CN115964035A - Data mining model management method and device, electronic equipment and storage medium - Google Patents
Data mining model management method and device, electronic equipment and storage medium Download PDFInfo
- Publication number
- CN115964035A CN115964035A CN202211547081.8A CN202211547081A CN115964035A CN 115964035 A CN115964035 A CN 115964035A CN 202211547081 A CN202211547081 A CN 202211547081A CN 115964035 A CN115964035 A CN 115964035A
- Authority
- CN
- China
- Prior art keywords
- model
- instruction
- data
- data mining
- registered
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000007726 management method Methods 0.000 title claims abstract description 50
- 238000007418 data mining Methods 0.000 title claims abstract description 41
- 238000003860 storage Methods 0.000 title claims abstract description 25
- 238000000034 method Methods 0.000 claims abstract description 35
- 238000011022 operating instruction Methods 0.000 claims abstract description 7
- 230000005540 biological transmission Effects 0.000 claims description 15
- 230000008859 change Effects 0.000 claims description 13
- 238000012544 monitoring process Methods 0.000 claims description 11
- 238000004422 calculation algorithm Methods 0.000 claims description 7
- 230000005012 migration Effects 0.000 claims description 6
- 238000013508 migration Methods 0.000 claims description 6
- 238000004140 cleaning Methods 0.000 claims description 5
- 238000000638 solvent extraction Methods 0.000 claims description 5
- 238000004590 computer program Methods 0.000 claims description 3
- 238000011161 development Methods 0.000 abstract description 16
- 230000008569 process Effects 0.000 abstract description 15
- 238000010586 diagram Methods 0.000 description 7
- 230000000694 effects Effects 0.000 description 7
- 238000010801 machine learning Methods 0.000 description 6
- 238000005516 engineering process Methods 0.000 description 5
- 230000006870 function Effects 0.000 description 4
- 238000007689 inspection Methods 0.000 description 4
- 238000011156 evaluation Methods 0.000 description 3
- 238000002360 preparation method Methods 0.000 description 3
- 238000012549 training Methods 0.000 description 3
- 230000009849 deactivation Effects 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 230000004044 response Effects 0.000 description 2
- 230000002159 abnormal effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000013499 data model Methods 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000009191 jumping Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Images
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention relates to a data mining model management method, a data mining model management device, electronic equipment and a storage medium. The data mining model management method comprises the following steps: responding to a model deployment instruction, and generating and displaying a model deployment interface, wherein the model deployment interface comprises an online model name, an online state corresponding to the model, and an operation instruction button for the registered model; responding to an operation instruction of the registered model, generating an operation approval message, and sending the operation approval message to a corresponding decision maker; the operating instructions for the registered model include at least one of: online, updating and stopping; and obtaining an approval result of the decision maker, and executing the operation instruction when the approval result indicates that the operation instruction passes through the operation instruction. The data mining model management method provided by the invention provides a standardized modeling process and a workflow template, reduces the development threshold and improves the development efficiency.
Description
Technical Field
The invention relates to the technical field of computer big data, in particular to a data mining model management method and device, electronic equipment and a storage medium.
Background
Data mining generally refers to a technique for searching for regular and valuable information from massive data through algorithm search. Data mining can be regarded as intersection of machine learning and a database, and mainly utilizes a technology provided by the machine learning to analyze mass data and a technology provided by the database to manage the mass data. In other words, machine learning provides a practical solution to data mining.
The data mining lifecycle consists of six phases: business understanding, data preparation, modeling, model evaluation, and method implementation. The model full life cycle refers to the whole process from "development" to "online" to "offline" of the model, and mainly comprises the following steps: data preparation, feature engineering, algorithm realization, model development, model release, model monitoring, model optimization, model offline and the like.
Today, data-driven decision making, more and more enterprises recognize that data is an important strategic asset of an enterprise. How to mine valuable information from mass data, guide intelligent operation and production of enterprises, and support real-time, accurate, quick and quick decisions of enterprises becomes an urgent business requirement of the enterprises at present, particularly in the industries such as finance, telecommunication, network and the like. In order to meet the above requirements of enterprises, various model development platforms emerge.
In terms of technical application, most of the current model development platforms are machine learning platforms, and the process from data preparation to model release can be basically realized. However, these platforms lack monitoring and early warning of model effects, lack management approval of processes such as online and offline of models, and are disordered in model versions, so that enterprises are difficult to effectively manage models. In addition, these methods also lack guidance for standardized modeling procedures and guidance for parameter selection, so that the development threshold of machine learning models is still high.
