CN111625567A - Data model matching method, device, computer system and readable storage medium - Google Patents
Data model matching method, device, computer system and readable storage medium Download PDFInfo
- Publication number
- CN111625567A CN111625567A CN202010341670.5A CN202010341670A CN111625567A CN 111625567 A CN111625567 A CN 111625567A CN 202010341670 A CN202010341670 A CN 202010341670A CN 111625567 A CN111625567 A CN 111625567A
- Authority
- CN
- China
- Prior art keywords
- data
- model
- user side
- machine learning
- information
- 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.)
- Withdrawn
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2457—Query processing with adaptation to user needs
- G06F16/24573—Query processing with adaptation to user needs using data annotations, e.g. user-defined metadata
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
- G16H50/70—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for mining of medical data, e.g. analysing previous cases of other patients
Abstract
The invention discloses a data model matching method, a device, a computer system and a readable storage medium, comprising the following steps: extracting structural metadata in the acquired structured data, and generating an object bullet frame with a grouping input field and the structural metadata; receiving object information sent by a user side according to an object bullet frame, acquiring unit data from the structured data according to the object information, and extracting grouping information and the unit data in the object information to form data to be operated; acquiring a machine learning model matched with data to be calculated from a preset model library according to model rules, sending a model bullet frame with an option label of the machine learning model to a user side, and receiving model information sent by the user side according to the model bullet frame; and extracting a target model from the matched machine learning model according to the model information, and calculating the data to be calculated through the target model to obtain a calculation result. The invention improves the use efficiency of the machine learning model and reduces the complexity of the machine learning model.
Description
Technical Field
The invention relates to the technical field of artificial intelligence, in particular to a data model matching method, a data model matching device, a computer system and a readable storage medium.
Background
The current data analysis system is usually that a data analyst inputs original bottom layer data into the system, so that the system generates a visual image, and analyzes the image to realize data analysis; however, the method for analyzing the data by analyzing the visual image often has inaccurate analysis result due to the doping of human factors; therefore, to address these issues, some businesses or institutions may employ machine learning models to analyze the underlying data of the visualized images.
However, the method for analyzing the underlying data through the machine learning model requires a large amount of data combing and environment building operations by professionals, which is not only inefficient, but also causes that people without professional knowledge cannot correctly and effectively use the machine learning model due to complicated data combing and environment building.
Disclosure of Invention
The invention aims to provide a data model matching method, a data model matching device, a computer system and a readable storage medium, which are used for solving the problems that the efficiency is low in the prior art, and a person without professional knowledge cannot correctly and effectively use a machine learning model due to complex data carding and environment building.
In order to achieve the above object, the present invention provides a data model matching method, including:
extracting structural metadata in the acquired structured data, generating an object bullet frame with a grouping input field and the structural metadata, and sending the object bullet frame to a user side;
receiving object information sent by a user side according to the object bullet frame, acquiring unit data from the structured data according to the object information, and extracting grouping information in the object information and the unit data to form data to be operated;
acquiring a machine learning model matched with the data to be calculated from a preset model library according to model rules, sending a model bullet frame with an option label of the machine learning model to the user side, and receiving model information sent by the user side according to the model bullet frame;
and extracting a target model from the matched machine learning model according to the model information, calculating the data to be calculated through the target model to obtain a calculation result, and returning the calculation result to the user side.
In the above scheme, before extracting the structural metadata in the acquired structured data, the method includes:
acquiring bottom layer data, and carrying out structuring processing on the bottom layer data to obtain structured data.
In the foregoing solution, the step of performing a structuring process on the bottom layer data to obtain structured data includes:
extracting metadata of each subdata in the bottom data, summarizing the metadata of all the bottom data and removing duplication to form structural metadata;
and drawing a two-dimensional table according to the structural metadata, sequentially recording the subdata in all the bottom data into the two-dimensional table according to the structural metadata to form structured data.
In the foregoing solution, after the packet information in the object information and the unit data are extracted to form data to be operated, the method may further include:
the method comprises the steps of obtaining a visual assembly matched with the data to be operated from a preset assembly library according to visual rules, sending a visual bullet frame with an option label of the visual assembly to a user side, and receiving assembly information sent by the user side according to the visual bullet frame.
In the above scheme, after receiving the component information sent by the user side according to the visual bullet frame, the method may further include:
extracting a target component from the matched visual components according to the component information, operating the unit data through the target component to generate a visual image, and returning the visual image to the user side.
In the above solution, the structural metadata may include: order date, shipping date, country, region, category, subcategory, product name, manufacturer.
In the above solution, the option label at least includes a name, a purpose, and a use of the machine learning model.
