CN109670267A - A kind of data processing method and device - Google Patents
A kind of data processing method and device Download PDFInfo
- Publication number
- CN109670267A CN109670267A CN201811643437.1A CN201811643437A CN109670267A CN 109670267 A CN109670267 A CN 109670267A CN 201811643437 A CN201811643437 A CN 201811643437A CN 109670267 A CN109670267 A CN 109670267A
- Authority
- CN
- China
- Prior art keywords
- data
- model
- user
- algorithm
- industry pattern
- 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.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Hardware Design (AREA)
- Evolutionary Computation (AREA)
- Geometry (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a kind of data processing method, device and equipment, the data processing method includes: to obtain the data identification request of user, carries data to be identified in the data identification request;Based on the attribute information of the data to be identified, determine that data identify industry pattern for the user;The data to be identified are input to the data identification industry pattern and carry out identifying processing, obtain recognition result;The recognition result is sent to the user.The application can find corresponding data by the data identification request of user and identify industry pattern, shorten the duration of user query data identification industry pattern corresponding with data to be identified, improve the working efficiency of user, identify industry pattern for the data to be identified in data identification request by the data that data identification request obtains, more targetedly, when so that data identification industry pattern handling data to be identified, obtained result is more acurrate.
Description
Technical field
This application involves data processing field more particularly to a kind of data processing method and device.
Background technique
With the rapid development of industry, all kinds of manufacturings, process flow, the operation of each company are managed in industrial circle
The various invisible digital assets such as thought, managerial experiences are more and more, and for the ease of the management to these digital assets, these are counted
Word assets are transformed into industrial algorithm model, and therefore, the quantity of industrial algorithm model is also more and more.
In the prior art, in industrial algorithm model, when the data to be tested of user's test in need, user cannot
Industrial algorithm model corresponding with data to be tested are wanted is quickly found, the data to be tested to user cannot be carried out in time
Test, reduces the working efficiency of user.
Summary of the invention
In view of this, the application's has been designed to provide a kind of data processing method and device, solves the prior art
The low problem of middle data-handling efficiency.
In a first aspect, the embodiment of the present application provides a kind of data processing method, comprising:
The data for obtaining user identify request, carry data to be identified in the data identification request;
Based on the attribute information of the data to be identified, determine that data identify industry pattern for the user;
The data to be identified are input to the data identification industry pattern and carry out identifying processing, obtain recognition result;
The recognition result is sent to the user.
Optionally, the attribute information based on the data to be identified determines that data identify industrial mould for the user
Type, comprising:
Obtain the right to use information of preset model corresponding with the attribute information;
According to the right to use information, determine that the user completes that the transaction of servitude of the preset model will be completed
The preset model of the transaction of servitude identifies industry pattern as the data.
Optionally, industry pattern is identified according to the following manner training data:
Training data is obtained from sample database;
Based on the training data and multiple preset models, the initial parameter of each preset model is configured;
The training data is configured in each preset model, each preset model is trained, is obtained each
Complete preset model, the corresponding training result of each preset model and the model accuracy rate of training;
Judge whether the training result meets preset condition, if not satisfied, then adjusting according to the training result each
The parameter of the preset model carries out retraining to the preset model after adjusting parameter, until the training result is full
The preset condition of foot;
The data are determined from each preset model for completing training based on the corresponding model accuracy rate of each preset model
Identify industry pattern.
Optionally, after determining that data identify industry pattern described for the user, further includes:
According to data identification request using the first number of users of data identification industry pattern, work is identified for the data
Industry model configures the first running environment;
According to data identification request using the second user number of data identification industry pattern, the data are identified into work
Industry model virtual turns to the several industry pattern copies of the second user;
It is the user configuration second according to the corresponding user information of the second user number and the first running environment
Running environment;
The data identification industry pattern copy of the user configuration is operated in second running environment.
Optionally, preset model is constructed according to following manner:
Polyalgorithm is edited in algorithm editing machine, the multiple algorithm is stored in algorithm list;
Selection algorithm handles preset training data from the algorithm list, obtains every in the algorithm list
The corresponding algorithm file of a algorithm;
Multiple algorithm files are compared, a kind of algorithm is selected from the algorithm list according to comparing result,
The preset model is generated according to the algorithm of selection.
Optionally, the sample database is constructed according to following manner, comprising:
Obtain the operation data of equipment;
According to preset mapping relations, the operation data is mapped as metadata;
The metadata is formed into the sample database.
Optionally, further includes:
Obtain the initial data of data source;
The source address that the initial data carries is parsed, the detailed data information of the initial data is obtained;
Data source belonging to the initial data and data structure are determined, according to the data source and data determined
The data characteristics of structure mapping extracts the field to match with the data characteristics of the mapping from the detailed data information,
Obtain the data characteristics of the initial data;
According to pre-set target metadata mapping table, the data characteristics of the initial data is extracted, is obtained
Target data feature to be stored;
By the target data characteristic storage to be stored in industrial metadatabase.
Optionally, after the preset model for obtaining each completion training, further includes:
Search instruction is obtained, the search instruction includes the keyword message of model;
According to the keyword message of the search instruction, the model information to match with the keyword message is extracted;
The model information is arranged according to pre-set ordering rule, the model information is loaded into advance
In the visualization interface of setting, the model of the load model information after showing visualization;
The data information that will acquire is input in the model and runs, and generates the operation result of the model;
The data information and the operation result are loaded into pre-set model running report template, mould is generated
Type operation report.
Optionally, the initial parameter of each preset model of configuration, comprising:
Visual configuration data source generates metadata table according to the corresponding metadata mapping relations of the data source, according to
The data of target side are combined by the metadata table;
According to the history parameters in industrial knowledge base, the initial parameter of each preset model is configured.
Second aspect, the embodiment of the present application provide a kind of data processing equipment, comprising:
Module is obtained, the data for obtaining user identify request, carry number to be identified in the data identification request
According to;
Determining module determines data identification industry for the attribute information based on the data to be identified for the user
Model;
Computing module carries out identifying processing for the data to be identified to be input to the data identification industry pattern,
Obtain recognition result;
Feedback module, for the recognition result to be sent to the user.
Data processing method provided in an embodiment of the present invention passes through the number to be identified carried in the data identification request of user
Industry pattern is identified according to data needed for determining user, and data to be identified are input in data identification industry pattern and are known
Not as a result, the result is fed back to user, user can be obtained by required as a result, the application is identified by the data of user
Request can find corresponding data identification industry pattern, shorten user query data identification corresponding with data to be identified
The duration of industry pattern improves the working efficiency of user, identifies industry pattern phase by the data that data identification request obtains
For the data to be identified in data identification request, more targetedly, so that data identification industry pattern treats knowledge
When other data are handled, obtained result is more acurrate.
To enable the above objects, features, and advantages of the application to be clearer and more comprehensible, preferred embodiment is cited below particularly, and cooperate
Appended attached drawing, is described in detail below.
Detailed description of the invention
Technical solution in ord to more clearly illustrate embodiments of the present application, below will be to needed in the embodiment attached
Figure is briefly described, it should be understood that the following drawings illustrates only some embodiments of the application, therefore is not construed as pair
The restriction of range for those of ordinary skill in the art without creative efforts, can also be according to this
A little attached drawings obtain other relevant attached drawings.
Fig. 1 is a kind of flow diagram of data processing method provided by the embodiments of the present application;
Fig. 2 is a kind of flow diagram of industry pattern method of commerce provided by the embodiments of the present application;
Fig. 3 is the flow diagram that a kind of training data provided by the embodiments of the present application identifies industry pattern method;
Fig. 4 is the process signal for the running environment configuration that a kind of data provided by the embodiments of the present application identify industry pattern
Figure;
Fig. 5 is a kind of flow diagram for constructing preset model method provided by the embodiments of the present application;
Fig. 6 is a kind of flow diagram for constructing sample database method provided by the embodiments of the present application;
Fig. 7 is a kind of flow diagram of method for constructing industrial metadatabase provided by the embodiments of the present application;
Fig. 8 is a kind of flow diagram of model running report-generating method provided by the embodiments of the present application;
Fig. 9 is the flow diagram that a kind of pair of preset model provided by the embodiments of the present application configures initial parameter;
Figure 10 is a kind of flow diagram of the construction method of industrial knowledge mapping provided by the embodiments of the present application;
Figure 11 is a kind of structural schematic diagram of data processing equipment provided by the embodiments of the present application;
Figure 12 is a kind of structural schematic diagram of computer equipment 1200 provided by the embodiments of the present application.
Specific embodiment
To keep the purposes, technical schemes and advantages of the embodiment of the present application clearer, below in conjunction with the embodiment of the present application
Middle attached drawing, the technical scheme in the embodiment of the application is clearly and completely described, it is clear that described embodiment is only
It is some embodiments of the present application, instead of all the embodiments.The application being usually described and illustrated herein in the accompanying drawings is real
The component for applying example can be arranged and be designed with a variety of different configurations.Therefore, below to the application's provided in the accompanying drawings
The detailed description of embodiment is not intended to limit claimed scope of the present application, but is merely representative of the selected reality of the application
Apply example.Based on embodiments herein, those skilled in the art institute obtained without making creative work
There are other embodiments, shall fall in the protection scope of this application.
In view of army's simulated training system is that single-chip microcontroller is arranged by embedded mode to set in operation in the prior art
In standby, and then the operation data of warfare equipment is acquired, single-chip microcontroller is arranged by embedded mode and needs to change warfare equipment
Dress, the warfare equipment after repacking is in actual combat in application, can adversely affect to actual combat result.Based on this, the present invention is implemented
Example provides a kind of fixed structure, sensor and simulation system, is described below by embodiment.