Due to the lack of professional knowledge and skills of model development, business personnel often feel abnormal difficulty in the process of implementing machine learning, which is mainly reflected in the following aspects:
(1) The learning threshold is high, the codes are complex, visual operation interfaces are lacked, standardized flow guidance is lacked, and parameter selection guidance is lacked.
(2) The development efficiency is low, the knowledge and experience of model management are lacked, so that the models cannot be reused, and the period of newly developed models is long.
For enterprises, in developing data mining technology, the following challenges are usually faced:
(1) The development cost is high, the cost for training business personnel is high, and the calculation requirement of the server is high.
(2) The monitoring of the model is less, the development and the deployment of the model have gaps, the operation effect of the production model is lack of monitoring, and the model cannot be prompted to be retrained or retired.
(3) The model is difficult to manage, and the processes of development, registration, deployment, retirement and the like of the model are lack of examination and approval.
Disclosure of Invention
Based on this, the invention aims to provide a data mining model management method, a data mining model management device, an electronic device and a storage medium, which provide a standardized modeling process and a workflow template, reduce a development threshold and improve development efficiency.
In a first aspect, the present invention provides a data mining model management method, including the following steps:
responding to a model deployment instruction, generating and displaying a model deployment interface, wherein the model deployment interface comprises an online model name, an online state corresponding to the model, and an operation instruction button for the registered model;
responding to an operation instruction of the registered model, generating an operation approval message, and sending the operation approval message to a corresponding decision maker; the operating instructions on the registered model comprise at least one of the following: online, updating and stopping;
obtaining an approval result of a decision maker, and executing the operation instruction when the approval result indicates that the operation instruction passes through the operation instruction;
wherein the registration of the registered model comprises:
obtaining a model registration instruction, and generating a corresponding standard modeling flow;
acquiring a parameter adjusting instruction, and modifying parameters corresponding to the current model according to the parameter adjusting instruction;
acquiring an operation instruction, executing an operation corresponding to the operation instruction, and displaying an operation result on a canvas;
and obtaining a model storage instruction, and storing the current model and the corresponding parameters to the registered model.
Further, the execution instructions include at least one of:
data source file selection, data partitioning, data quality inspection, data cleaning, data binning, characteristic importance evaluation, variable collinearity inspection and algorithm realization.
Further, the method also comprises the following steps:
acquiring the system authority of a current user according to the login information of the user;
and generating a system interface of the user according to the system authority of the user.
Further, the method also comprises the following steps:
for the online model, acquiring a service monitoring instruction, and generating and displaying running data corresponding to the current model; the operational data includes at least one of: the promotion degree, the AUC value, the accuracy rate, the precision rate and the recall rate;
and generating an operation performance change report of the current model according to the historical change condition of the operation data.
Further, the model deployment interface displays the name, version, model state and update time of the current online deployment model in a table form.
Further, the method also comprises the following steps:
acquiring a timing task instruction for an online model;
and executing the timing task according to the timing scheduling information corresponding to the timing task instruction, and generating corresponding reminding information after execution.
Further, the timing task includes at least one of:
the model task is used for periodically updating the registered model;
a transmission task for providing a data migration service;
and the prediction task is used for providing automatic batch reasoning.
In a second aspect, the present invention further provides a data mining model management apparatus, including:
the model deployment interface display module is used for responding to a model deployment instruction and generating and displaying a model deployment interface, and the model deployment interface comprises an online model name, an online state corresponding to the model and an operation instruction button for the registered model;
the operation approval message generation module is used for responding to an operation instruction of the registered model, generating an operation approval message and sending the operation approval message to a corresponding decision maker; the operating instructions on the registered model comprise at least one of the following: online, updating and stopping;
the operation instruction execution module is used for acquiring the approval result of the decision maker and executing the operation instruction when the approval result indicates that the operation instruction passes;
wherein the registering of the registered model comprises:
obtaining a model registration instruction, and generating a corresponding standard modeling flow;
acquiring a parameter adjusting instruction, and modifying the parameter corresponding to the current model according to the parameter adjusting instruction;
acquiring an operation instruction, executing an operation corresponding to the operation instruction, and displaying an operation result on a canvas;
and obtaining a model storage instruction, and storing the current model and the corresponding parameters to the registered model.