In order to achieve the above object, the present invention further provides a data model matching apparatus, including:
the object management module is used for extracting structural metadata in the acquired structural data, generating an object bullet frame with a grouping input field and the structural metadata, and sending the object bullet frame to the user side;
the operation data generation module is used for receiving object information sent by a user side according to the object bullet frame, acquiring unit data from the structured data according to the object information, and extracting grouping information in the object information and the unit data to form data to be operated;
the model matching module is used for acquiring a machine learning model matched with the data to be calculated from a preset model library according to model rules, sending a model pop frame with an option tag of the machine learning model to the user side, and receiving model information sent by the user side according to the model pop frame;
and the model selection operation module is used for extracting a target model from the matched machine learning model according to the model information, operating the data to be operated through the target model to obtain a calculation result, and returning the calculation result to the user side.
In order to achieve the above object, the present invention further provides a computer system, which includes a plurality of computer devices, each computer device including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processors of the plurality of computer devices jointly implement the steps of the data model matching method when executing the computer program.
In order to achieve the above object, the present invention further provides a computer-readable storage medium, which includes a plurality of storage media, each storage medium having a computer program stored thereon, wherein the computer programs stored in the storage media, when executed by a processor, collectively implement the steps of the data model matching method.
According to the data model matching method, the data model matching device, the computer system and the readable storage medium, the structural metadata in the structured data are extracted and the object bullet frame is generated, so that a user can conveniently select the required structural metadata on the object bullet frame; the method comprises the steps that sub-data corresponding to object data are obtained from structured data and set as unit data through structural metadata selected by a user on an object bullet frame, a machine learning model matched with data to be calculated is obtained from a preset model base, and the model bullet frame describing the purpose of the machine learning model is sent to a user side, so that the user can select a required machine learning model on the model bullet frame according to an option label and generate model information, the user can analyze the unit data quickly and conveniently, and meanwhile, the machine learning model is prestored in the model base, so that complex data carding and environment building are avoided, and the use efficiency of the machine learning model is improved; extracting a model corresponding to the model information from a machine learning model matched with the data to be operated through the model information generated by the user side, setting the model as a target model, and operating the unit data by using the target model to obtain a calculation result; because complex data combing and environment building are not needed, a user without professional knowledge can call a correct machine learning model and effectively use the machine learning model only through the elastic frame, the use complexity of the machine learning model is greatly reduced, and the convenience of the user in using the machine learning model is improved.
Drawings
FIG. 1 is a flowchart of a first embodiment of a data model matching method according to the present invention;
FIG. 2 is a schematic diagram of an environment application of a data model matching method according to a second embodiment of the data model matching method of the present invention;
FIG. 3 is a flowchart of a data model matching method according to a second embodiment of the present invention;
FIG. 4 is an information interaction diagram of a data model matching method according to a second embodiment of the data model matching method of the present invention;
FIG. 5 is a flowchart of obtaining structured data according to a second embodiment of the data model matching method of the present invention:
FIG. 6 is a schematic diagram of program modules of a third embodiment of a data model matching apparatus according to the present invention;
fig. 7 is a schematic diagram of a hardware structure of a computer device according to a fourth embodiment of the computer system of the present invention.
Reference numerals:
1. data model matching device 2, server 3, network 4, user side
5. Computer device 11, structured processing module 12, object management module
13. Operation data generation module 14, model matching module 15 and component matching module
16. Model selection operation module 17, component selection operation module 51, memory 52 and processor
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The invention provides a data model matching method, a data model matching device, a computer system and a readable storage medium, which are suitable for the technical field of artificial intelligence and are used for providing a data model matching method based on an object management module, an operation data generation module, a model matching module and a model selection operation module. According to the method, the structural metadata in the acquired structural data is extracted, the object bullet frame with the grouping input field and the structural metadata is generated, and the object bullet frame is sent to a user side; receiving object information sent by a user side according to an object bullet frame, acquiring unit data from the structured data according to the object information, and extracting grouping information and the unit data in the object information to form data to be operated; acquiring a machine learning model matched with data to be calculated from a preset model library according to model rules, sending a model bullet frame with an option label of the machine learning model to a user side, and receiving model information sent by the user side according to the model bullet frame; and extracting a target model from the matched machine learning model according to the model information, calculating the data to be calculated through the target model to obtain a calculation result, and returning the calculation result to the user side.