As shown in Figure 1, the embodiment of the present application provides a kind of data processing method, comprising:
101, the data identification request of user is obtained, carries data to be identified in the data identification request;
Here, data identification request can be the requests such as cutter life prediction, cutter classification, and data to be identified are to be identified
The data of object.
For example, data identification request is cutter life prediction, object to be identified is cutter A, and data to be identified are cutter A
Date of manufacture, thickness, width, length etc..
102, based on the attribute information of the data to be identified, determine that data identify industry pattern for the user;
Here, attribute information is the information for characterizing data to be identified, e.g., cutter life prediction, cutter classification etc., number
It is the model for meeting user demand and completing training according to identification industry pattern.
Specifically, including multiple categorical data identification industry when determining that data identify industry pattern, in default view
Model finds data identification work corresponding with the attribute information of data to be identified in the data identification industry pattern of multiple classifications
Industry model, wherein default view is the interface for showing user's information needed.
It include classification in default view is cutter for example, the attribute information of data to be identified is cutter life prediction
The data identification industry pattern and classification of life prediction are that the data of cutter classification identify industry pattern, according to data to be identified
Attribute information, determining data identification industry pattern are that the data that classification is cutter life prediction identify industry pattern.
103, the data to be identified are input to the data identification industry pattern and carry out identifying processing, obtain identification knot
Fruit;
Here, recognition result includes the calculated result of data identification industry pattern.
Specifically, including at least one parameter in data identification industry pattern, according to each number in data to be identified
According to corresponding parameter in data identification industry pattern is assigned to, based on the parameter value after assignment in data identification industry pattern, meter
It counts according to identification industry pattern, obtains recognition result.
104, the recognition result is sent to the user.
Specifically, recognition result can feed back to user by being displayed in the user interface, or by by recognition result
The mobile terminal for being sent to user feeds back to user.
Number needed for the data to be identified carried in the data identification request that the embodiment of the present invention passes through user determine user
According to identification industry pattern, and data to be identified are input in data identification industry pattern and obtain recognition result, the result is anti-
Feed user, and user can be obtained by required as a result, the application can be found pair by the identification request of the data of user
The data identification industry pattern answered, shortens the duration of user query data identification industry pattern corresponding with data to be identified,
The working efficiency for improving user identifies that industry pattern is identified relative to data by the data that data identification request obtains and requests
In data to be identified for, more targetedly so that data identification industry pattern is when handling data to be identified,
Obtained result is more acurrate.
In a step 102, the attribute information based on the data to be identified determines that data identify work for the user
Industry model, as shown in Fig. 2, this application provides a kind of industry pattern method of commerce, comprising:
201, obtain the right to use information of preset model corresponding with the attribute information;
202, according to the right to use information, determine that the user completes the transaction of servitude to the preset model, it will
The preset model for completing the transaction of servitude identifies industry pattern as the data.
Here, preset model is the model pre-set, and the right to use information of preset model includes but is not limited to: single
Use and use price, be used for multiple times and it is corresponding use price, monthly payment use and use price, packet use and use valence year
Lattice, transfering use right and transfer price etc., transaction of servitude include: that the power that is intended for single use trades, power transaction is used for multiple times, monthly payment makes
With power transaction, packet year transaction of servitude, transfering use right transaction etc..
Specifically, user selects according to self-demand after user gets corresponding preset model according to attribute information
The corresponding right to use information of the preset model shows that corresponding transaction is believed in default view according to the right to use information of user's selection
Breath, user trade according to the Transaction Information, and the preset model is exactly the data identification that user can be used after the completion of transaction
Industry pattern.
For example, getting preset model according to attribute information, user clicks the list in the right to use information of the preset model
It is secondary to use characterized virtual push button, the transaction interface that the preset model is intended for single use, user's foundation are shown in default view
The prompt of transaction interface is transferred to the use price of single use to corresponding account, determines that corresponding account receives the single and makes
Using after price, determining that the user completes the transaction of servitude to the preset model, then the preset model is exactly that user can
Industry pattern is identified with the data used.
The embodiment of the present application realizes the market-oriented mode of doing business and business model of preset model by using power transaction
Conversion enhances the utilization rate of enterprise digital assets, improves the benefit of enterprise.In this way, all kinds of numbers of each enterprise can be made
According to tangible digitlization assets are become, the transfer transformation efficiency of digitlization assets is promoted, benefit is created for enterprise, is conducive to enterprise
Digitize the effective use of assets.
It is after the user determines that data identify industry pattern, as shown in figure 3, being instructed according to following manner in step 102
Practice the data and identify industry pattern:
301, training data is obtained from sample database;
302, it is based on the training data and multiple preset models, configures the initial parameter of each preset model;
In step 302, for each preset model, the initial parameter of each preset model is configured, comprising the following steps:
Step 1, according to preset model, the data of multiple and different features are obtained.
Specifically, for example, the difference of acquisition is special if multiple model will be handled the digital asset of real estate industry
Levying data can be with are as follows: usable floor area, occupied area, pattern, floor etc..
Step 2, data are arranged, forms matrix model
Specifically, using the feature of data as column, the specific value of data is made when the multiple data that will acquire are arranged
For row, matrix model is formed, for example, first row can after the usable floor area that will acquire, occupied area, pattern, floor are arranged
Think usable floor area, second is classified as occupied area, and third is classified as pattern, and the 4th is classified as floor, usable floor area, occupied area, lattice
Office, the corresponding specific value of floor are row, constitute matrix model.
Step 3, to matrix model carry out elasticity distribution formula data set (Resilient Distributed Datasets,
RDD) Data Serialization is handled, and is associated analysis to the RDD of matrix model, and the degree of association is selected to be less than the square of default association angle value
The RDD of battle array model is combined, and obtains the corresponding training data of industry pattern.
Specifically, analyzed by feature of the trained algorithm to the RDD of matrix model, judgment matrix model
The degree of association of the feature of RDD, for example, the degree of association of the corresponding RDD of fisrt feature RDD corresponding with second feature is judged, if should
The degree of association is greater than default association angle value, then it is assumed that the corresponding RDD of fisrt feature RDD corresponding with second feature can be replaced mutually
In generation, instructs industry pattern using the training data that the corresponding RDD of fisrt feature RDD corresponding with second feature is combined
When practicing, the training result inaccuracy of industry pattern can be made therefore when obtaining training data, to close to the RDD of matrix model
Connection analysis carries out screening and optimizing to training data, improves the accuracy rate of industry pattern training result.
Further, obtain training data when, available multiple and different training data, by different training datas into
Row Conjoint Analysis is therefrom selected optimal training data and is trained to industry pattern, and the standard of industry pattern training is improved
True rate, wherein the training data that optimal training data is minimum for the degree of association of data characteristics.
303, the training data is configured in each preset model, each preset model is trained, is obtained
To each preset model, the corresponding training result of each preset model and model accuracy rate for completing training;
Training data includes test data and processing result, specifically, step 302 includes:
Obtain preset model and corresponding test data.
The test data corresponding in test data is labelled with processing result for testing preset model.
Test data is configured in preset model, preset model calls pre-set polyalgorithm respectively to test number
According to being calculated, multiple calculated results are obtained.
Specifically, polyalgorithm is respectively handled test data, multiple calculated results are obtained.
The multiple calculated results and corresponding processing result of test data are analyzed, accuracy rate highest in calculated result is selected
Benchmark algorithm of the algorithm as preset model, and the calculated result that benchmark algorithm is calculated is as the training result.
The calculated result that each algorithm is obtained is for statistical analysis with processing result, obtains the corresponding calculating knot of the algorithm
The accuracy rate of fruit, and so on, statistical analysis obtains the accuracy rate of each algorithm in pre-set polyalgorithm, and selection is accurate
Benchmark algorithm of the highest algorithm of rate as preset model, and the calculated result that benchmark algorithm is calculated is tied as training
Fruit.
304, judge whether the training result meets preset condition, if not satisfied, then according to the training result tune
The parameter of whole each preset model carries out retraining to the preset model after adjusting parameter, until the training knot
Fruit meets preset condition;
Specifically, the training result of preset model includes the accuracy rate of benchmark algorithm in preset model, if the training result
More than or equal to scheduled value, then it is assumed that the training result meets preset condition, then preset model training terminates;If training result
Less than scheduled value, then it is unsatisfactory for preset condition, the parameter of preset model is adjusted according to training result, specifically, based on training
As a result, preset model can adjust automatically parameter, to after adjusting parameter preset model carry out retraining, during retraining,
The benchmark algorithm for reselecting preset model is tied the calculated result of the benchmark algorithm reselected as the training of preset model
Fruit, until training result meets preset condition, by being constantly trained to preset model, algorithm in preset model
Accuracy rate is constantly promoted, and the operation result of preset model constantly adjusts, and improves the precise degrees of preset model operation result.
305, it is based on the corresponding model accuracy rate of each preset model, from each preset model for completing training, described in determination
Data identify industry pattern.
Specifically, the corresponding preset model for completing training of the model accuracy rate for characterizing maximum value is determined as described same
Default corresponding data of classifying identify industry pattern.
The example for continuing the calculating preset model accuracy in step 102, according to prediction model A1Accuracy be 80%,
Prediction model B1Accuracy be 70%, prediction model C1Accuracy be 85%, accuracy in three preset models is maximum
Be prediction model C1, then prediction model C1It is object module needed for user.
