WO2019061991A1

WO2019061991A1 - Multi-element universal model platform modeling method, electronic device, and computer readable storage medium

Info

Publication number: WO2019061991A1
Application number: PCT/CN2018/076178
Authority: WO
Inventors: 安欣; 王建明; 肖京
Original assignee: 平安科技（深圳）有限公司
Priority date: 2017-09-30
Filing date: 2018-02-10
Publication date: 2019-04-04
Also published as: CN107844634A; CN107844634B

Abstract

A multi-element universal model platform modeling method. The method comprises steps of: establishing a universal data processing process, accessing a universal data format, and establishing, according to the universal data processing process and the universal data format, a sample database and a feature database for pre-stored user information (S31); establishing a model algorithm database for processing multi-element scene services and providing the model algorithm database for a preset configuration file to invoke (S32); and receiving a modeling instruction of a particular scene service, invoking, by means of the configuration file, a model algorithm corresponding to the particular scene service from the model algorithm database, and performing modeling by means of the model algorithm corresponding to the particular scene service according to user information, in accordance with the scene service, in the sample database and the feature database (S33). The method can quickly implement modeling by utilizing a multi-element universal model platform, thereby achieving accumulation of resource parameters and model parameters and outputting periodical analysis results.

Description

Multi-purpose universal model platform modeling method, electronic device and computer readable storage medium

The present application claims priority to the Chinese Patent Application filed on Sep. 30, 2017, the Chinese Patent Application No. 201710940520.4, entitled "Multi-Universal Model Platform Modeling Method, Electronic Device, and Computer-Readable Storage Media". All content is incorporated by reference in the application.

Technical field

The present application relates to the field of computer information technology, and in particular, to a multi-purpose universal model platform modeling method, an electronic device, and a computer readable storage medium.

Background technique

At present, there is no universal solution for predicting each scenario service. In addition, the high learning cost of Spark big data leads to slow access speed and low code reuse rate of new project or new service modeling tasks. Therefore, the design method of the prior art is not reasonable enough and needs to be improved.

Summary of the invention

In view of this, the present application proposes a multivariate general model platform modeling method, an electronic device, and a computer readable storage medium, which can quickly access a modeling service by establishing a multivariate general model algorithm library, save costs, and use different types of models. The algorithm joins the unified model platform and can be expanded infinitely.

First, in order to achieve the above object, the present application provides an electronic device including a memory and a processor, where the memory stores a multi-purpose universal model platform modeling system operable on the processor, The multivariate universal model platform modeling system is implemented by the processor to implement the following steps:

Establish a general process of data processing, access a general format of data, and establish a sample library and a feature library for pre-stored user information according to the general process of data processing and a common format of data;

Establishing a model algorithm library for processing multi-scenario services, providing a preset configuration file call; and

Receiving a modeling instruction of the specific scenario service, calling, by the configuration file, a model algorithm corresponding to the specific scenario service from the model algorithm library, and passing the user information corresponding to the specific scenario service in the sample library and the feature database Model algorithms corresponding to specific scenario services are modeled.

Preferably, the general process of data processing includes: normalizing data to make the data conform to the usage specifications of the offline platform; cleaning the data with dirty data, removing null and outliers in the data; and converting the data format to offline The specific format used by the platform.

Preferably, the data common format is a narrow table format, and the sample library adopts a narrow table format;

The sample library includes user sample information required for modeling the multivariate general model platform, and the user sample information required for modeling is input into the sample library according to the data common format through the data processing general process; and

The feature library is a narrow table set conforming to the data common format, and the narrow table set stores user feature information.

Preferably, the multivariate universal model platform modeling system is further used to implement the following steps when executed by the processor:

Determining whether the data in the sample library and the feature database meets the business requirements; and

The determining whether the data in the sample library and the feature library meets the business requirements includes:

Determining whether a sample matching the service of the specific scenario is stored in the sample library, and determining whether the feature in the feature library conforms to the service of the specific scenario.

After the modeling is completed by the model algorithm corresponding to the specific scenario service, the model training and prediction are performed by using the user feature information in the feature database, and the user who meets the specific feature is extracted from the sample library, and the user is recommended to the extracted user. Specific scenario business.