The first embodiment is as follows:
referring to fig. 1, a data model matching method of the present embodiment includes:
s102: extracting structural metadata in the acquired structured data, generating an object bullet frame with a grouping input field and the structural metadata, and sending the object bullet frame to a user side;
s103: receiving object information sent by a user side according to the object bullet frame, acquiring unit data from the structured data according to the object information, and extracting grouping information in the object information and the unit data to form data to be operated;
s105: acquiring a machine learning model matched with the data to be calculated from a preset model library according to model rules, sending a model bullet frame with an option label of the machine learning model to the user side, and receiving model information sent by the user side according to the model bullet frame;
s107: and extracting a target model from the matched machine learning model according to the model information, calculating the data to be calculated through the target model to obtain a calculation result, and returning the calculation result to the user side.
In an exemplary embodiment, the problem of how to facilitate the user to select the data to be analyzed on the user side is solved by extracting the structural metadata in the structured data and generating the object bounding box, so that the user can select the required structural metadata on the object bounding box, and the grouping input field of the object bounding box is an information input field into which numbers can be input, in this embodiment, the structural metadata may include: order date, shipping date, country, region, category, subcategory, product name, manufacturer.
Through the structural metadata selected by the user on the object bullet frame, the subdata corresponding to the object data is obtained from the structured data and is set as unit data, and the problem of how to obtain a data set according to the requirements of the user is solved.
The machine learning model matched with the unit data is obtained from the preset model library, and the model bullet frame with the function of describing the machine learning model is sent to the user side, so that a user can select the needed machine learning model on the model bullet frame according to the option labels and generate model information, and the problem of how to recommend the machine learning model suitable for analyzing the unit data to the user so that the user can quickly and conveniently analyze the unit data is solved; meanwhile, the machine learning model is prestored in the model base, so that complex data combing and environment building are avoided, and the use efficiency of the machine learning model is improved.
Extracting a model corresponding to the model information from a machine learning model matched with unit data of the data to be operated through the model information generated by the user side, setting the model as a target model, and operating the unit data by using the target model to obtain a calculation result; the user does not need to carry out professional training, and only needs to select the required unit data and the target model in the object bullet frame and the model bullet frame, so that the analysis of the underlying data can be completed and the calculation result can be obtained, the use complexity of the machine learning model is greatly reduced, the convenience of using the machine learning model by the user is improved, and the problem of directly obtaining the required calculation result through the selected unit data and the selected target model by the user is solved.
Because complex data combing and environment building are not needed, a user without professional knowledge can call a correct machine learning model and effectively use the machine learning model only through the elastic frame, the use complexity of the machine learning model is greatly reduced, and the convenience of the user in using the machine learning model is improved.
Example two:
the embodiment is a specific application scenario of the first embodiment, and the method provided by the present invention can be more clearly and specifically explained through the embodiment.
The method provided in this embodiment is specifically described below by taking an example of performing transaction management on at least one user side at a server running a data model matching method. It should be noted that the present embodiment is only exemplary, and does not limit the protection scope of the embodiments of the present invention.
Fig. 2 schematically shows an environment application diagram of the data model matching method according to the second embodiment of the present application.
In an exemplary embodiment, the server 2 where the data model matching method is located is connected to the user terminal 4 through the network 3; the server 2 may provide services through one or more networks 3, which networks 3 may include various network devices, such as routers, switches, multiplexers, hubs, modems, bridges, repeaters, firewalls, proxy devices, and/or the like. The network 3 may include physical links, such as coaxial cable links, twisted pair cable links, fiber optic links, combinations thereof, and/or the like. The network 3 may include wireless links, such as cellular links, satellite links, Wi-Fi links, and/or the like; the user terminal 4 may be a server installed with a face recognition algorithm program.
Fig. 3 is a flowchart of a specific method of a data model matching method according to an embodiment of the present invention, where the method specifically includes steps S201 to S208.
Fig. 4 is an information interaction diagram of a data model matching method according to an embodiment of the present invention.
S201: acquiring bottom layer data, and performing structuring processing on the bottom layer data to obtain structured data;
in order to facilitate regression analysis, cluster analysis, numerical prediction, text analysis and relationship analysis of the obtained bottom layer data, the bottom layer data is subjected to structured processing to obtain structured data, so that character type data, numerical type data and date type data in the bottom layer data can be arranged and classified according to metadata of the character type data, the numerical type data and the date type data, data analysis of the bottom layer data is facilitated by a machine learning model, a visualization component performs visualization processing on the bottom layer data to generate a visualization image, and the problem that a data analyst has entry errors is avoided.