In a third aspect, the present invention also provides an electronic device, including:
at least one memory and at least one processor;
the memory to store one or more programs;
when executed by the at least one processor, cause the at least one processor to carry out the steps of a data mining model management method according to any one of the first aspect of the invention.
In a fourth aspect, the present invention also provides a computer-readable storage medium,
the computer readable storage medium stores a computer program which, when executed by a processor, implements the steps of a data mining model management method according to any one of the first aspect of the invention.
According to the data mining model management method and device, the electronic equipment and the storage medium, when a user executes model creation, the system provides an industry-matched workflow template and a standardized modeling flow component, and the threshold of the user for model development is greatly reduced. When a user needs to check the registered or deployed model information, the user is strictly controlled by role authority, so that the condition that the user authority is too large or too small is avoided. When a user needs to execute relevant operations of model deployment, the user also needs to ask a decision maker for approval, so that misoperation or unauthorized operation of the user is prevented. The real-time monitoring of the model effect can help a user to know the change of the model effect more timely and accurately, and the model is optimized or stopped timely.
For a better understanding and practice, the invention is described in detail below with reference to the accompanying drawings.
Drawings
FIG. 1 is a functional framework diagram of a model full lifecycle management system used in one embodiment of the present invention;
FIG. 2 is a diagram of the steps of operation of the management system shown in FIG. 1;
FIG. 3 is a schematic diagram illustrating steps of a data mining model management method according to the present invention;
FIG. 4 is a schematic diagram of a model deployment interface of the management system shown in FIG. 1;
FIG. 5 is a flow chart of an approval process of the management system shown in FIG. 1;
FIG. 6 is a schematic diagram of a model task interface of the management system shown in FIG. 1;
FIG. 7 is a schematic diagram of a transport task interface of the management system shown in FIG. 1;
FIG. 8 is a diagrammatic illustration of a predictive tasks interface of the management system shown in FIG. 1;
fig. 9 is a schematic structural diagram of a data mining model management apparatus provided in the present invention.
Detailed Description
To make the objects, technical solutions and advantages of the present application more clear, embodiments of the present application will be described in further detail below with reference to the accompanying drawings.
It should be understood that the embodiments described are only some embodiments of the present application, and not all embodiments. All other embodiments obtained by a person of ordinary skill in the art based on the embodiments in the present application without any creative effort belong to the protection scope of the embodiments in the present application.
The terminology used in the embodiments of the present application is for the purpose of describing particular embodiments only and is not intended to be limiting of the embodiments of the present application. As used in the examples of this application and the appended claims, the singular forms "a", "an", and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It should also be understood that the term "and/or" as used herein refers to and encompasses any and all possible combinations of one or more of the associated listed items.
The following description refers to the accompanying drawings in which the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present application. Rather, they are merely examples of apparatus and methods consistent with certain aspects of the application, as detailed in the appended claims. In the description of the present application, it is to be understood that the terms "first," "second," "third," and the like are used solely to distinguish one from another and are not necessarily used to describe a particular order or sequence, nor are they to be construed as indicating or implying relative importance. The specific meaning of the above terms in the present application can be understood by those of ordinary skill in the art as appropriate.
In addition, in the description of the present application, "a plurality" means two or more unless otherwise specified. "and/or" describes the association relationship of the associated objects, meaning that there may be three relationships, e.g., a and/or B, which may mean: a exists alone, A and B exist simultaneously, and B exists alone. The character "/" generally indicates that the former and latter associated objects are in an "or" relationship.
To solve the problems in the background art, in a specific application scenario, an embodiment of the present application provides a model full lifecycle management system, as shown in fig. 1, the system includes the following functional modules: the system comprises a workflow management module, a data source management module, a model building module and a model deployment module.
In a specific application scenario, as shown in fig. 2, the steps of using the system by the user include:
step 1, creating a workflow and filling related information of the workflow; and selecting related cases from the workflow template library as the workflow template.
And 2, configuring a data source and supporting a common relational database and a common cloud database.
And 3, dragging any model component in the standard process to a modeling canvas, and automatically generating the corresponding standard modeling process.
And 4, opening corresponding components in the canvas, and adjusting parameters such as data source file selection, data partitioning, data quality inspection, data cleaning, data binning, characteristic importance evaluation, variable collinearity inspection, algorithm realization and the like.
And 5, clicking a 'run' button to perform operation on one or more components and also to perform the whole workflow.