The embodiment of the present invention is by providing a kind of preset model training method comprising: obtain preset model and training
Data configure the initial parameter of preset model;Training data is configured in the preset model, preset model is instructed
Practice, obtains training result;Whether training of judgement result meets preset condition, if not satisfied, then being adjusted according to training result pre-
If the parameter of model, retraining is carried out to the preset model after adjusting parameter, until training result meets preset condition.This
A kind of preset model training method provided is provided, preset model is trained using training data, during training
The parameter of continuous adjustment preset model, is continuously improved the accuracy of preset model, so that the preset model convergence after training, and
Suitable for industrial circle, this method has generality, can be applied to each industrial circle, and by update training data energy and
When preset model is updated, extenuated in the prior art, it is existing not have to the method that preset model is trained
Generality can only meet the needs of fixed production, and cannot timely be trained to preset model, so that preset model
The technical issues of cannot effectively being updated for a long time.
As shown in figure 4, step 102 it is described for the user determine data identify industry pattern after, further includes:
401, according to data identification request using the first number of users of data identification industry pattern, know for the data
Other industry pattern configures the first running environment;
In the embodiment of the present application, after user logs in the comprehensive transaction platform of industry pattern, packet is inputted from the search box of displaying
Use demand containing key word information, the comprehensive transaction platform of industry pattern analyze the identification request of received data.As
One alternative embodiment, user it should be understood that the equipment of a certain model can also operate normally how long, thus, the identification of the data of input
Request may is that model XXXX.
In step 401, according to data identification request using the first number of users of data identification industry pattern, for institute
It states data identification industry pattern and configures the first running environment, comprising:
Data when counting current request identify industry pattern number;
Count the corresponding number of users of each data identification industry pattern;
According to the corresponding number of users of each data identification industry pattern, data identification industry pattern number, data identification industry
Types of models, current available resource and pre-set running environment configuration strategy, for the first operation of industry pattern configuration
Environment.
Here, if running environment configuration strategy may is that data identification industry pattern, corresponding number of users is more, and data are known
Other industry pattern type is the type of operation complexity, and current available resource is more, and data identify that industry pattern number is fewer, for the number
The corresponding resource of the first running environment according to identification industry pattern configuration is also more.
Data identification industry pattern type includes but is not limited to: equipment class industry pattern, production class industry pattern, Decision Classes
Industry pattern, enterprise-class industry pattern.For example, for Decision Classes industry pattern, operation is complex.
In the embodiment of the present application, as an alternative embodiment, each number is carried out by industry pattern engine atom computing unit
According to the integration and allotment of the running environment of identification industry pattern, the virtual rear Taiwan investment of industry pattern distribution is identified for the data of operation
Source space, forms the running environment comprising Virtual Space, and the running environment of each data identification industry pattern is mutually indepedent.And at this
Data identify in the subsequent operation of industry pattern, and the application and calling of resource are carried out with Virtual Space.
402, according to data identification request using the second user number of data identification industry pattern, the data are known
Other industry pattern virtually turns to the several industry pattern copies of the second user;
Specifically, being virtually the industry pattern copy to match with the second user number by target industry pattern.It is each
Use the corresponding industry pattern copy for using the target industry pattern of the user of the target industry pattern.Wherein, described in login
Industry pattern integrates transaction platform, from the tabulation of displaying, chooses target industrial classification, according to the target point shown
Next stage target classification is chosen in the corresponding next stage tabulation of class, arrives required target industry pattern until choosing.
403, it is the user configuration according to the corresponding user information of the second user number and the first running environment
Second running environment;
It in step 403, is described according to the corresponding user information of the second user number and the first running environment
The second running environment of user configuration, comprising:
According to using the target industry pattern each user rank and permission, each user configuring condition and
The use price that each user can bear carries out overall merit, the second running environment is respectively configured for each user, for institute
The second running environment and first running environment for stating each user configuration match, the second fortune configured between each user
It is mutually indepedent between row environment.
Specifically, sharing the algorithm and resource of target industry pattern, user's input between data identification industry pattern copy
Data identify that the data of industry pattern copy are mutually isolated.
In the configuration feelings of the rank according to each user using the target industry pattern and permission, each user
After the use price that condition and each user can bear carries out overall merit, the second operation is respectively configured for each user
Before environment, this method further include:
Determine whether the user completes the transaction of servitude to the target industry pattern, if so, being described in executing
The step of the second running environment is respectively configured in each user.
Transaction of servitude includes: that power transaction is intended for single use, power transaction is used for multiple times, monthly payment transaction of servitude, uses in packet year
Power transaction, transfering use right transaction.
Specifically, can identify the specific requirement and specification of industry pattern according to each data, data identification industry is provided
The use price of model.After user clicks use, the transaction for carrying out the data identification industry pattern right to use is determined, in the right to use
After the completion of transaction, the comprehensive transaction platform of industry pattern is collected the deduction of transaction in the ratio in allocation strategy.And it can be according to pre-
The right to use price distribution side being first arranged, for example, identifying industrial mould to the owner of data identification industry pattern and the data
The developer of type carries out amount of money distribution by allocation strategy.
404, the data identification industry pattern copy of the user configuration is operated in second running environment.
Specifically, the follow-up operation of data identification industry pattern copy is based on the second running environment of the predistribution.
When one industrial classification model prediction engine of multiple user's simultaneous selections to carry out in use, by the industrial classification mould
Type is virtually multiple data identification industry pattern copies concurrently to be executed, to meet the use demand of user, each user
A corresponding virtual data identify industry pattern copy.Wherein, pass through resource between virtual data identification industry pattern copy
Resource-sharing and Data Concurrent are realized in isolation, including but not limited to: sharing model parameters information, user-isolated data resource.With
User data resource includes but is not limited to: the operation control of the target data, the prediction result, user setting of task output of user's input
Parameter processed etc..
When same data identification industry pattern with different users to concurrently being executed, it can construct comprising void
Second running environment in quasi- calculation resources space, and execute the use demand for concurrently executing user.For example, passing through target industry
The parameter of the running environment configuration of model, operation resource needed for obtaining data identification industry pattern, for example, memory size,
CPU/GPU nucleus number, cache size, interactive space etc..Then, integrated by distributed physical resource, construct virtualization resource
Pond, for example, according to current request use industry pattern the first number of users, in conjunction with the concurrency of operation, GPU or CPU type,
The nucleus number of arithmetic element, the size of virtual memory, size of spatial cache etc. select idleness higher from preset resource pool
Resource, formed an independent virtualization subspace, obtain the industry pattern virtual operation resource space (first operation ring
Border).Finally, the virtual operation resource space according to the target industry pattern, is that the corresponding data of each user identify industrial mould
Type copy distributes virtual calculation resources space.In this way, one user, industry pattern, virtual resource space formation multi dimensional resource
Pond virtualizes space, so as to be shown with the mode of three-dimensional space.For example, user is set as x coordinate, industry pattern
It is set as y-coordinate, virtual resource space is set as z coordinate, by three-dimensional system of coordinate, uniquely determines a certain specific user and industry
In resource pool occupied by model the case where virtual resource space.
It is needed with 5 users using 2 industry patterns, wherein 2 users use the first industry pattern, and 2 users use
Second industry pattern, 1 user need while with for the first industry patterns and the second industry pattern, and 5 users respectively correspond
5 points of x-axis, two industry patterns correspond to 2 points of y-axis, for the virtual resource space that user uses industry pattern to distribute, make
For the point of z-axis.Three-dimensional resource pool framework is formd by x, y, z three-dimensional coordinate, resource pool is independent to each user.
In the embodiment of the present application, the running environment of data identification industry pattern is configured according to the demand of user,
It can satisfy the diversified demand of user, and resource utilization can be promoted, improve data identification industry pattern running environment and match
Set efficiency.
As shown in figure 5, constructing preset model according to following manner before being trained to preset model:
501, polyalgorithm is edited in algorithm editing machine, the multiple algorithm is stored in algorithm list;
In step 501, polyalgorithm is edited in algorithm editing machine, polyalgorithm is stored in algorithm list, wrapped
It includes:
Algorithm engine technology is integrated in algorithm editing machine.
By algorithm engine technology, the online editing algorithmic code in algorithm editing machine adjusts algorithmic code online
Examination, online execution, obtain output result online.
When editing algorithmic code in algorithm editing machine, the event attribute of algorithmic code is configured, in algorithmic code
Execute preset operation in operational process automatically according to event attribute.
Specifically, event attribute includes: preposition algorithm, postposition algorithm, runs succeeded, executes failure, executes time-out etc., thing
Part attribute value includes: to send mail, stop algorithm, waiting etc., can be according to preset transmission if event attribute is to execute time-out
This event attribute value of mail makes the operation for sending mail reminder.
The primary attribute of placement algorithm code on algorithm editing machine.
Specifically, primary attribute includes the title of algorithm, the version of algorithm, the creation time of algorithm, parameter list etc.,
The primary attribute of placement algorithm code is convenient for that algorithmic code is managed and is inquired on algorithm editing machine.
Event attribute, primary attribute are saved with corresponding algorithmic code, algorithm is stored in algorithm by formation algorithm
In list.
502, selection algorithm handles preset training data from the algorithm list, obtains the algorithm list
In the corresponding algorithm file of each algorithm;
Specifically, including multiple historical datas for being labelled with processing result in training data, one is selected from algorithm list
A algorithm handles preset training data, obtains the corresponding processing result of the algorithm and algorithm file, in algorithm file
Operation accuracy rate including algorithm is somebody's turn to do specifically, the processing result of the processing result of the algorithm and mark is analyzed
The operation accuracy rate of algorithm;And so on, each of acquisition algorithm list algorithm handles training data, obtains every
The corresponding processing result of one algorithm and algorithm file.