It should be noted that the character type field is defined as a character data type without calculation capability, such as "product name", "manufacturer", etc., and includes chinese characters, english characters, numeric characters, and other ascii characters. The definition of the numerical field is to describe the numerical characteristics of things, for example, "unit price", "quantity", "sales", etc. are numerical types, and these fields may take different numerical values. The defined data form can be directly loaded into a memory or a register to carry out mathematical operations such as addition, subtraction, multiplication, division and the like; the definition of the date type field is a value containing information of date including year and time including hour, minute and second.
In a preferred embodiment, referring to fig. 5, the step of performing a structuring process on the underlying data to obtain structured data includes:
s201-1: extracting metadata of each subdata in the bottom data, summarizing the metadata of all the bottom data and removing duplication to form structural metadata;
s201-2: and drawing a two-dimensional table according to the structural metadata, sequentially recording the subdata in all the bottom data into the two-dimensional table according to the structural metadata to form structured data.
Illustratively, aggregating the metadata of all the underlying data and de-duplicating to form structural metadata includes: the order date, delivery date, country, region, category, subcategory, product name and manufacturer, and the structured data formed by inputting the bottom layer data into the drawn two-dimensional table according to the structural metadata are as follows:
s202: and extracting the structural metadata in the acquired structured data, generating an object bullet frame with a grouping input field and the structural metadata, and sending the object bullet frame to a user side.
In order to facilitate the user to select the data to be analyzed on the user side, the step extracts the structural metadata in the structured data and generates the object bounding box, so that the user can select the required structural metadata on the object bounding box conveniently. Illustratively, the grouping input field of the object bounding box is an information input field into which numbers can be input, and the structural metadata includes: order date, shipping date, country, region, category, subcategory, product name, manufacturer.
S203: and receiving object information sent by the user side according to the object bullet frame, acquiring unit data from the structured data according to the object information, and extracting grouping information in the object information and the unit data to form data to be operated.
In order to obtain a data set according to the requirements of a user, the step obtains sub data corresponding to the object data from the structured data through the structural metadata selected by the user on the object bullet frame and sets the sub data as unit data.
Illustratively, the user forms grouping information by inputting numbers in a grouping input field; the structural metadata of the object bullet frame is provided with an object option bar, and a user can select the structural metadata on the object bullet frame by checking at least one object option bar, so that the user can acquire required data from the structured data; and the server operating the data model matching method obtains subdata corresponding to the object data from the structured data through an object option bar selected by a user, sets the subdata as unit data, and summarizes the grouping information and the unit data to form data to be operated.
Furthermore, the user can also input demand information in the object bullet frame, and the type of the demand information can represent the model operation demand of performing model operation on the data to be processed through the machine learning model, and can also represent the visualization demand of generating a visualization image on the data to be processed through the visualization component.
For example: the user inputs a number in the grouping input field to form grouping information "1"; if the user has selected two structural metadata, namely "region" and "category", on the object tab, the following unit data will be obtained:
| region of land | Categories |
| East China | Over the counter medicine |
| Southwest | Over the counter medicine |
| Southwest | Prescription medicine |
| East China | Over the counter medicine |
| Zhongnan province | Over the counter medicine |
| East China | Prescription medicine |
| East China | Prescription medicine |
| East China | Over the counter medicine |
| East China | Over the counter medicine |
| East China | Over the counter medicine |
S204: extracting demand information in object information, and judging the type of the demand information; if the type of the user requirement is a model operation requirement, executing S205 to obtain a machine learning model; if the type of the user requirement is a visualization requirement, S206 is executed to obtain a visualization component.
S205: and acquiring a machine learning model matched with the data to be calculated from a preset model library according to model rules, sending a model bullet frame with an option label of the machine learning model to the user side, and receiving model information sent by the user side according to the model bullet frame.
In order to recommend a machine learning model suitable for analyzing the unit data to a user so that the user can quickly and conveniently analyze the unit data, the step obtains the machine learning model matched with the unit data of the data to be calculated from a preset model library, and sends a model pop-up frame with the purpose of describing the machine learning model to a user side, so that the user can select the required machine learning model on the model pop-up frame according to an option label and generate model information; wherein the option label includes at least a name, a purpose, and a use of the machine learning model. In this embodiment, the user generates the model information by clicking the option tab on the model pop-up box of the user side.
Illustratively, the model rules may be as shown in the following table:
the data to be calculated includes grouping information "1" expressing one grouping, and structural metadata expressing two indexes, that is: "region" and "category"; then, according to the model rule, a machine learning model corresponding to "one group, 2 indexes" is obtained from a preset model library as a machine learning model matched with the data to be calculated, that is: a linear regression model, a Logistic model, a density clustering model and a K mean value model;
loading the names, purposes and purposes of the linear regression model, the Logistic model, the density clustering model and the K-means model into a model bullet frame as option labels to generate the model bullet frame with the machine learning model option labels, wherein the option labels are shown in the following table:
it should be noted that the model library is a database for storing machine learning models, and the machine learning models are computer algorithms that can recognize specific types after being trained. Value in the structured data can be deeply mined through a machine learning model and a visual component.