In addition, to achieve the above object, the present application further provides a multi-purpose universal model platform modeling method, which is applied to an electronic device, and the method includes:

Preferably, the method further comprises the steps of:

Determining whether the data in the sample library and the feature library meets the business requirement, and determining whether the data in the sample library and the feature database meets the business requirement comprises: determining whether the sample in the sample library stores the service corresponding to the specific scenario, and determining the feature database Whether the characteristics are consistent with the business of the particular scenario; and

Further, in order to achieve the above object, the present application further provides a computer readable storage medium storing a multivariate general model platform modeling system, and the multivariate universal model platform modeling system may be at least one The processor executes to cause the at least one processor to perform the steps of the multivariate general model platform modeling method as described above.

Compared with the prior art, the electronic device, the multi-purpose universal model platform modeling method and the computer readable storage medium proposed by the present application can quickly access the modeling service by establishing a multi-purpose general model algorithm library, thereby saving costs; Different types of model algorithms are added to the unified model platform, which can be expanded infinitely. Users can quickly use the multi-purpose general model platform to model without the coding foundation or Spark big data foundation. Further, through the preset parameter configuration file, this application The accumulation of resource parameters and model parameters is realized, and the results of the phase analysis are output.

DRAWINGS

1 is a schematic diagram of an optional hardware architecture of an electronic device of the present application;

2 is a schematic diagram of a program module of an embodiment of a multi-purpose universal model platform modeling system in an electronic device of the present application;

FIG. 3 is a schematic diagram of an implementation process of an embodiment of a multi-purpose universal model platform modeling method according to the present application.

Reference mark:

电子设备 Electronic equipment	22
存储器Memory	21twenty one
处理器processor	22twenty two
网络接口Network Interface	23twenty three
多元通用模型平台建模系统Multivariate universal model platform modeling system	2020
第一创建模块 First creation module	201201
第二创建模块 Second creation module	202202
第三创建模块 Third creation module	203203
流程步骤Process step	S31-S33S31-S33

The implementation, functional features and advantages of the present application will be further described with reference to the accompanying drawings.

Detailed ways

In order to make the objects, technical solutions, and advantages of the present application more comprehensible, the present application will be further described in detail below with reference to the accompanying drawings and embodiments. It is understood that the specific embodiments described herein are merely illustrative of the application and are not intended to be limiting. All other embodiments obtained by a person of ordinary skill in the art based on the embodiments of the present application without departing from the inventive scope are the scope of the present application.

It should be noted that the descriptions of "first", "second" and the like in the present application are for the purpose of description only, and are not to be construed as indicating or implying their relative importance or implicitly indicating the number of technical features indicated. . Thus, features defining "first" and "second" may include at least one of the features, either explicitly or implicitly. In addition, the technical solutions between the various embodiments may be combined with each other, but must be based on the realization of those skilled in the art, and when the combination of the technical solutions is contradictory or impossible to implement, it should be considered that the combination of the technical solutions does not exist. Nor is it within the scope of protection required by this application.

It is further to be understood that the term "comprises", "comprises" or any other variations thereof is intended to encompass a non-exclusive inclusion, such that a process, method, article, or device that comprises a And includes other elements not explicitly listed, or elements that are inherent to such a process, method, article, or device. An element that is defined by the phrase "comprising a ..." does not exclude the presence of additional equivalent elements in the process, method, item, or device that comprises the element.

First of all, the present application proposes an electronic device 2.

Referring to FIG. 1, it is a schematic diagram of an optional hardware architecture of the electronic device 2 of the present application. In this embodiment, the electronic device 2 may include, but is not limited to, a memory 21, a processor 22, and a network interface 23 that can communicate with each other through a system bus. It is pointed out that FIG. 1 only shows the electronic device 2 with the components 21-23, but it should be understood that not all illustrated components are required to be implemented, and more or fewer components may be implemented instead.

The electronic device 2 may be a computing device such as a rack server, a blade server, a tower server, or a rack server. The electronic device 2 may be an independent server or a server cluster composed of multiple servers. .