503, multiple algorithm files are compared, a kind of calculation is selected from the algorithm list according to comparing result
Method generates the preset model according to the algorithm of selection.
Specifically, operation accuracy rate, the runing time of algorithm etc. in the algorithm file including algorithm;To polyalgorithm text
Part compares, the operation accuracy rate of algorithm in comparison algorithm file, the highest algorithm of operation accuracy rate of selection algorithm, according to
The algorithm of selection generates preset model.
The industry pattern is for handling digital asset in the embodiment of the present application, this method comprises: compiling in algorithm
It collects in device and edits polyalgorithm, polyalgorithm is stored in algorithm list;Selection algorithm is to preset instruction from algorithm list
Practice data to be handled, obtains the corresponding algorithm file of each algorithm in algorithm list;Algorithm file is compared, according to right
A kind of algorithm is selected from algorithm list than result, preset model is generated according to the algorithm of selection.The present invention is pre- by what is provided
If model generating method, preset model can be constructed by this method, which can convert the digital asset of enterprise
At preset model, and then the industry pattern that can be will convert into is put on model fairground and trades, and enables digital asset
It recycles, has played the market value of digital asset, while improving enterprise to the efficiency of management of digital asset, further in fact
The preservation of digital asset is showed and has spread.
Before obtaining training data in sample database, need to construct sample database, as shown in fig. 6, the application
Provide a kind of method for constructing sample database, comprising:
601, obtain the operation data of equipment;
Here, equipment is the industrial equipment of data to be obtained, and operation data includes the detection dress being arranged on industrial equipment
Set the storage address etc. of the hardware address of collected data, industrial equipment itself and the operating parameter of storage industry equipment
Several in data.The form of above-mentioned data is not unique, can be clock signal or analog signal, can also for it is constant not
The parameter of change, or image.
Specifically, setting data model carries out preliminary treatment, data model packet to the operation data of the equipment acquired
Include source data source and target side data source, source data source linking objective end data source, the fortune of the equipment for parsing encryption
Row data.
Obtain operation data the following steps are included:
Step 1, equipment end collector obtains the operation data of equipment in real time, and by the operation data in the predetermined section time
Carry out packing processing.
Step 2, source data source obtains from equipment end and is packaged treated operation data.
Here, equipment end collector refers to the equipment for acquiring equipment running status in real time close to running equipment installation,
Specifically, the new data that equipment end collector transmission device generates or the data changed, target data
Source obtains the operation data that is packaged that treated from equipment end collector according to preset equipment end collector address;Target
Treated that operation data is sent to source data source by the packing obtained from equipment end collector for data source.
602, according to preset mapping relations, the operation data is mapped as metadata;
Here, preset mapping relations refer to the numerical value of the operation data of equipment and type pass corresponding with metadata
System.Metadata refers to that the feature of the current time equipment operating data corresponding sample data planting modes on sink characteristic in mapping relations (refers specifically to
Characteristic value).
The operation data of equipment is that generally have different forms, that is, more from the collected data of different equipment ends
Source isomeric data.It needs according to preset mapping relations, the data type of operation data and numerical value is mapped to metadata.
603, the metadata is formed into the sample database.
Specifically, sample database stores the corresponding sample data Al Kut of operation data of all devices of last moment
Sign, the specific form of sample data planting modes on sink characteristic are characterized value.The operation data of all devices corresponds to several characteristic values, because
The practical corresponding sample data planting modes on sink characteristic of the operation data of this all devices is feature vector.It is closed according to preset mapping
System, by the address of cache of equipment end at logical implication;According to logical implication and metadata, sample database is formed.Above-mentioned mapping
Relationship further includes address feature vector the depositing in sample database corresponding with the operation data that the equipment generates of equipment end
Storage space is set.According to the address of equipment end, the corresponding feature vector of operation data of current time equipment according to above-mentioned mapping pass
System is put into the corresponding address of sample database, updates the feature of the sample database of the equipment, forms new sample database.
The embodiment of the present application is used the feature of the operation data Feature Mapping of equipment to sample database, forms sample number
According to the method in library, above-mentioned complicated and diversified industrial data is identified with corresponding technology not yet in the prior art, is located
Reason, and then the method that can not form sample database is compared, which form standard identification models, improve the operation data from equipment
The middle efficiency for obtaining useful information.
Metadata is stored in industrial metadatabase, as shown in fig. 7, the embodiment of the present application also provides a kind of building industry
The method of metadatabase, comprising: 701, obtain the initial data of data source.
Data source includes but is not limited to ordinary file, distributed file system (HDFS, Hadoop Distributed
File System) the different file system such as file, MySQL, Hbase, MongoDB etc. different databases.From data source
It can be initial data auto-configuration data source attribute when obtaining initial data, which mainly includes the source of initial data
Location, source address uses URL format (Uniform Resource Locator, uniform resource locator), including the original number
The port that is used according to title or IP address, the data source of place data source and the path for reaching this initial data and original
The title of data itself.For example, the source address of configuration is HDFS: //xxx.xxx.x.x/a and jdbc:mysql: //
x.x.x.:3306/db。
702, the source address that parsing initial data carries obtains the detailed data information of initial data.
In a step 702, the source address that parsing initial data carries, obtains the detailed data information of initial data, wraps
Include following steps:
Step 1, the source address that parsing initial data carries.
Parsing to source address includes determining that the source address whether there is, and whether source address can connect, if
Possess the permission for reading file.
Step 2, the mapping relations for inquiring source address and data source connection type, with obtaining the source of initial data carrying
The data source connection type of location mapping.
Step 3, according to obtained data source connection type, the detailed data information of initial data is obtained.
Different data sources extracts initial data, ordinary file or HDFS file according to the data source connection type of mapping
The asynchronous extraction of muti-piece can be splitted the file into, text file extracts data using mr mode and cleaning text file, HDFS file are adopted
Data are extracted with spark mode;The data sources such as Hbase, MongoDB extract in such a way that common SQL is directly inquired meets item
The data of part, the databases such as MySQL, Hbase extract data by sqoop mode.
703, data source belonging to initial data and data structure are determined, according to the data source and data structure determined
The data characteristics of mapping extracts the field to match with the data characteristics of mapping from detailed data information, obtains initial data
Data characteristics.
Due to including bulk information content in detailed data information, if subsequent data storage management can be made by all receiving
At difficulty, therefore according to the data characteristics of pre-set initial data, detailed data information is screened, is extracted required
Data characteristics, wherein data characteristics includes but is not limited to data field, data constraint, data value, data type and data knot
Structure etc..Data type includes but is not limited to Int, long, String, varchar, timestamp, date etc.;Data structure packet
It includes but is not limited to, Xml, json, Parquet, keyvalue, lp, datagrid etc..
704, according to pre-set target metadata mapping table, the data characteristics of initial data is extracted, is obtained
Target data feature to be stored.
From the data characteristics of initial data, it is special to extract the data to match with the data characteristics of target metadata mapping table
Sign.
The basic data type of metadata in data source is defined in advance, and number of targets is defined using object oriented language
According to data type, by the data type of any metadata by Protobuf protocol conversion be target data data type,
And be associated the data type of metadata, the data type of target data by structure I D, it is stored in target metadata and reflects
In firing table.Other than carrying out data type conversion, the conversion between the data characteristicses such as data constraint can also be carried out.
The data characteristics of initial data includes data structure, the data structure of target metadata mapping table is chosen, from original
In the data characteristics of data, the data structure to match with the data structure chosen is extracted, to obtain target data to be stored.Such as
Fruit user needs to carry out data certain processing, some secondary data structures is provided in the embodiment of the present invention, it is only necessary to input
Data, so that it may generate some other indexs for needing to calculate by this data or some reports according to secondary data structure
The output of sheet form.
705, will target data characteristic storage be stored in industrial metadatabase.
According to the attribute of pre-set target data, for target data configuration attribute, the storage including target data
Location, target data can the essential attributes configurations such as method of calling, the path URL of target data, account number cipher.It is divided into target data
Static data and dynamic data, as device id, equipment march into the arena the time as static data;Time series data is dynamic data, is being stored
When dynamic data, need to add the information that data structure is keyvalue, data field is timestamp.The industry built
Metadatabase can will be combined from the data that the Data Integration of different data sources, different type structure is same data structure
Body facilitates storage management and additions and deletions to change the application service looked into.
In industrial circle, by from different data sources and the different data of data structure are known as multi-source heterogeneous data, in the future
From same data source but the different data of data structure are known as complicated isomeric data.
When initial data is multi-source heterogeneous data, a specific embodiment is lifted, multi-source heterogeneous data are respectively from MongoDB
Data source, MySQL data source and ftp data source obtain the detailed data information of initial data, press by parsing source address
According to the data characteristics of pre-set initial data, corresponding data characteristics is extracted from detailed data information, it is specific such as table (1)
It is shown:
Table (1)
Multi-source heterogeneous data are converted into HDFS data by metadata mapping table, specific as shown in table (2):
Table (2)
The target data for being integrated into same data structure is stored in industrial metadata.
When initial data is complicated isomeric data, a specific embodiment is lifted, complicated isomeric data comes from ftp data source,
Data including two kinds of structures of xml format and json format.By parsing source address, the detailed data letter of initial data is obtained
Breath, according to the data characteristics of pre-set initial data, extracts corresponding data characteristics, specifically such as from detailed data information
Shown in table (3):
Table (3)
It by complicated M IS is HDFS data by metadata mapping table, specific as shown in table (4):
Table (4)
The target data for being integrated into same data structure is stored in industrial metadata.