S206: the method comprises the steps of obtaining a visual assembly matched with the data to be operated from a preset assembly library according to visual rules, sending a visual bullet frame with an option label of the visual assembly to a user side, and receiving assembly information sent by the user side according to the visual bullet frame.
In order to recommend a visual component suitable for analyzing the unit data to a user, so that the user can quickly and conveniently analyze the unit data, the visual component matched with the unit data of the data to be operated is obtained from a preset component library, and a visual bullet frame describing the use of the visual component is sent to a user side, so that the user can select the required visual component on the visual bullet frame according to an option label and generate component information, wherein the name of the visual component is described in the option label. In this embodiment, the user generates the component information by clicking the option tab on the visual pop-up box of the user side.
Illustratively, the visualization rules include:
| data to be operated on-unit data | Visual analysis component |
| One group, 2 indexes | Scatter diagram |
| One group, more than 2 indexes | Contrast scattergram |
| A plurality of groups of 1 index or more | Parallel coordinate axes |
| One group, more than 2 indexes | Scatter diagram |
| A plurality of groups of 1 index or more | Grouping table |
| One group, 2 indexes | Scatter diagram |
| Two groups, greater than 1 index | Parallel coordinate axes |
| Multiple grouping, greater than 1 index | Parallel coordinate axes |
| One group, a plurality of indexes | Scatter diagram |
| Multiple grouping, greater than 1 index | Grouping table |
| One group, a plurality of indexes | Scatter diagram and map |
| One group, a plurality of indexes | Scatter diagram and map |
| One group, a plurality of indexes | Scatter diagram and map |
| Multiple packets | Grouping table |
| One group, a plurality of indexes | Grouping table |
| A date grouping, an index | Line graph |
| A date grouping, an index | Line graph |
| One date grouping, multiple indexes | Line graph |
| One wordGrouping strings | Grouping table |
| A character string grouping | Grouping table |
| Grouping two strings | Grouping table |
| Grouping two strings | Relational graph and sang-base graph |
| Grouping two strings | Relational graph and sang-base graph |
The data to be calculated includes grouping information "1" expressing one grouping, and structural metadata expressing two indexes, that is: "region" and "category"; then, according to the model rule, a visualization component corresponding to "one group, 2 indexes" is obtained from a preset component library as a visualization component matched with the data to be calculated, that is: scatter diagrams, maps, grouping tables; loading the names of the scatter diagram, the map and the grouping table into a model bullet frame as option labels to generate the model bullet frame with the option labels of the visual components, wherein the option labels are shown in the following table:
the component library is a database of visual image components including a scatter diagram component, a map component, a grouping table component, a parallel coordinate axis component, a relation diagram component and a sang-based diagram component. The visual image component is a computer program for making a chart on unit data in data to be operated.
The data visualization realized through the visual image mainly aims at clearly and effectively transmitting and communicating information by means of a graphical means, the thought concept can be effectively transmitted, the aesthetic form and function need to be performed in a parallel manner, and the key aspects and features are intuitively transmitted, so that the deep insight of a quite sparse and complex data set is realized. However, designers often cannot well grasp the balance between design and function, so as to create a Chinese and unrealistic data visualization form, and cannot achieve the main purpose, namely, to communicate and communicate information; value in structured data can be deeply mined through a machine learning model and a visualization component, such as: a data set of a sales pharmacy is provided, a scatter diagram can be made according to the product name, the sales amount and the sales amount, a data analysis component based on the scatter diagram is converted into a cluster analysis model, and automatic classification of products is achieved.
S207: and extracting a target model from the matched machine learning model according to the model information, calculating the data to be calculated through the target model to obtain a calculation result, and returning the calculation result to the user side.
In order to facilitate a user to directly obtain a required calculation result through the selected unit data and the selected target model, in the step, a model corresponding to the model information is extracted from a machine learning model matched with the unit data of the data to be calculated through the model information generated by the user side, the model is set as a target model, and the target model is used for calculating the unit data to obtain a calculation result; the user does not need to carry out professional training, and only needs to select the required unit data and the target model from the object bullet frame and the model bullet frame, so that the analysis of the underlying data can be completed and the calculation result can be obtained, the use complexity of the machine learning model is greatly reduced, and the convenience of the user in using the machine learning model is improved.