The memory 21 includes at least one type of readable storage medium including a flash memory, a hard disk, a multimedia card, a card type memory (eg, SD or DX memory, etc.), a random access memory (RAM), a static Random access memory (SRAM), read only memory (ROM), electrically erasable programmable read only memory (EEPROM), programmable read only memory (PROM), magnetic memory, magnetic disk, optical disk, and the like. In some embodiments, the memory 21 may be an internal storage unit of the electronic device 2, such as a hard disk or memory of the electronic device 2. In other embodiments, the memory 21 may also be an external storage device of the electronic device 2, such as a plug-in hard disk equipped on the electronic device 2, a smart memory card (SMC), and a secure digital device. (Secure Digital, SD) card, flash card, etc. Of course, the memory 21 may also include both an internal storage unit of the electronic device 2 and an external storage device thereof. In this embodiment, the memory 21 is generally used to store an operating system installed in the electronic device 2 and various types of application software, such as program code of the multi-purpose general model platform modeling system 20, and the like. Further, the memory 21 can also be used to temporarily store various types of data that have been output or are to be output.

The processor 22 may be a Central Processing Unit (CPU), controller, microcontroller, microprocessor, or other data processing chip in some embodiments. The processor 22 is typically used to control the overall operation of the electronic device 2, such as performing control and processing related to data interaction or communication with the electronic device 2. In this embodiment, the processor 22 is configured to run program code or process data stored in the memory 21, such as running the multi-purpose general model platform modeling system 20 and the like.

The network interface 23 may comprise a wireless network interface or a wired network interface, which is typically used to establish a communication connection between the electronic device 2 and other electronic devices. For example, the network interface 23 is configured to connect the electronic device 2 to an external data platform through a network, and establish a data transmission channel and a communication connection between the electronic device 2 and an external data platform. The network may be an intranet, an Internet, a Global System of Mobile communication (GSM), a Wideband Code Division Multiple Access (WCDMA), a 4G network, or a 5G network. Wireless or wired networks such as network, Bluetooth, Wi-Fi, etc.

So far, the application environment of the various embodiments of the present application and the hardware structure and functions of related devices have been described in detail. Hereinafter, various embodiments of the present application will be proposed based on the above-described application environment and related devices.

Referring to FIG. 2, it is a program module diagram of an embodiment of a multi-purpose universal model platform modeling system 20 in the electronic device 2 of the present application. In this embodiment, the multivariate universal model platform modeling system 20 may be divided into one or more program modules, and the one or more program modules are stored in the memory 21 and are composed of one or more The processor (which is the processor 22 in this embodiment) is executed to complete the application. For example, in FIG. 2, the multivariate universal model platform modeling system 20 can be segmented into a first creation module 201, a second creation module 202, and a third creation module 203. A program module as referred to in this application refers to a series of computer program instructions that are capable of performing a particular function, and are more suitable than the program to describe the execution of the multi-purpose model platform modeling system 20 in the electronic device 2. The function of each program module 201-203 will be described in detail below.

The first creating module 201 is configured to establish a general process of data processing, access a general format of data, and establish a sample library and a feature library for pre-stored user information according to the data processing general process and the data common format to implement different Feature sharing between scene services. The different scenario services may be an Internet scenario service and a financial scenario service.

Preferably, in the embodiment, the general process of data processing includes, but is not limited to, standardizing operations on data (such as user information), conforming the data to the usage specifications of the offline platform, and performing dirty data cleaning on the data to remove data. Null and outliers in ; and convert the data format to the specific format used by the offline platform. In this embodiment, the specific format used by the offline platform may adopt a narrow table format, such as a KV (Key-Value) key table format, that is, a table form of a KV database.

Preferably, in this embodiment, the data common format is a narrow table format. The tables currently in use are all in a wide table format, while the table data in a wide table format is poorly shared. For data sharing and convenient management, a general data format KV table (narrow table format) is used in this embodiment.

Preferably, in this embodiment, the sample library adopts a narrow table format (such as a KV table), including but not limited to: user sample information required for modeling the multi-purpose general model platform, such as a user label (Label), a user name. (User), and business type or item type (Item). In this embodiment, the user sample information (such as user tag, user name, and service type) required for modeling is input into the sample library according to the data common format by using the data processing general process. The user sample information may be Spark big data information.