As shown in figure 8, after the preset model for obtaining each completion training, further includes:
801, search instruction is obtained, search instruction includes the keyword message of model.
The invisible digital asset conversion of industrial circle is trained to data identification industry pattern by user, is sent out in the market in model
Cloth data identify industry pattern, and data identify that industry pattern after the approval, can be shown in model city through working platform personnel
On field.Classify according to keyword message to the data identification industry pattern in model market, wherein keyword message includes
Algorithm, technical field, the scope of application etc. for constructing data identification industry pattern, such as equipment class, predictive maintenance, discrete manufacturing business
Correlation, Flow Manufacturing correlation etc..Model market identifies the algorithm of industry pattern by obtaining building data, generates corresponding
Algorithm icon;The training data and test data for obtaining data identification industry pattern, generate corresponding data model icon, by
Algorithm icon and data model icon collectively form data identification industry pattern in model corresponding icon in the market.General mould
Type is shown as default layout, but the model display scheme of different patterns, model display side are provided in the backstage in model market
Case is divided into free and two kinds of forms of charge can personalized application in the background after user's selection or purchasing model exhibition scheme
Model display scheme carries out model display.In addition to this, user can carry out model display in the background according to their own needs
Layout customization, can also be according to the development interface of industry pattern marketplace platform, user's self-developing model display layout, and issues
It is used into backstage for other users.
Model market display data identifies industry pattern, and provides the value-added service of data identification industry pattern, passes through use
Use of the family to data identification industry pattern identifies that price, the resource situation used and the data of industry pattern are known according to data
The feature of other industry pattern, data identify that the industry pattern owner obtains the income of the right to use of model, and industry pattern market is flat
The expense that platform obtains the price difference of data identification industry pattern transaction and resource uses, the industry pattern market for providing a kind of opening are flat
The business model of platform sustainable growth.Data identification industry pattern is user to the pre- of lesser value showing before model market
What if model refinement obtained, it is higher by being worth possessed by improved data identification industry pattern, therefore, improved number
Preset model price of the price than before after uploading to model market according to identification industry pattern is high, solves preset model value
Low problem improves user's income.802, according to the keyword message of search instruction, extracts from database and believe with keyword
The matched model information of manner of breathing.
803, model information is arranged according to pre-set ordering rule, model information is loaded into and is preset
Visualization interface in, show visualization after stress model information model.
Include the information that user uploads model time in model information, arranged from the near to the distant according to model time is uploaded,
User search can be allowed to identify industry pattern to the newest data for uploading to platform;It include the download time of model in model information
Information can allow user search more to access times are downloaded by other users according to download time information by mostly and less arranging
Data identify industry pattern;Further include the pricing information of model in model information, carries out ascending order or drop according to pricing information
Sequence arrangement.
Optionally, model information includes model prices information, generate model prices information the following steps are included:
Step 1, the actual amount of data of model prediction is obtained from model information, the algorithm that model uses.
Step 2, the predictablity rate of computation model.
Step 3, according to the comprehensive of the actual amount of data of model prediction, developing algorithm and predictablity rate computation model
Point.
It step 4, include the evaluating deg and positive rating of model in model information, according to evaluating deg, positive rating and comprehensive score
The overall cost of computation model.
Wherein, the evaluating deg of model is to carry out weight calculation to the predictablity rate of user's scoring and model to obtain, and is used
Family scoring be ten point system, user scoring be at least one point, [1,6) commented for difference, [6,8) be in comment, [8,10] be favorable comment.
Step 5, it according to overall cost, calculates and generates model prices information.
804, the data information that will acquire, which is input in model, to be run, and generates the operation result of model.
805, data information and operation result are loaded into pre-set model running report template, model fortune is generated
Row report.
The generation of model running report template in step 805 specifically includes the following steps:
Step 1, developing algorithm is determined, the attribute information of developing algorithm is arranged in automatic patching system template.
Step 2, metadata configurations, developing algorithm and knowledge base model are obtained.
Step 3, determine metadata configurations, developing algorithm and knowledge base model RDD (elasticity distribution formula data set,
Resilient Distributed Datasets) relationship.
A specific embodiment, metadata configurations to industrial algorithm 1 are lifted, industrial algorithm 1 arrives industrial algorithm 2, and industrial algorithm 2 arrives
Knowledge base model.
Step 4, according to RDD relationship and attribute information, model running report template is generated.
Step 5, storage model runs report template.
In addition to the model running report template of the default style, user can be with personalized customization template, so that each template
Report style and content it is different.
The update of data information can be divided into following three kinds of situations: transport again after model is by adjusting parameter or input data
Row, model running public lecture are updated automatically on the basis of original report;When the algorithm of model or the data of model are automatic
After matching, model running public lecture automatic adaptation updates;When the user of model change or running environment change, model running
Report also can automatic adaptation update.
Model running report-generating method provided by the embodiments of the present application, user are uniformly stored in model by retrieving to use
The normal data identification industry pattern being converted by the invisible digital asset in industrial circle in market, not only makes enterprise
All kinds of soft assets of experience become the digitlization assets having the honor, and also help the permanent preservation and circulation of enterprise digital assets.
Meanwhile model market also avoids false, ropy data identification industry pattern and fishes in troubled waters in the market, illegally in model
Profit, can not be managed collectively and commercialized technical problem with solving digitlization assets existing in the prior art.
As shown in figure 9, the initial parameter of each preset model of configuration, comprising:
901, visual configuration data source generates metadata table according to the corresponding metadata mapping relations of the data source,
According to the metadata table, the data of target side are combined;
Wherein, data source, connection attribute, target data, data characteristics etc. are included in metadata.
Specifically, in industry pattern management backstage by way of dragging, disposition data source is closed according to the mapping of metadata
System, the corresponding metadata of the data source is mapped, obtains metadata table after mapping result is arranged, target side need by
When the data of the corresponding target side of metadata are combined and configure, fast and flexible can be carried out by above-mentioned metadata table
Mapping.
Wherein, in disposition data source, if data source is a certain information system, knowledge base or ftp server etc.,
Corresponding connection attribute, such as the address URL, the data connection of Lai Jianli source to target side will be configured.
Above-mentioned steps 901: visual configuration data source is generated according to the corresponding metadata mapping relations of above-mentioned data source
The data of target side are combined by metadata table according to above-mentioned metadata table, comprising: when above-mentioned data source is complicated isomery
It is when data source, the multiple data sources in above-mentioned complicated heterogeneous data source are whole according to respective connection type and reading manner progress
It closes, obtains complicated heterogeneous data source allocation plan.
Specifically, it when above-mentioned data source is composed of the data source of a variety of different data features, needs first to obtain
Take the corresponding processing method of each data source, i.e. connection type and reading manner.And all processing methods are integrated into one
Kind is directed to the allocation plan of the data source.
According to above-mentioned complicated heterogeneous data source allocation plan, above-mentioned complicated heterogeneous data source is configured.
Specifically, the configuration of corresponding connection attribute is carried out to data source according to above-mentioned allocation plan.
According to the multiple data sources in above-mentioned complicated heterogeneous data source, data source assembly is generated.
Specifically, the multiple data sources in above-mentioned complicated heterogeneous data source are mapped in such a way that metadata maps same
Kind data source, then remerges into data source assembly.
Above-mentioned data source assembly and auxiliary data source feature are combined, the corresponding first number of the data source assembly is generated
According to.
Specifically, after data source assembly and auxiliary data source feature being combined, the member of data source assembly is generated
Data.
According to the corresponding metadata mapping relations of the data source assembly, assembly metadata table is generated.
Specifically, it is mapped for the metadata of above-mentioned data source assembly, result is organized into combination voxel data
Table.
According to the assembly metadata table, the data of target side are combined.
Specifically, target side need by the data of the corresponding target side of the corresponding metadata of above-mentioned data source assembly into
When row combination and configuration, the mapping of fast and flexible can be carried out by above-mentioned metadata table.
902, according to the history parameters in industrial knowledge base, the initial parameter of each preset model is configured.
Specifically, training parameter is stored in industrial knowledge base, it can be according to history parameters to industrial mould before model training
Type carries out preliminary configuration.
Data identify to include industrial knowledge mapping in the report generated after the completion of industry pattern operation, as shown in Figure 10, this
Application provides a kind of construction method of industrial knowledge mapping, comprising:
1001, obtain the feature vector of pending data.
In the embodiment of the present application, pending data can be the operation data in industrial equipment, such as the sensing in equipment
Several in the data of device acquisition, the hardware address of equipment and device memory in the data such as data for storing.It is to be processed
Data can also be the parameter of part or product, by taking cutter as an example, can be size, material and sharpness of cutter etc.
Several in parameter.Pending data is multi-source heterogeneous data, can not directly as the data that algorithm model is directly handled, because
Pending data, is mapped to the form of feature vector by this.This implementation is provided for predicting cutting-tool's used life below
The construction method of industrial knowledge mapping be described.
In step 1001, the feature vector of pending data is obtained, comprising the following steps:
Step 1, data model is established, wherein data model includes source data source, target side data source and source number
According to the mapping relations in source and target end data source.
Step 2, pending data is obtained from source by source data source.
Step 3, the mapping relations based on source data source and target side data source obtain the target of pending data mapping
End data source obtains the feature vector of pending data.