Illustratively, a user clicks an option tag of linear regression on a model bullet frame to generate model information, extracts a machine learning model named linear regression from machine learning models matched with the data to be operated as a target model according to the model information, and operates unit data in the data to be operated through the target model to obtain a calculation result and returns the calculation result to a user side.
S208: extracting a target component from the matched visual components according to the component information, operating the unit data through the target component to generate a visual image, and returning the visual image to the user side.
In order to facilitate a user to directly obtain a required calculation result through the selected unit data and the selected target component, in the step, a model corresponding to component information is extracted from a visual component matched with the unit data of the data to be calculated through the component information generated by the user side, the model is set as the target component, and the target component is used for calculating the unit data to obtain the calculation result; the user does not need to carry out professional training, and only needs to select the required unit data and the target assembly from the object bullet frame and the model bullet frame, so that the analysis of the underlying data can be completed, the calculation result can be obtained, the use complexity of the visual assembly is greatly reduced, and the convenience of using the visual assembly by the user is improved.
Illustratively, a user clicks an option label of a scatter diagram on a model pop-up box to generate component information, a visual component named as the scatter diagram is extracted from visual components matched with the data to be operated as a target component according to the model information, unit data in the data to be operated is operated through the target component to obtain a visual image, and the visual image is returned to a user side.
Example three:
referring to fig. 6, a data model matching apparatus 1 of the present embodiment includes:
the object management module 12 is configured to extract structural metadata in the acquired structured data, generate an object bounding box with a grouping input field and the structural metadata, and send the object bounding box to the user side;
the operation data generation module 13 is configured to receive object information sent by a user side according to the object bounding box, obtain unit data from the structured data according to the object information, and extract grouping information in the object information and the unit data to form data to be operated;
the model matching module 15 is configured to acquire a machine learning model matched with the data to be calculated from a preset model library according to a model rule, send a model bullet frame with an option tag of the machine learning model to the user side, and receive model information sent by the user side according to the model bullet frame;
and the model selection operation module 17 is configured to extract a target model from the matched machine learning model according to the model information, operate the data to be operated through the target model to obtain a calculation result, and return the calculation result to the user side.
Optionally, the data model matching apparatus 1 further includes:
the judging module 14 is configured to extract the requirement information in the object information, and judge the type of the requirement information; if the type of the user requirement is a model operation requirement, acquiring a machine learning model; and if the type of the user requirement is a visualization requirement, acquiring a visualization component. Optionally, the data model matching apparatus 1 further includes:
and the structuring processing module 11 is configured to obtain bottom layer data, and perform structuring processing on the bottom layer data to obtain structured data.
Optionally, the data model matching apparatus 1 further includes:
the component matching module 16 is configured to acquire a visual component matched with the to-be-computed data from a preset component library according to a visual rule, send a visual bullet frame with an option tag of the visual component to the user side, and receive component information sent by the user side according to the visual bullet frame.
Optionally, the data model matching apparatus 1 further includes:
and the component selection operation module 18 is used for extracting a target component from the matched visual components according to the component information, operating the unit data through the target component to generate a visual image, and returning the visual image to the user side.
Example four:
in order to achieve the above object, the present invention further provides a computer system, which includes a plurality of computer devices 5, components of the data model matching apparatus 1 according to the second embodiment can be distributed in different computer devices, and the computer devices can be smartphones, tablet computers, notebook computers, desktop computers, rack-mounted servers, blade servers, tower servers, or rack-mounted servers (including independent servers, or a server cluster formed by a plurality of servers) which execute programs, and the like. The computer device of the embodiment at least includes but is not limited to: a memory 51, a processor 52, which may be communicatively coupled to each other via a system bus, as shown in FIG. 7. It should be noted that fig. 7 only shows a computer device with components, but it should be understood that not all of the shown components are required to be implemented, and more or fewer components may be implemented instead.
In this embodiment, the memory 51 (i.e., a readable storage medium) includes a flash memory, a hard disk, a multimedia card, a card-type memory (e.g., SD or DX memory, etc.), a Random Access Memory (RAM), a Static Random Access Memory (SRAM), a read-only memory (ROM), an electrically erasable programmable read-only memory (EEPROM), a programmable read-only memory (PROM), a magnetic memory, a magnetic disk, an optical disk, and the like. In some embodiments, the memory 51 may be an internal storage unit of the computer device, such as a hard disk or a memory of the computer device. In other embodiments, the memory 51 may be an external storage device of a computer device, such as a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash memory Card (Flash Card), or the like, provided on the computer device. Of course, the memory 51 may also include both internal and external storage devices of the computer device. In this embodiment, the memory 51 is generally used for storing an operating system and various application software installed on the computer device, such as the program codes of the data model matching apparatus in the first embodiment. Further, the memory 51 may also be used to temporarily store various types of data that have been output or are to be output.