Preferably, in this embodiment, the feature library is a narrow table set (such as a KV table set) that conforms to the data common format, and the narrow table set stores user feature information (such as a user portrait feature). It should be noted that some service scenarios cannot meet the modeling requirements by using the standard feature database. In this embodiment, specific user image features (such as age, height, gender, region, user preference, etc.) under different scenario services are generated for the multi-scenario scenario. For example, in an Internet scene service, if an Internet user portrait feature such as an advertisement image or a heartbeat feature is required, the user portrait feature under these Internet scene services can be set to a narrow table set and stored in the feature library for sharing.

The second creating module 202 is configured to establish a model algorithm library for processing the multi-scenario service, and provide the preset configuration file call, thereby implementing rapid modeling. The model algorithm library includes a model algorithm for processing services in different scenarios. Preferably, in this embodiment, the model algorithm library can access an LR (Logistic Regression) model, a GBD (Gradient Boosting Decision Tree) model, a random forest model, and a K-Means aggregation. Class algorithms, etc., to achieve multi-scene business processing functions, such as CTR (Click-Through-Rate) click prediction function in the Internet scenario, user group analysis function in financial scenarios, and other multi-personalized business analysis requirements.

The third creating module 203 is configured to receive a modeling instruction of a specific scenario service, and invoke, by using the configuration file (just modifying a parameter), a model algorithm corresponding to the specific scenario service from the model algorithm library, and according to the sample The user information of the library and the feature library that conforms to the service of the specific scenario (if the data in the sample library and the feature library meets the business requirements) is modeled by the model algorithm corresponding to the service of the specific scenario. The user information includes user sample information and user feature information.

Preferably, in this embodiment, by modifying parameters in the configuration file, such as the selected algorithm name, the sample table name used, and the feature table name, the user can quickly utilize the multi-generic universal without coding base or Spark big data foundation. Model platform modeling.

Preferably, in the embodiment, the third creating module 203 is further configured to: determine whether data in the sample library and the feature library meets a service requirement. The determining whether the data in the sample library and the feature database meets the business requirement comprises: determining whether a sample corresponding to the specific scenario service is stored in the sample library (such as a sample of CTR click prediction in an Internet scenario); Whether the feature conforms to the specific scenario service, for example, determining whether the feature image of the user, the heartbeat feature, etc., which meet the CTR click prediction service requirement in the Internet scenario are stored in the feature library.

For example, if the specific scenario service received is: analyze whether a particular user is willing to use a specific service (such as Ping An Life Insurance). The model algorithm corresponding to the service of the specific scenario in the model algorithm library is an LR model algorithm, and the LR model algorithm required to directly retrieve the model is used for modeling.

Further, in other embodiments, the third creating module 203 is further configured to:

Preferably, the model training refers to: learning new knowledge from characteristics of existing users, and establishing new features of the user. The model prediction means that, for an unknown user, the model predicts whether the unknown user will use the specific scenario service (such as life insurance business).

Preferably, the extracting the user that meets the specific feature from the sample library comprises: determining, by feature importance analysis, specific features related to the specific scenario service (ie, important features, such as age greater than 30 years old, user preference to purchase A record of insurance, or a telephone consultation with a life insurance business, etc., and a user having the particular feature is extracted from the sample library.

Analyze the effect of model prediction (such as analysis of recommended feedback rate and success rate) and feed the analysis results to other application interfaces.

Through the above program modules 201-203, the multivariate general model platform modeling system 20 proposed by the present application can quickly access the modeling service by establishing a multivariate general model algorithm library, thereby saving costs; in addition, adding different types of model algorithms to the unified The model platform can be expanded infinitely. The user can quickly use the multi-generic model platform to model without the coding foundation or the Spark big data foundation. Further, the resource parameters and model parameters are realized by the preset parameter configuration file. Accumulate and output the results of the phase analysis.

In addition, the present application also proposes a multivariate general model platform modeling method.

Referring to FIG. 3, it is a schematic diagram of an implementation process of an embodiment of a multi-purpose universal model platform modeling method of the present application. In this embodiment, the order of execution of the steps in the flowchart shown in FIG. 3 may be changed according to different requirements, and some steps may be omitted.