Specifically, the source data source of data model connects source, wherein when pending data refers to equipment, source can
Think memory, file or the database of storage device parameter, or acquisition component;Pending data nulling part or
When product, source can be storage product or memory, file or the database of Parameters of The Parts.From target side data source
The feature vector of generation can be obtained.The mapping relations of source data source and target side data source refer specifically to handle pending data
At the configuration of feature vector.Above-mentioned mapping relations are stored in the metadatabase of data model.When source data source is adopted from source
Collect pending data, metadata map component transfers above-mentioned mapping relations in metadatabase in data model, by pending data
It is mapped to feature vector.
Source can upload the file of the dimension information of the type cutter with user, edge length, the length of cutter hub such as cutter
Several parameter informations in the length of degree, the width of cutter hub and hilt.Source is also possible to the cutter stored in database
Material information, such as the material of cutter hub and the material of hilt.Source data source obtains above- mentioned information, the metadata in data model
Component is closed by the mapping relations of source data source and target side data source in calling metadatabase as a kind of possible mapping
System, the dimension information of all cutters corresponding characteristic value in above-mentioned mapping relations, as alternatively possible mapping relations,
The length of the length of cutter hub, the width of cutter hub and hilt a corresponding characteristic value, blade of cutter in above-mentioned mapping relations
Length corresponds to another characteristic value in above-mentioned mapping relations.By obtained characteristic value according to the sequential concatenation of setting, shape
At high-dimensional feature vector.
1002, establish simultaneously initialization algorithm model, wherein algorithm model includes the first deep learning network and the second depth
Learning network.
In the embodiment of the present application, feature vector is trained by the first deep learning network of algorithm model, shape
At knowledge (feature for extracting data), the second deep learning network that above-mentioned knowledge brings algorithm model into is trained, and is formed new
Knowledge (extracting new feature), and establish contacting between above-mentioned knowledge and above-mentioned new knowledge.
As an optional implementation manner, the method for initialization algorithm model includes (1) and (2), is specifically included:
(1) by the index configurations of pending data in the first deep learning network and the second deep learning network.
(2) training algorithm of the first deep learning network and the training algorithm of the second deep learning network are set.
Specifically, the index of pending data refer to pending data in the address of source, the first deep learning network and
The second available pending data of algorithm layer is trained, and forms knowledge.First deep learning network and the second algorithm layer obtain
The pending data taken can be different.In the page for configuring the first deep learning network, the index of pending data is configured, is passed through
It pulls algorithm and completes algorithm configuration.The configuration method of second deep learning network is identical, the algorithm of configuration and the first deep learning
The placement algorithm of network is different.
1003, based on the feature vector of the first deep learning network and pending data, generate the first blocks of knowledge.
Step 1003 specifically includes as follows: the feature vector of pending data being input to the first deep learning network, is obtained
Obtain model training result.Model training result is input in knowledge base, instructs knowledge base according to model training result and model
Practice the mapping relations of result and blocks of knowledge, generates the first blocks of knowledge.
Specifically, the first deep learning network is used to extract the feature of pending data, prediction result refers to number to be processed
According to assessment grade, the first blocks of knowledge refers to the corresponding evaluation of the assessment grade of pending data.Assessment cutting-tool's used life
Length in short-term, the parameter (several parameters such as size, sharpness of material and blade) for acquiring cutter is input to the first depth
Learning network is trained, and output indicates the symbol of the assessment grade of the length of cutting-tool's used life, such as opinion rating symbol
Including tetra- grades of A, B, C, D.It stores the corresponding evaluation of opinion rating in knowledge base, is grown very much as A corresponds to service life,
It is very longer that B corresponds to service life, and C corresponds to that service life is general, and it is short that A corresponds to service life.
1004, it is based on the second deep learning network and the first blocks of knowledge, generates the second blocks of knowledge and the first knowledge list
The weighted value of member and the second blocks of knowledge.
As an alternative embodiment, the feature vector of the first blocks of knowledge and pending data is input to second
The training layer of deep learning network obtains the second blocks of knowledge of training layer output;By the first blocks of knowledge and the second knowledge list
Member is input to the weighted value generation layer of the second deep learning network, generates the weight of the first blocks of knowledge and the second blocks of knowledge
Value.
Specifically, the second blocks of knowledge refers to the new blocks of knowledge that the first blocks of knowledge combination pending data is formed.
The feature vector of the first blocks of knowledge generated in step 1003 and pending data is input to the second deep learning network
Training layer is trained again, forms the second blocks of knowledge.By blocks of knowledge and the new blocks of knowledge (instruction of the second deep learning network
Practice the training result of layer) it is input to the weighted value generation layer of the second deep learning network, the second deep learning network exports weight
Value.Here weighted value indicates the tightness degree between two blocks of knowledge, and weighted value is bigger, the pass between two blocks of knowledge
It is closer.
After obtaining the evaluation of cutting-tool's used life, by the data and evaluation with cutter life length, input the
The training layer re -training of two deep learning networks, obtains more accurate evaluation, such as the influence factor of the length of service life.?
The influence factor of the length of the evaluation and service life of cutter life length is input to weight generation layer simultaneously, obtains cutter
The weighted value for contacting tightness degree of the influence factor of the length of the evaluation and service life of service life length, numerical value is bigger,
Tightness degree is higher.
As another optional embodiment, is input to by the first blocks of knowledge and with the feature vector of other data
The training layer of two deep learning networks obtains the second blocks of knowledge of training layer output.
After obtaining the evaluation of cutting-tool's used life, such as by the noise data of mechanical movement and above-mentioned evaluation, input the
The training layer re -training of two deep learning networks, the noise generated when obtaining using cutter.Cutter life length
Noise that evaluation and using cutter when generate while it being input to weight generation layer, obtain the evaluation of cutter life length and is made
The weighted value of the noise generated when with cutter.
1005, according to weighted value, generate the industry of the digraph comprising being directed toward the second blocks of knowledge by the first blocks of knowledge
Knowledge mapping.
Specifically, the first blocks of knowledge and the second blocks of knowledge respectively represent the knowledge list after blocks of knowledge and training
Member.Blocks of knowledge enabled node itself indicates that the blocks of knowledge after training is obtained by blocks of knowledge training, the stream of knowledge
Dynamic direction is from the blocks of knowledge after blocks of knowledge flow direction training, and usable direction indicates, the big little finger of toe blocks of knowledge of weighted value and instruction
The close relation of blocks of knowledge after white silk can be indicated with the length of line.According to the above method, after forming blocks of knowledge and training
Blocks of knowledge digraph.Constantly according to blocks of knowledge after being trained knowledge, according to weighted value building know
The digraph of blocks of knowledge after knowing unit and training, can form knowledge mapping.User can choose the displaying side of knowledge mapping
Formula, wherein if exhibition method can be that perhaps mind map exhibition method is topological diagram or mind map for topological diagram, icon,
User can edit knowledge mapping by towing, and edited knowledge mapping can export.
As shown in figure 11, the embodiment of the present application provides a kind of data processing equipment, comprising:
Module 1101 is obtained, the data for obtaining user identify request, carry in the data identification request wait know
Other data;
Determining module 1102 determines that data identify for the attribute information based on the data to be identified for the user
Industry pattern;
Computing module 1103 carries out at identification for the data to be identified to be input to the data identification industry pattern
Reason, obtains recognition result;
Feedback module 1104, for the recognition result to be sent to the user.
Optionally, the determining module 1102 is specifically used for:
Obtain the right to use information of preset model corresponding with the attribute information;
According to the right to use information, determine that the user completes that the transaction of servitude of the preset model will be completed
The preset model of the transaction of servitude identifies industry pattern as the data.
Optionally, the data processing equipment further include: training module 1105, the training module 1105 are used for:
Training data is obtained from sample database;
Based on the training data and multiple preset models, the initial parameter of each preset model is configured;
The training data is configured in each preset model, each preset model is trained, is obtained each
Complete preset model, the corresponding training result of each preset model and the model accuracy rate of training;
Judge whether the training result meets preset condition, if not satisfied, then adjusting according to the training result each
The parameter of the preset model carries out retraining to the preset model after adjusting parameter, until the training result is full
The preset condition of foot;
The data are determined from each preset model for completing training based on the corresponding model accuracy rate of each preset model
Identify industry pattern.
Optionally, the data processing equipment further include: environment configurations module 1106, the environment configurations module 1106 have
Body is used for:
According to data identification request using the first number of users of data identification industry pattern, work is identified for the data
Industry model configures the first running environment;
According to data identification request using the second user number of data identification industry pattern, the data are identified into work
Industry model virtual turns to the several industry pattern copies of the second user;
It is the user configuration second according to the corresponding user information of the second user number and the first running environment
Running environment;
The data identification industry pattern copy of the user configuration is operated in second running environment.
Optionally, the data processing equipment further include: model construction module 1107, the model construction module 1107 have
Body is used for:
Polyalgorithm is edited in algorithm editing machine, the multiple algorithm is stored in algorithm list;
Selection algorithm handles preset training data from the algorithm list, obtains every in the algorithm list
The corresponding algorithm file of a algorithm;
Multiple algorithm files are compared, a kind of algorithm is selected from the algorithm list according to comparing result,
The preset model is generated according to the algorithm of selection.Optionally, the data processing equipment further include: data preparation module
1008, the data preparation module 1008 is specifically used for:
Obtain the operation data of equipment;
According to preset mapping relations, the operation data is mapped as metadata;
The metadata is formed into the sample database.