And 6, opening a result page of each component, and seeing the operation result of each component.
And 7, repeating the operations of the steps 2 to 6, constructing a plurality of different models in the same canvas, and comparing the advantages and disadvantages of the models by using a model comparison component, so that the optimal model can be selected conveniently.
And 8, importing the table generated in the process of training the model into a certain data source according to a certain form by using a 'writing out data table' component, and well saving and backing up files.
And 9, clicking a 'save' button to save the canvas operation.
Step 10, clicking a 'model registration' button, jumping to a 'model management' module, and filling relevant information of the model, wherein the method comprises the following steps: model name, model version, etc., to facilitate management of the model. At the same time, the user can view the existing model according to the authority.
And step 11, entering a model deployment interface, wherein a user can monitor the state of the on-line model according to the authority and can also perform operations such as on-line, updating and stopping on the registered model.
And step 12, when the user performs operations such as online, updating and stopping the registered model, the system automatically sends an application to an approval system, and the operation is executed after approval by a decision maker.
Based on the above model full lifecycle management system, the embodiment of the application provides a data mining model management method, as shown in fig. 3, the method comprises the steps of:
s01: and responding to the model deployment instruction, and generating and displaying a model deployment interface, wherein the model deployment interface comprises the name of the on-line model, the state corresponding to the on-line model, and an operation instruction button for the registered model.
Preferably, the model deployment interface displays the name, version, model state and update time of the currently online deployment model in a tabular form.
In a specific application, as shown in fig. 4, a user can view model information established by the user and shared public model information established by others according to account permissions. The specific operations/functions are as follows:
S02: responding to an operation instruction of the registered model, generating an operation approval message, and sending the operation approval message to a corresponding decision maker; the operating instructions for the registered model include at least one of: online, updating and deactivating.
S03: and obtaining an approval result of the decision maker, and executing the operation instruction when the approval result indicates that the operation instruction passes through the operation instruction.
In a preferred embodiment, the process of online deployment and approval of the model is shown in fig. 5, and the approval process is performed by a decision maker, so as to prevent misoperation or unauthorized operation of a user.
Wherein the registering of the registered model comprises:
s11: and obtaining a model registration instruction, and generating a corresponding standard modeling flow.
S12: and acquiring a parameter adjusting instruction, and modifying the parameters corresponding to the current model according to the parameter adjusting instruction.
S13: and acquiring an operation instruction, executing the operation corresponding to the operation instruction, and displaying the operation result on a canvas.
Selecting a data source file, partitioning data, checking data quality, cleaning data, binning data, evaluating feature importance, checking variable collinearity and realizing an algorithm.
S14: and acquiring a model storage instruction, and storing the current model and the corresponding parameters to the registered model.
In a more preferred embodiment, the data mining model management method provided by the present invention further includes the steps of:
s21: acquiring the system authority of a current user according to the login information of the user;
s22: and generating a system interface of the user according to the system authority of the user.
Through the permission setting, when a user needs to check the registered or deployed model information, the user is strictly controlled by the role permission, so that the condition that the user permission is too large or too small is avoided. And effective authority control is performed, the registered model viewing authority is controlled through the user role authority, and the model deployment related operation is controlled.
In another preferred embodiment, the data mining model management method provided by the present application further includes the following steps:
s31: for the on-line model, acquiring a service monitoring instruction, and generating and displaying operation data corresponding to the current model; the operational data includes at least one of: the degree of improvement, AUC value, rate of accuracy, recall.
S32: and generating an operation performance change report of the current model according to the historical change condition of the operation data.
In specific application, a user clicks an icon in service monitoring to control a system to display the operation effect (such as the promotion degree, the AUC value, the accuracy, the precision, the recall rate and the like) of a model, and the performance change of the model is informed to the user in time, so that the user can evaluate whether the model needs to be retrained, put on shelf and the like.
In another preferred embodiment, the data mining model management method provided by the present application further includes the following steps:
s41: and acquiring a timing task instruction for the on-line model.
S42: and executing the timing task according to the timing scheduling information corresponding to the timing task instruction, and generating corresponding reminding information after execution.
Preferably, the timing task includes at least one of:
the model task is used for periodically updating the registered model;
a transmission task for providing a data migration service;
and the prediction task is used for providing automatic batch reasoning.