Step S31, establishing a general process of data processing, accessing a general format of data, and establishing a sample library and a feature library for pre-stored user information according to the general process of data processing and a common format of data, so as to realize feature sharing between services in different scenarios. . The different scenario services may be an Internet scenario service and a financial scenario service.

Step S32, establishing a model algorithm library for processing the multi-scenario service, and providing the preset configuration file call, thereby implementing rapid modeling. The model algorithm library includes a model algorithm for processing services in different scenarios. Preferably, in this embodiment, the model algorithm library can access an LR (Logistic Regression) model, a GBD (Gradient Boosting Decision Tree) model, a random forest model, and a K-Means aggregation. Class algorithms, etc., to achieve multi-scene business processing functions, such as CTR (Click-Through-Rate) click prediction function in the Internet scenario, user group analysis function in financial scenarios, and other multi-personalized business analysis requirements.

Step S33, receiving a modeling instruction of a specific scenario service, calling, by using the configuration file (just modifying the parameter), a model algorithm corresponding to the specific scenario service from the model algorithm library, and conforming to the sample library and the feature library according to the User information of a specific scenario service (if the data in the sample library and the feature library meets business requirements), and modeled by a model algorithm corresponding to the specific scenario service. The user information includes user sample information and user feature information.

Preferably, in this embodiment, the step of determining whether the data in the sample library and the feature library meets the service requirement is further included. The determining whether the data in the sample library and the feature database meets the business requirement comprises: determining whether a sample corresponding to the specific scenario service is stored in the sample library (such as a sample of CTR click prediction in an Internet scenario); Whether the feature conforms to the specific scenario service, for example, determining whether the feature image of the user, the heartbeat feature, etc., which meet the CTR click prediction service requirement in the Internet scenario are stored in the feature library.

For example, if the specific scenario service received is: analyze whether a particular user is willing to use a specific service (such as Ping An Life Insurance). The model algorithm corresponding to the service of the specific scene in the model algorithm library is an LR model algorithm, and the required LR model algorithm is directly used to perform modeling by using the configuration file.

Further, in other embodiments, the multivariate universal model platform modeling method further includes the following steps:

Through the above steps S31-S33 and other related steps, the multivariate general model platform modeling method proposed by the present application can quickly access the modeling service by establishing a multivariate general model algorithm library, and save costs; in addition, different types of model algorithms are used. Joining a unified model platform, the level can be expanded infinitely. The user can quickly use the multi-purpose general model platform to model without the coding foundation or Spark big data foundation. Further, through the preset parameter configuration file, the application implements the resource parameters and The accumulation of model parameters and the output of phased analysis results.

Further, in order to achieve the above object, the present application further provides a computer readable storage medium (such as a ROM/RAM, a magnetic disk, an optical disk), where the computer readable storage medium stores a multivariate universal model platform modeling system 20, The multivariate universal model platform modeling system 20 can be executed by at least one processor 22 to cause the at least one processor 22 to perform the steps of the multivariate general model platform modeling method as described above.

Through the description of the above embodiments, those skilled in the art can clearly understand that the foregoing embodiment method can be implemented by means of software plus a necessary general hardware platform, and can also be implemented by hardware, but in many cases, the former is A better implementation. Based on such understanding, the technical solution of the present application, which is essential or contributes to the prior art, may be embodied in the form of a software product stored in a storage medium (such as ROM/RAM, disk, The optical disc includes a number of instructions for causing a terminal device (which may be a mobile phone, a computer, a server, an air conditioner, or a network device, etc.) to perform the methods described in various embodiments of the present application.

The preferred embodiments of the present application have been described above with reference to the drawings, and are not intended to limit the scope of the application. The serial numbers of the embodiments of the present application are merely for the description, and do not represent the advantages and disadvantages of the embodiments. Additionally, although logical sequences are shown in the flowcharts, in some cases the steps shown or described may be performed in a different order than the ones described herein.

A person skilled in the art can implement the present application in various variants without departing from the scope and spirit of the present application. For example, the features of one embodiment can be used in another embodiment to obtain another embodiment. The equivalent structure or equivalent process transformations made by the present specification and the contents of the drawings, or directly or indirectly applied to other related technical fields, are all included in the scope of patent protection of the present application.