Optionally, the data processing equipment further include: industrial metadatabase constructs module 1109, the industry metadata
Library building module 1109 is specifically used for:
Obtain the initial data of data source;
The source address that the initial data carries is parsed, the detailed data information of the initial data is obtained;
Data source belonging to the initial data and data structure are determined, according to the data source and data determined
The data characteristics of structure mapping extracts the field to match with the data characteristics of the mapping from the detailed data information,
Obtain the data characteristics of the initial data;
According to pre-set target metadata mapping table, the data characteristics of the initial data is extracted, is obtained
Target data feature to be stored;
By the target data characteristic storage to be stored in industrial metadatabase.
Optionally, the data processing equipment further include: generation module 1110, the generation module 1110 are specifically used for:
Search instruction is obtained, the search instruction includes the keyword message of model;
According to the keyword message of the search instruction, the model information to match with the keyword message is extracted;
The model information is arranged according to pre-set ordering rule, the model information is loaded into advance
In the visualization interface of setting, the model of the load model information after showing visualization;
The data information that will acquire is input in the model and runs, and generates the operation result of the model;
The data information and the operation result are loaded into pre-set model running report template, mould is generated
Type operation report.
Optionally, the data processing equipment further include: initial parameter configuration module 1111, the initial parameter configure mould
Block 1111 is specifically used for:
Visual configuration data source generates metadata table according to the corresponding metadata mapping relations of the data source, according to
The data of target side are combined by the metadata table;
According to the history parameters in industrial knowledge base, the initial parameter of each preset model is configured.
Corresponding to the data processing method in Fig. 1, the embodiment of the present application also provides a kind of computer equipments 1200, such as scheme
Shown in 12, which includes memory 1201, processor 1202 and is stored on the memory 1201 and can be in the processor
The computer program run on 1202, wherein above-mentioned processor 1202 is realized at above-mentioned data when executing above-mentioned computer program
The step of reason method.
Specifically, above-mentioned memory 1201 and processor 1202 can be general memory and processor, not do here
It is specific to limit, when the computer program of 1202 run memory 1201 of processor storage, it is able to carry out above-mentioned data processing side
Method, for solving the problems, such as that data-handling efficiency is low in the prior art, by being carried in the data identification request of user wait know
Data needed for other data determine user identify industry pattern, and data to be identified are input in data identification industry pattern and are obtained
To recognition result, which is fed back into user, user can be obtained by required as a result, the data that the application passes through user
Identification request can find corresponding data identification industry pattern, shorten user query data corresponding with data to be identified
The duration for identifying industry pattern, improves the working efficiency of user, identifies industrial mould by the data that data identification request obtains
Type is for the data to be identified in data identification request, more targetedly, so that the data identify industry pattern pair
When data to be identified are handled, obtained result is more acurrate.
Corresponding to the data processing method in Fig. 1, the embodiment of the present application also provides a kind of computer readable storage medium,
It is stored with computer program on the computer readable storage medium, which executes above-mentioned data when being run by processor
The step of processing method.
Specifically, which can be general storage medium, such as mobile disk, hard disk, on the storage medium
Computer program when being run, above-mentioned model selection method is able to carry out, for solving data-handling efficiency in the prior art
Low problem, data needed for determining user by the data to be identified carried in the data identification request of user identify industrial mould
Type, and data to be identified are input in data identification industry pattern and obtain recognition result, which is fed back into user, user
It can be obtained by required as a result, the application, which can find corresponding data by the data identification request of user, identifies work
Industry model shortens the duration of user query data identification industry pattern corresponding with data to be identified, improves the work of user
Make efficiency, identifies industry pattern relative to the data to be identified in data identification request by the data that data identification request obtains
For, more targetedly, when so that data identification industry pattern handling data to be identified, obtained result is more quasi-
Really.
In embodiment provided herein, it should be understood that disclosed device and method, it can be by others side
Formula is realized.The apparatus embodiments described above are merely exemplary, for example, the division of the unit, only one kind are patrolled
Function division is collected, there may be another division manner in actual implementation, in another example, multiple units or components can combine or can
To be integrated into another system, or some features can be ignored or not executed.Another point, shown or discussed is mutual
Coupling, direct-coupling or communication connection can be INDIRECT COUPLING or communication link by some communication interfaces, device or unit
It connects, can be electrical property, mechanical or other forms.
The unit as illustrated by the separation member may or may not be physically separated, aobvious as unit
The component shown may or may not be physical unit, it can and it is in one place, or may be distributed over multiple
In network unit.It can select some or all of unit therein according to the actual needs to realize the mesh of this embodiment scheme
's.
In addition, each functional unit in embodiment provided by the present application can integrate in one processing unit, it can also
To be that each unit physically exists alone, can also be integrated in one unit with two or more units.
It, can be with if the function is realized in the form of SFU software functional unit and when sold or used as an independent product
It is stored in a computer readable storage medium.Based on this understanding, the technical solution of the application is substantially in other words
The part of the part that contributes to existing technology or the technical solution can be embodied in the form of software products, the meter
Calculation machine software product is stored in a storage medium, including some instructions are used so that a computer equipment (can be a
People's computer, server or network equipment etc.) execute each embodiment the method for the application all or part of the steps.
And storage medium above-mentioned includes: that USB flash disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), arbitrary access are deposited
The various media that can store program code such as reservoir (RAM, Random Access Memory), magnetic or disk.
It should also be noted that similar label and letter indicate similar terms in following attached drawing, therefore, once a certain Xiang Yi
It is defined in a attached drawing, does not then need that it is further defined and explained in subsequent attached drawing, in addition, term " the
One ", " second ", " third " etc. are only used for distinguishing description, are not understood to indicate or imply relative importance.
Finally, it should be noted that embodiment described above, the only specific embodiment of the application, to illustrate the application
Technical solution, rather than its limitations, the protection scope of the application is not limited thereto, although with reference to the foregoing embodiments to this Shen
It please be described in detail, those skilled in the art should understand that: anyone skilled in the art
Within the technical scope of the present application, it can still modify to technical solution documented by previous embodiment or can be light
It is readily conceivable that variation or equivalent replacement of some of the technical features;And these modifications, variation or replacement, do not make
The essence of corresponding technical solution is detached from the spirit and scope of the embodiment of the present application technical solution.The protection in the application should all be covered
Within the scope of.Therefore, the protection scope of the application shall be subject to the protection scope of the claim.
Claims (10)
1. a kind of data processing method characterized by comprising
The data for obtaining user identify request, carry data to be identified in the data identification request;
Based on the attribute information of the data to be identified, determine that data identify industry pattern for the user;
The data to be identified are input to the data identification industry pattern and carry out identifying processing, obtain recognition result;
The recognition result is sent to the user.
2. the method as described in claim 1, which is characterized in that the attribute information based on the data to be identified, for institute
It states user and determines that data identify industry pattern, comprising:
Obtain the right to use information of preset model corresponding with the attribute information;
According to the right to use information, determine that the user completes the transaction of servitude to the preset model, it will be described in completion
The preset model of transaction of servitude identifies industry pattern as the data.
3. the method as described in claim 1, which is characterized in that identify industry pattern according to the following manner training data:
Training data is obtained from sample database;
Based on the training data and multiple preset models, the initial parameter of each preset model is configured;
The training data is configured in each preset model, each preset model is trained, each completion is obtained
Trained preset model, the corresponding training result of each preset model and model accuracy rate;
Judge whether the training result meets preset condition, if not satisfied, then each described according to training result adjustment
The parameter of preset model carries out retraining to the preset model after adjusting parameter, until the training result meets in advance
If condition;
Based on the corresponding model accuracy rate of each preset model, from each preset model for completing training, the data identification is determined
Industry pattern.
4. the method as described in claim 1, which is characterized in that it is described for the user determine data identify industry pattern it
Afterwards, further includes:
According to data identification request using the first number of users of data identification industry pattern, industrial mould is identified for the data
Type configures the first running environment;
According to data identification request using the second user number of data identification industry pattern, the data are identified into industrial mould
Type virtually turns to the several industry pattern copies of the second user;
According to the corresponding user information of the second user number and the first running environment, run for the user configuration second
Environment;
The data identification industry pattern copy of the user configuration is operated in second running environment.
5. method as claimed in claim 3, which is characterized in that construct preset model according to following manner:
Polyalgorithm is edited in algorithm editing machine, the multiple algorithm is stored in algorithm list;
Selection algorithm handles preset training data from the algorithm list, obtains each calculation in the algorithm list
The corresponding algorithm file of method;
Multiple algorithm files are compared, a kind of algorithm is selected from the algorithm list according to comparing result, according to
The algorithm of selection generates the preset model.
6. method as claimed in claim 3, which is characterized in that construct the sample database according to following manner, comprising:
Obtain the operation data of equipment;
According to preset mapping relations, the operation data is mapped as metadata;
The metadata is formed into the sample database.
7. the method as described in claim 1, which is characterized in that further include:
Obtain the initial data of data source;
The source address that the initial data carries is parsed, the detailed data information of the initial data is obtained;
Data source belonging to the initial data and data structure are determined, according to the data source and data structure determined
The data characteristics of mapping is extracted the field to match with the data characteristics of the mapping from the detailed data information, is obtained
The data characteristics of the initial data;
According to pre-set target metadata mapping table, the data characteristics of the initial data is extracted, is obtained wait deposit
Store up target data feature;
By the target data characteristic storage to be stored in industrial metadatabase.
