: The data receiving module is used to obtain modeling data and user portraits based on the preset data modeling interface; A scenario construction module, configured to generate one or more data modeling scenarios based on the user profile; A node generation model is used to obtain the user's intended data modeling scenario from the modeling scenario, generate multiple model nodes according to the intended data modeling scenario, generate multiple schematic labels for the model nodes, and generate schematic connectors for the configuration relationships of the model nodes, display a model node configuration interface, and display the schematic labels and the schematic connectors on the model node configuration interface, wherein the schematic connectors include: a data output symbol, a data input symbol, a data flow schematic line, a data sharing symbol, and a sub-model data symbol; the data sharing symbol is used to indicate that different model nodes are allowed to share parameters; the sub-model data symbol is used to represent data generated by a machine learning model; the data input symbol and the corresponding data output symbol are connected via the data flow schematic line; A node modification module, configured to modify the configuration relationship of the model nodes in response to a user operation; The model generation module is used to obtain the calculation parameters corresponding to the model nodes, perform modeling according to the calculation parameters and the configuration relationship, and generate a data model of the modeling data.

7. A computer device, characterized in that: The computer device includes a memory and a processor; The memory is used to store computer programs; The processor is configured to execute the computer program and implement the intelligent data modeling method according to any one of claims 1 to 5 when executing the computer program.

8. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, which, when executed by a processor, enables the processor to implement the intelligent data modeling method according to any one of claims 1 to 5.

Images