Claims

An electronic device, comprising: a memory and a processor, wherein the memory stores a multivariate universal model platform modeling system operable on the processor, the multivariate universal model platform modeling The system implements the following steps when executed by the processor:

Establish a general process of data processing, access a general format of data, and establish a sample library and a feature library for pre-stored user information according to the general process of data processing and a common format of data;

Establishing a model algorithm library for processing multi-scenario services, providing a preset configuration file call; and

Receiving a modeling instruction of the specific scenario service, calling, by the configuration file, a model algorithm corresponding to the specific scenario service from the model algorithm library, and passing the user information corresponding to the specific scenario service in the sample library and the feature database Model algorithms corresponding to specific scenario services are modeled.
The electronic device according to claim 1, wherein the general process of data processing comprises: normalizing data to make the data conform to the usage specification of the offline platform; performing dirty data cleaning on the data to remove null values in the data; And outliers; and the specific format used to convert the data format to an offline platform.
The electronic device according to claim 1, wherein the data common format is a narrow table format, and the sample library adopts a narrow table format;

The sample library includes user sample information required for modeling the multivariate general model platform, and the user sample information required for modeling is input into the sample library according to the data common format through the data processing general process; and

The feature library is a narrow table set conforming to the data common format, and the narrow table set stores user feature information.
The electronic device of claim 3, wherein the multivariate universal model platform modeling system is further configured to implement the following steps when executed by the processor:

Determining whether the data in the sample library and the feature database meets the business requirements; and

The determining whether the data in the sample library and the feature library meets the business requirements includes:

Determining whether a sample matching the service of the specific scenario is stored in the sample library, and determining whether the feature in the feature library conforms to the service of the specific scenario.
The electronic device according to claim 1, wherein the multivariate universal model platform modeling system is further used to implement the following steps when executed by the processor:

After the modeling is completed by the model algorithm corresponding to the specific scenario service, the model training and prediction are performed by using the user feature information in the feature database, and the user who meets the specific feature is extracted from the sample library, and the user is recommended to the extracted user. Specific scenario business.
The electronic device according to claim 2, wherein the multivariate universal model platform modeling system is further used to implement the following steps when executed by the processor:

After the modeling is completed by the model algorithm corresponding to the specific scenario service, the model training and prediction are performed by using the user feature information in the feature database, and the user who meets the specific feature is extracted from the sample library, and the user is recommended to the extracted user. Specific scenario business.
The electronic device according to claim 3, wherein the multivariate universal model platform modeling system is further used to implement the following steps when executed by the processor:

After the modeling is completed by the model algorithm corresponding to the specific scenario service, the model training and prediction are performed by using the user feature information in the feature database, and the user who meets the specific feature is extracted from the sample library, and the user is recommended to the extracted user. Specific scenario business.
The electronic device according to claim 4, wherein the multivariate universal model platform modeling system is further used to implement the following steps when executed by the processor:

After the modeling is completed by the model algorithm corresponding to the specific scenario service, the model training and prediction are performed by using the user feature information in the feature database, and the user who meets the specific feature is extracted from the sample library, and the user is recommended to the extracted user. Specific scenario business.
A multi-purpose universal model platform modeling method is applied to an electronic device, characterized in that the method comprises:

Establish a general process of data processing, access a general format of data, and establish a sample library and a feature library for pre-stored user information according to the general process of data processing and a common format of data;

Establishing a model algorithm library for processing multi-scenario services, providing a preset configuration file call; and

Receiving a modeling instruction of the specific scenario service, calling, by the configuration file, a model algorithm corresponding to the specific scenario service from the model algorithm library, and passing the user information corresponding to the specific scenario service in the sample library and the feature database Model algorithms corresponding to specific scenario services are modeled.
The multivariate general model platform modeling method according to claim 9, wherein the general process of data processing comprises: normalizing data to make the data conform to the usage specification of the offline platform; and cleaning the data by dirty data. Null and outliers in the data; and the specific format used to convert the data format to an offline platform.
The multivariate general model platform modeling method according to claim 9, wherein the data common format is a narrow table format, and the sample library adopts a narrow table format;