Processor 52 may be a Central Processing Unit (CPU), controller, microcontroller, microprocessor, or other data Processing chip in some embodiments. The processor 52 is typically used to control the overall operation of the computer device. In this embodiment, the processor 52 is configured to run the program code stored in the memory 51 or process data, for example, run a data model matching apparatus, so as to implement the data model matching method of the first embodiment.
Example five:
to achieve the above objects, the present invention also provides a computer-readable storage system including a plurality of storage media such as a flash memory, a hard disk, a multimedia card, a card type memory (e.g., SD or DX memory, etc.), a Random Access Memory (RAM), a Static Random Access Memory (SRAM), a Read Only Memory (ROM), an Electrically Erasable Programmable Read Only Memory (EEPROM), a Programmable Read Only Memory (PROM), a magnetic memory, a magnetic disk, an optical disk, a server, an App application store, etc., on which a computer program is stored, which when executed by a processor 52, implements corresponding functions. The computer readable storage medium of this embodiment is used for storing a data model matching apparatus, and when being executed by the processor 52, the data model matching method of the first embodiment is implemented.
The above-mentioned serial numbers of the embodiments of the present invention are merely for description and do not represent the merits of the embodiments.
Through the above description of the embodiments, those skilled in the art will clearly understand that the method of the above embodiments can be implemented by software plus a necessary general hardware platform, and certainly can also be implemented by hardware, but in many cases, the former is a better implementation manner.
The above description is only a preferred embodiment of the present invention, and not intended to limit the scope of the present invention, and all modifications of equivalent structures and equivalent processes, which are made by using the contents of the present specification and the accompanying drawings, or directly or indirectly applied to other related technical fields, are included in the scope of the present invention.
Claims (10)
1. A data model matching method, comprising:
extracting structural metadata in the acquired structured data, generating an object bullet frame with a grouping input field and the structural metadata, and sending the object bullet frame to a user side;
receiving object information sent by a user side according to the object bullet frame, acquiring unit data from the structured data according to the object information, and extracting grouping information in the object information and the unit data to form data to be operated;
acquiring a machine learning model matched with the data to be calculated from a preset model library according to model rules, sending a model bullet frame with an option label of the machine learning model to the user side, and receiving model information sent by the user side according to the model bullet frame;
and extracting a target model from the matched machine learning model according to the model information, calculating the data to be calculated through the target model to obtain a calculation result, and returning the calculation result to the user side.
2. The data model matching method according to claim 1, wherein before extracting the structural metadata in the acquired structured data, the method includes:
acquiring bottom layer data, and carrying out structuring processing on the bottom layer data to obtain structured data.
3. The data model matching method of claim 2, wherein the step of structuring the underlying data to obtain structured data comprises:
extracting metadata of each subdata in the bottom data, summarizing the metadata of all the bottom data and removing duplication to form structural metadata;
and drawing a two-dimensional table according to the structural metadata, sequentially recording the subdata in all the bottom data into the two-dimensional table according to the structural metadata to form structured data.
4. The data model matching method according to claim 1, wherein after the group information and the unit data in the object information are extracted to form data to be operated, the method further comprises:
the method comprises the steps of obtaining a visual assembly matched with the data to be operated from a preset assembly library according to visual rules, sending a visual bullet frame with an option label of the visual assembly to a user side, and receiving assembly information sent by the user side according to the visual bullet frame.
5. The data model matching method of claim 4, wherein after receiving the component information sent by the user end according to the visual bullet box, the method further comprises:
extracting a target component from the matched visual components according to the component information, operating the unit data through the target component to generate a visual image, and returning the visual image to the user side.
6. The data model matching method of claim 1, wherein the structural metadata comprises: order date, shipping date, country, region, category, subcategory, product name, manufacturer.
7. The data model matching method of claim 1, wherein the option label includes at least a name, a purpose, and a use of the machine learning model.
8. A data model matching apparatus, comprising:
the object management module is used for extracting structural metadata in the acquired structural data, generating an object bullet frame with a grouping input field and the structural metadata, and sending the object bullet frame to the user side;
the operation data generation module is used for receiving object information sent by a user side according to the object bullet frame, acquiring unit data from the structured data according to the object information, and extracting grouping information in the object information and the unit data to form data to be operated;
the model matching module is used for acquiring a machine learning model matched with the data to be calculated from a preset model library according to model rules, sending a model pop frame with an option tag of the machine learning model to the user side, and receiving model information sent by the user side according to the model pop frame;
and the model selection operation module is used for extracting a target model from the matched machine learning model according to the model information, operating the data to be operated through the target model to obtain a calculation result, and returning the calculation result to the user side.