In a specific application, as shown in fig. 6, the specific operations/functions of the model task are as follows:
As shown in fig. 7, the specific operation/function of the transmission task is as follows:
As shown in fig. 8, the specific operation/function of the prediction task is as follows:
An embodiment of the present application further provides a data mining model management apparatus, as shown in fig. 9, the data mining model management apparatus 400 includes:
the model deployment interface display module 401 is configured to generate and display a model deployment interface in response to a model deployment instruction, where the model deployment interface includes an online model name, an online state corresponding to the model, and an operation instruction button for the registered model;
an operation approval message generation module 402, configured to generate an operation approval message in response to an operation instruction for the registered model, and send the operation approval message to a corresponding decision maker; the operating instructions on the registered model comprise at least one of the following: online, updating and stopping;
an operation instruction execution module 403, configured to obtain an approval result of a decision maker, and execute the operation instruction when the approval result indicates that the operation instruction passes through the approval result;
wherein the registering of the registered model comprises:
obtaining a model registration instruction, and generating a corresponding standard modeling flow;
the parameter adjustment instruction is obtained, and the parameter adjustment instruction is obtained, modifying the parameters corresponding to the current model according to the parameter adjusting instruction;
acquiring an operation instruction, executing an operation corresponding to the operation instruction, and displaying an operation result on a canvas;
and obtaining a model storage instruction, and storing the current model and the corresponding parameters to the registered model.
Preferably, the operation instruction comprises at least one of the following items:
selecting a data source file, partitioning data, checking data quality, cleaning data, binning data, evaluating feature importance, checking variable collinearity and realizing an algorithm.
Preferably, the target detection network model is a YOLOv5 model.
Preferably, the method further comprises the following steps:
a system authority obtaining module for obtaining the login information of the current user, acquiring the system authority of the user;
and the system interface generating module is used for generating the system interface of the user according to the system authority of the user.
Preferably, the method further comprises the following steps:
the operation data generation module is used for acquiring a service monitoring instruction for the on-line model, and generating and displaying operation data corresponding to the current model; the operational data includes at least one of: the promotion degree, the AUC value, the accuracy, the precision and the recall rate;
and the running performance change report generation module is used for generating a running performance change report of the current model according to the historical change condition of the running data.
Preferably, the model deployment interface displays the name, version, model state and update time of the currently online deployment model in a tabular form.
Preferably, the method further comprises the following steps:
the timing task instruction acquisition module is used for acquiring a timing task instruction for the on-line model;
and the timing task execution module is used for executing the timing task according to the timing scheduling information corresponding to the timing task instruction and generating corresponding reminding information after the timing task is executed.
Preferably, the timing task comprises at least one of:
the model task is used for periodically updating the registered model;
a transmission task for providing a data migration service;
and the prediction task is used for providing automatic batch reasoning.
For the device embodiments, since they substantially correspond to the method embodiments, reference may be made to the partial description of the method embodiments for relevant points. The above-described embodiments of the apparatus are merely illustrative, and the units described as separate parts may or may not be physically separate, and the parts displayed as units may or may not be physical units. It is clear to those skilled in the art that, for convenience and brevity of description, the specific working processes of the above-described systems, apparatuses and units may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again.
An embodiment of the present application further provides an electronic device, including:
at least one memory and at least one processor;
the memory for storing one or more programs;
when executed by the at least one processor, cause the at least one processor to implement the steps of a data mining model management method as described above.
For the device embodiment, since it substantially corresponds to the method embodiment, so that the relevant points can be found in the description of the method embodiments. The above-described device embodiments are merely illustrative, wherein the components described as separate parts may or may not be physically separate, and the parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the disclosed solution. Without inventive effort by one of ordinary skill in the art, i.e., can be understood and implemented.
Embodiments of the present application also provide a computer-readable storage medium,
the computer readable storage medium stores a computer program which, when executed by a processor, implements the steps of a data mining model management method as previously described.
Computer-usable storage media include permanent and non-permanent, removable and non-removable media, and information storage may be implemented by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of the storage medium of the computer include, but are not limited to: phase change memory (PRAM), static Random Access Memory (SRAM), dynamic Random Access Memory (DRAM), other types of random access memory (rram), read only memory (ro M), electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technologies, compact disc read only memory (CD-ROM), digital Versatile Discs (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices, or any other non-transmission medium, which may be used to store information that may be accessed by a computing device.