The sample library includes user sample information required for modeling the multivariate general model platform, and the user sample information required for modeling is input into the sample library according to the data common format through the data processing general process; and

The feature library is a narrow table set conforming to the data common format, and the narrow table set stores user feature information.
The method of modeling a multivariate universal model platform according to claim 11, wherein the method further comprises the steps of:

Determining whether the data in the sample library and the feature library meets the business requirement, and determining whether the data in the sample library and the feature database meets the business requirement comprises: determining whether the sample in the sample library stores the service corresponding to the specific scenario, and determining the feature database Whether the characteristics are consistent with the business of the particular scenario; and

After the modeling is completed by the model algorithm corresponding to the specific scenario service, the model training and prediction are performed by using the user feature information in the feature database, and the user who meets the specific feature is extracted from the sample library, and the user is recommended to the extracted user. Specific scenario business.
The multivariate general model platform modeling method according to claim 9, wherein the multivariate general model platform modeling system is further used to implement the following steps when executed by the processor:

After the modeling is completed by the model algorithm corresponding to the specific scenario service, the model training and prediction are performed by using the user feature information in the feature database, and the user who meets the specific feature is extracted from the sample library, and the user is recommended to the extracted user. Specific scenario business.
The multivariate general model platform modeling method according to any one of claims 10 to 12, wherein the multivariate universal model platform modeling system is further used to implement the following steps when executed by the processor:

After the modeling is completed by the model algorithm corresponding to the specific scenario service, the model training and prediction are performed by using the user feature information in the feature database, and the user who meets the specific feature is extracted from the sample library, and the user is recommended to the extracted user. Specific scenario business.
A computer readable storage medium storing a multivariate universal model platform modeling system executable by at least one processor, the multivariate universal model platform modeling The system implements the following steps when executed by the processor:

Establishing a general process of data processing, accessing a general format of data, and establishing a sample library and a feature library for pre-stored user information according to the general process of data processing and a common format of data;

Establishing a model algorithm library for processing multi-scenario services, providing a preset configuration file call; and

Receiving a modeling instruction of the specific scenario service, calling, by the configuration file, a model algorithm corresponding to the specific scenario service from the model algorithm library, and passing the user information corresponding to the specific scenario service in the sample library and the feature database Model algorithms corresponding to specific scenario services are modeled.
The computer readable storage medium according to claim 15, wherein the general process of data processing comprises: normalizing data to conform to data usage specifications of the offline platform; cleaning data for dirty data, and removing data. Null and outliers; and the specific format used to convert the data format to an offline platform.
The computer readable storage medium of claim 15 wherein said data common format is a narrow table format and said sample library is in a narrow table format;

The sample library includes user sample information required for modeling the multivariate general model platform, and the user sample information required for modeling is input into the sample library according to the data common format through the data processing general process; and

The feature library is a narrow table set conforming to the data common format, and the narrow table set stores user feature information.
The computer readable storage medium of claim 17, wherein the multivariate universal model platform modeling system is further configured to implement the following steps when executed by the processor:

Determining whether the data in the sample library and the feature database meets the business requirements; and

The determining whether the data in the sample library and the feature library meets the business requirements includes:

Determining whether a sample matching the service of the specific scenario is stored in the sample library, and determining whether the feature in the feature library conforms to the service of the specific scenario.
The computer readable storage medium of claim 15, wherein the multivariate universal model platform modeling system is further configured to implement the following steps when executed by the processor:

After the modeling is completed by the model algorithm corresponding to the specific scenario service, the model training and prediction are performed by using the user feature information in the feature database, and the user who meets the specific feature is extracted from the sample library, and the user is recommended to the extracted user. Specific scenario business.
The computer readable storage medium according to claim 16 or 17 or 18, wherein the multivariate universal model platform modeling system is further used to implement the following steps when executed by the processor:

After the modeling is completed by the model algorithm corresponding to the specific scenario service, the model training and prediction are performed by using the user feature information in the feature database, and the user who meets the specific feature is extracted from the sample library, and the user is recommended to the extracted user. Specific scenario business.