9. A computer system comprising a plurality of computer devices, each computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processors of the plurality of computer devices when executing the computer program collectively implement the steps of the data model matching method of any of claims 1 to 7.
10. A computer-readable storage medium comprising a plurality of storage media, each storage medium having a computer program stored thereon, wherein the computer programs stored in the storage media, when executed by a processor, collectively implement the steps of the data model matching method of any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010341670.5A CN111625567A (en) | 2020-04-27 | 2020-04-27 | Data model matching method, device, computer system and readable storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010341670.5A CN111625567A (en) | 2020-04-27 | 2020-04-27 | Data model matching method, device, computer system and readable storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111625567A true CN111625567A (en) | 2020-09-04 |
Family
ID=72260563
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010341670.5A Withdrawn CN111625567A (en) | 2020-04-27 | 2020-04-27 | Data model matching method, device, computer system and readable storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111625567A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112434201A (en) * | 2020-12-04 | 2021-03-02 | 高慧军 | Big data based data visualization method and big data cloud server |
| CN112612872A (en) * | 2020-12-17 | 2021-04-06 | 第四范式（北京）技术有限公司 | Method, device, equipment and storage medium for realizing artificial intelligence interpretability |
| CN113792081A (en) * | 2021-08-31 | 2021-12-14 | 吉林银行股份有限公司 | Method and system for automatically checking data assets |
-
2020
- 2020-04-27 CN CN202010341670.5A patent/CN111625567A/en not_active Withdrawn
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112434201A (en) * | 2020-12-04 | 2021-03-02 | 高慧军 | Big data based data visualization method and big data cloud server |
| CN112612872A (en) * | 2020-12-17 | 2021-04-06 | 第四范式（北京）技术有限公司 | Method, device, equipment and storage medium for realizing artificial intelligence interpretability |
| CN113792081A (en) * | 2021-08-31 | 2021-12-14 | 吉林银行股份有限公司 | Method and system for automatically checking data assets |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109189888B (en) | Electronic device, infringement analysis method, and storage medium | |
| CN111625567A (en) | Data model matching method, device, computer system and readable storage medium | |
| US20140229497A1 (en) | Automated data analysis | |
| CN112016273A (en) | Document directory generation method and device, electronic equipment and readable storage medium | |
| WO2022048363A1 (en) | Website classification method and apparatus, computer device, and storage medium | |
| CN112052138A (en) | Service data quality detection method and device, computer equipment and storage medium | |
| CN112528616B (en) | Service form generation method and device, electronic equipment and computer storage medium | |
| CN112541745A (en) | User behavior data analysis method and device, electronic equipment and readable storage medium | |
| CN110765101B (en) | Label generation method and device, computer readable storage medium and server | |
| CN109933502B (en) | Electronic device, user operation record processing method and storage medium | |
| CN112016967B (en) | Transaction data processing method and device | |
| CN111143505A (en) | Document processing method, device, medium and electronic equipment | |
| CN112860905A (en) | Text information extraction method, device and equipment and readable storage medium | |
| CN113886708A (en) | Product recommendation method, device, equipment and storage medium based on user information | |
| CN113688923A (en) | Intelligent order abnormity detection method and device, electronic equipment and storage medium | |
| CN113505273B (en) | Data sorting method, device, equipment and medium based on repeated data screening | |
| CN114491047A (en) | Multi-label text classification method and device, electronic equipment and storage medium | |
| CN111444368B (en) | Method and device for constructing user portrait, computer equipment and storage medium | |
| CN113268665A (en) | Information recommendation method, device and equipment based on random forest and storage medium | |
| CN111752958A (en) | Intelligent associated label method, device, computer equipment and storage medium | |
| CN111723210A (en) | Method and device for storing data table, computer equipment and readable storage medium | |
| US20170017678A1 (en) | System and method for recursive metadata layers on big data sets | |
| CN112507098B (en) | Question processing method, question processing device, electronic equipment, storage medium and program product | |
| CN115018588A (en) | Product recommendation method and device, electronic equipment and readable storage medium | |
| CN114780688A (en) | Text quality inspection method, device and equipment based on rule matching and storage medium |
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 | ||
| WW01 | Invention patent application withdrawn after publication |
Application publication date: 20200904 |
|
| WW01 | Invention patent application withdrawn after publication |