According to the data mining model management method, the data mining model management device, the electronic equipment and the storage medium, when a user executes model creation, the system provides an industry-matched workflow template and a standardized modeling flow component, and the threshold of model development of the user is greatly reduced. When a user needs to check the registered or deployed model information, the user is strictly controlled by role authority, so that the condition that the user authority is too large or too small is avoided. When a user needs to execute relevant operations of model deployment, the user also needs to ask a decision maker for approval, so that misoperation or unauthorized operation of the user is prevented. The real-time monitoring of the model effect can help a user to know the change of the model effect more timely and accurately, and the model is optimized or stopped timely.
The above-mentioned embodiments only express several embodiments of the present invention, and the description thereof is more specific and detailed, but not construed as limiting the scope of the invention. It should be noted that, for those skilled in the art, without departing from the inventive concept, several variations and modifications can be made, which are within the scope of the invention.
Claims (10)
1. A data mining model management method is characterized by comprising the following steps:
responding to a model deployment instruction, generating and displaying a model deployment interface, wherein the model deployment interface comprises an online model name, an online state corresponding to the model, and an operation instruction button for the registered model;
responding to an operation instruction of the registered model, generating an operation approval message, and sending the operation approval message to a corresponding decision maker; the operating instructions on the registered model comprise at least one of the following: online, updating and stopping;
obtaining an approval result of a decision maker, and executing the operation instruction when the approval result indicates that the operation instruction passes through the operation instruction;
wherein the registering of the registered model comprises:
obtaining a model registration instruction, and generating a corresponding standard modeling flow;
acquiring a parameter adjusting instruction, and modifying the parameter corresponding to the current model according to the parameter adjusting instruction;
acquiring an operation instruction, executing the operation corresponding to the operation instruction, displaying the operation result on the canvas;
and obtaining a model storage instruction, and storing the current model and the corresponding parameters to the registered model.
2. The data mining model management method of claim 1, wherein the operational instructions comprise at least one of:
selecting a data source file, partitioning data, checking data quality, cleaning data, binning data, evaluating feature importance, checking variable collinearity and realizing an algorithm.
3. The data mining model management method of claim 1, further comprising the steps of:
acquiring the system authority of the user according to the login information of the current user;
and generating a system interface of the user according to the system authority of the user.
4. The method of managing a data mining model of claim 1, further comprising the steps of:
for the on-line model, acquiring a service monitoring instruction, and generating and displaying operation data corresponding to the current model; the operational data includes at least one of: the promotion degree, the AUC value, the accuracy, the precision and the recall rate;
and generating an operation performance change report of the current model according to the historical change condition of the operation data.
5. The data mining model management method of claim 4, wherein:
the model deployment interface displays the name of the current online deployment model in a tabular form version, model state, update time.
6. The data mining model management method of claim 1, further comprising the steps of:
acquiring a timing task instruction for the online model;
and executing the timing task according to the timing scheduling information corresponding to the timing task instruction, and generating corresponding reminding information after execution.
7. The data mining model management method of claim 6, wherein the timing task comprises at least one of:
the model task is used for periodically updating the registered model;
the task of the transmission is carried out, for providing data migration services;
and the prediction task is used for providing automatic batch reasoning.
8. A data mining model management apparatus, comprising:
the model deployment interface display module is used for responding to a model deployment instruction and generating and displaying a model deployment interface, and the model deployment interface comprises an online model name, an online state corresponding to the model and an operation instruction button for the registered model;
the operation approval message generation module is used for responding to an operation instruction of the registered model, generating an operation approval message and sending the operation approval message to a corresponding decision maker; the operating instructions on the registered model comprise at least one of the following: online, updating and stopping;
the operation instruction execution module is used for acquiring the approval result of the decision maker and executing the operation instruction when the approval result indicates that the operation instruction passes;
wherein the registering of the registered model comprises:
obtaining a model registration instruction, and generating a corresponding standard modeling flow;
acquiring a parameter adjusting instruction, and modifying the parameter corresponding to the current model according to the parameter adjusting instruction;
acquiring an operation instruction, executing an operation corresponding to the operation instruction, and displaying an operation result on a canvas;
and acquiring a model storage instruction, and storing the current model and the corresponding parameters to the registered model.
9. An electronic device, comprising:
at least one memory and at least one processor;
the memory for storing one or more programs;
when executed by the at least one processor, cause the at least one processor to perform the steps of a data mining model management method according to any one of claims 1 to 7.
10. A computer-readable storage medium characterized by:
the computer readable storage medium stores a computer program which when executed by a processor implements the steps of a data mining model management method as claimed in any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211547081.8A CN115964035A (en) | 2022-12-05 | 2022-12-05 | Data mining model management method and device, electronic equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211547081.8A CN115964035A (en) | 2022-12-05 | 2022-12-05 | Data mining model management method and device, electronic equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115964035A true CN115964035A (en) | 2023-04-14 |
Family
ID=87351890
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211547081.8A Pending CN115964035A (en) | 2022-12-05 | 2022-12-05 | Data mining model management method and device, electronic equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115964035A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116627434A (en) * | 2023-07-24 | 2023-08-22 | 北京寄云鼎城科技有限公司 | Model deployment service method, electronic equipment and medium |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109583796A (en) * | 2019-01-08 | 2019-04-05 | 河南省灵山信息科技有限公司 | A kind of data digging system and method for Logistics Park OA operation analysis |
| CN110427398A (en) * | 2018-04-28 | 2019-11-08 | 北京资采信息技术有限公司 | A kind of model management tool based on data mining and analysis |
| CN111523812A (en) * | 2020-04-24 | 2020-08-11 | 同盾控股有限公司 | Model life cycle management method and system, equipment and storage medium |
-
2022
- 2022-12-05 CN CN202211547081.8A patent/CN115964035A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110427398A (en) * | 2018-04-28 | 2019-11-08 | 北京资采信息技术有限公司 | A kind of model management tool based on data mining and analysis |
| CN109583796A (en) * | 2019-01-08 | 2019-04-05 | 河南省灵山信息科技有限公司 | A kind of data digging system and method for Logistics Park OA operation analysis |
| CN111523812A (en) * | 2020-04-24 | 2020-08-11 | 同盾控股有限公司 | Model life cycle management method and system, equipment and storage medium |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116627434A (en) * | 2023-07-24 | 2023-08-22 | 北京寄云鼎城科技有限公司 | Model deployment service method, electronic equipment and medium |
| CN116627434B (en) * | 2023-07-24 | 2023-11-28 | 北京寄云鼎城科技有限公司 | Model deployment service method, electronic equipment and medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US7974896B2 (en) | Methods, systems, and computer program products for financial analysis and data gathering | |
| US8126768B2 (en) | Application change request to deployment maturity model | |
| CN111159157B (en) | Index processing method and device for enterprise report data | |
| US20180129483A1 (en) | Developing software project plans based on developer sensitivity ratings detected from monitoring developer error patterns | |
| US20160086121A1 (en) | Providing Gamification Analytics in an Enterprise Environment | |
| Krauß et al. | Selection and application of machine learning-algorithms in production quality | |
| US11017339B2 (en) | Cognitive labor forecasting | |
| EP3961523A1 (en) | Managing production pipelines | |
| US10135680B2 (en) | Assessing computing environment changes using specification documents | |
| US20200097867A1 (en) | Visualization of cross-project dependency risk | |
| US20210042214A1 (en) | Generating test plans for testing computer products based on product usage data | |
| CN115964035A (en) | Data mining model management method and device, electronic equipment and storage medium | |
| Weber et al. | A New Process Model for the Comprehensive Management of Machine Learning Models. | |
| CN115587720A (en) | Configuration management system | |
| US10324821B2 (en) | Oracle cemli analysis tool | |
| CN116149824A (en) | Task re-running processing method, device, equipment and storage medium | |
| CN115065597A (en) | Container resource allocation method, device, medium and equipment | |
| CN107766306A (en) | A kind of method and apparatus of data display | |
| CN115699042A (en) | Collaborative system and method for validating analysis of device failure models in crowd-sourced environments | |
| US20240212059A1 (en) | Enhanced automated entity close process management | |
| Santos et al. | Diagnostic assessment of product lifecycle management based on Industry 4.0 requirements | |
| US11977997B1 (en) | Tagging systems and methods for efficient cloud service provisioning | |
| US20240029031A1 (en) | Machine learning recommendation for maintenance targets in preventive maintenance plans | |
| US20240220899A1 (en) | Method and system to use real-time activity to provide workforce monitoring | |
| Hamlet et al. | IDC Re-Engineering Phase 2 Iteration E1 Use Case Realizations version 1.2. |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20230414 |
|
| RJ01 | Rejection of invention patent application after publication |