8. method as claimed in claim 3, which is characterized in that after the preset model for obtaining each completion training, also
Include:
Search instruction is obtained, the search instruction includes the keyword message of model;
According to the keyword message of the search instruction, the model information to match with the keyword message is extracted;
The model information is arranged according to pre-set ordering rule, the model information is loaded into and is preset
Visualization interface in, the model of the load model information after showing visualization;
The data information that will acquire is input in the model and runs, and generates the operation result of the model;
The data information and the operation result are loaded into pre-set model running report template, model fortune is generated
Row report.
9. method as claimed in claim 3, which is characterized in that the initial parameter of each preset model of configuration, comprising:
Visual configuration data source generates metadata table, according to described according to the corresponding metadata mapping relations of the data source
The data of target side are combined by metadata table;
According to the history parameters in industrial knowledge base, the initial parameter of each preset model is configured.
10. a kind of data processing equipment characterized by comprising
Module is obtained, the data for obtaining user identify request, carry data to be identified in the data identification request;
Determining module determines that data identify industry pattern for the attribute information based on the data to be identified for the user;
Computing module carries out identifying processing for the data to be identified to be input to the data identification industry pattern, obtains
Recognition result;
Feedback module, for the recognition result to be sent to the user.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811643437.1A CN109670267B (en) | 2018-12-29 | 2018-12-29 | Data processing method and device |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811643437.1A CN109670267B (en) | 2018-12-29 | 2018-12-29 | Data processing method and device |
Publications (2)
Publication Number | Publication Date |
---|---|
CN109670267A true CN109670267A (en) | 2019-04-23 |
CN109670267B CN109670267B (en) | 2023-06-13 |
Family
ID=66147474
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201811643437.1A Active CN109670267B (en) | 2018-12-29 | 2018-12-29 | Data processing method and device |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109670267B (en) |
Cited By (15)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109741141A (en) * | 2018-12-30 | 2019-05-10 | 北京航天数据股份有限公司 | Configure the method and industry pattern running environment configurator of industry pattern running environment |
CN110334955A (en) * | 2019-07-08 | 2019-10-15 | 北京字节跳动网络技术有限公司 | Processing method, device, equipment and the storage medium of index evaluation |
CN110347845A (en) * | 2019-07-15 | 2019-10-18 | 北京明略软件系统有限公司 | The method for drafting and device of knowledge mapping |
CN110705312A (en) * | 2019-09-30 | 2020-01-17 | 贵州航天云网科技有限公司 | Development system for rapidly developing industrial mechanism model based on semantic analysis |
CN110909065A (en) * | 2019-12-06 | 2020-03-24 | 光大兴陇信托有限责任公司 | Multi-data-source intelligent calling method |
CN111338988A (en) * | 2020-02-20 | 2020-06-26 | 西安芯瞳半导体技术有限公司 | Memory access method and device, computer equipment and storage medium |
CN111813846A (en) * | 2020-06-29 | 2020-10-23 | 北京九章云极科技有限公司 | Data analysis processing system and data processing method |
CN111861648A (en) * | 2020-07-06 | 2020-10-30 | 南京大学 | Price negotiation strategy model learning method based on simulation training |
CN112035414A (en) * | 2020-08-31 | 2020-12-04 | 山东浪潮通软信息科技有限公司 | Metadata streaming method, device and computer readable medium |
CN112204547A (en) * | 2020-05-26 | 2021-01-08 | 深圳市智物联网络有限公司 | Data processing method, device and equipment based on industrial object model |
CN112906717A (en) * | 2021-03-04 | 2021-06-04 | 广联达科技股份有限公司 | Method and system for identifying engineering file content |
CN113284141A (en) * | 2021-07-09 | 2021-08-20 | 武汉精创电子技术有限公司 | Model determination method, device and equipment for defect detection |
CN113377785A (en) * | 2021-08-16 | 2021-09-10 | 宜科(天津)电子有限公司 | Industrial data processing system |
CN113890899A (en) * | 2021-09-13 | 2022-01-04 | 北京交通大学 | Protocol conversion method based on knowledge graph |
CN116738081A (en) * | 2023-08-08 | 2023-09-12 | 贵州优特云科技有限公司 | Front-end component binding method, device and storage medium |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2011092793A1 (en) * | 2010-01-29 | 2011-08-04 | パナソニック株式会社 | Data processing device |
CN106951407A (en) * | 2017-03-15 | 2017-07-14 | 南京理工大学 | Quick high accuracy semantic relevancy computational methods based on Spark platforms |
CN107391569A (en) * | 2017-06-16 | 2017-11-24 | 阿里巴巴集团控股有限公司 | Identification, model training, Risk Identification Method, device and the equipment of data type |
CN108520324A (en) * | 2018-04-13 | 2018-09-11 | 北京京东金融科技控股有限公司 | Method and apparatus for generating information |
CN108733778A (en) * | 2018-05-04 | 2018-11-02 | 百度在线网络技术(北京)有限公司 | The industry type recognition methods of object and device |
CN108984500A (en) * | 2018-06-19 | 2018-12-11 | 平安科技(深圳)有限公司 | Extracting method, terminal device and the medium of amount information |
-
2018
- 2018-12-29 CN CN201811643437.1A patent/CN109670267B/en active Active
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2011092793A1 (en) * | 2010-01-29 | 2011-08-04 | パナソニック株式会社 | Data processing device |
CN106951407A (en) * | 2017-03-15 | 2017-07-14 | 南京理工大学 | Quick high accuracy semantic relevancy computational methods based on Spark platforms |
CN107391569A (en) * | 2017-06-16 | 2017-11-24 | 阿里巴巴集团控股有限公司 | Identification, model training, Risk Identification Method, device and the equipment of data type |
CN108520324A (en) * | 2018-04-13 | 2018-09-11 | 北京京东金融科技控股有限公司 | Method and apparatus for generating information |
CN108733778A (en) * | 2018-05-04 | 2018-11-02 | 百度在线网络技术(北京)有限公司 | The industry type recognition methods of object and device |
CN108984500A (en) * | 2018-06-19 | 2018-12-11 | 平安科技(深圳)有限公司 | Extracting method, terminal device and the medium of amount information |
Cited By (21)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109741141A (en) * | 2018-12-30 | 2019-05-10 | 北京航天数据股份有限公司 | Configure the method and industry pattern running environment configurator of industry pattern running environment |
CN110334955A (en) * | 2019-07-08 | 2019-10-15 | 北京字节跳动网络技术有限公司 | Processing method, device, equipment and the storage medium of index evaluation |
CN110334955B (en) * | 2019-07-08 | 2021-09-14 | 北京字节跳动网络技术有限公司 | Index evaluation processing method, device, equipment and storage medium |
CN110347845A (en) * | 2019-07-15 | 2019-10-18 | 北京明略软件系统有限公司 | The method for drafting and device of knowledge mapping |
CN110705312A (en) * | 2019-09-30 | 2020-01-17 | 贵州航天云网科技有限公司 | Development system for rapidly developing industrial mechanism model based on semantic analysis |
CN110909065B (en) * | 2019-12-06 | 2020-10-27 | 光大兴陇信托有限责任公司 | Multi-data-source intelligent calling method |
CN110909065A (en) * | 2019-12-06 | 2020-03-24 | 光大兴陇信托有限责任公司 | Multi-data-source intelligent calling method |
CN111338988A (en) * | 2020-02-20 | 2020-06-26 | 西安芯瞳半导体技术有限公司 | Memory access method and device, computer equipment and storage medium |
CN112204547A (en) * | 2020-05-26 | 2021-01-08 | 深圳市智物联网络有限公司 | Data processing method, device and equipment based on industrial object model |
CN111813846B (en) * | 2020-06-29 | 2021-04-02 | 北京九章云极科技有限公司 | Data analysis processing system and data processing method |
CN111813846A (en) * | 2020-06-29 | 2020-10-23 | 北京九章云极科技有限公司 | Data analysis processing system and data processing method |
CN111861648A (en) * | 2020-07-06 | 2020-10-30 | 南京大学 | Price negotiation strategy model learning method based on simulation training |
CN112035414B (en) * | 2020-08-31 | 2024-05-03 | 浪潮通用软件有限公司 | Metadata streaming method, apparatus and computer readable medium |
CN112035414A (en) * | 2020-08-31 | 2020-12-04 | 山东浪潮通软信息科技有限公司 | Metadata streaming method, device and computer readable medium |
CN112906717A (en) * | 2021-03-04 | 2021-06-04 | 广联达科技股份有限公司 | Method and system for identifying engineering file content |
CN112906717B (en) * | 2021-03-04 | 2024-05-28 | 广联达科技股份有限公司 | Method and system for identifying engineering file content |
CN113284141A (en) * | 2021-07-09 | 2021-08-20 | 武汉精创电子技术有限公司 | Model determination method, device and equipment for defect detection |
CN113377785A (en) * | 2021-08-16 | 2021-09-10 | 宜科(天津)电子有限公司 | Industrial data processing system |
CN113890899A (en) * | 2021-09-13 | 2022-01-04 | 北京交通大学 | Protocol conversion method based on knowledge graph |
CN116738081B (en) * | 2023-08-08 | 2023-10-27 | 贵州优特云科技有限公司 | Front-end component binding method, device and storage medium |
CN116738081A (en) * | 2023-08-08 | 2023-09-12 | 贵州优特云科技有限公司 | Front-end component binding method, device and storage medium |
Also Published As
Publication number | Publication date |
---|---|
CN109670267B (en) | 2023-06-13 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109670267A (en) | A kind of data processing method and device | |
CN109902708A (en) | A kind of recommended models training method and relevant apparatus | |
Khatibi Bardsiri et al. | A PSO-based model to increase the accuracy of software development effort estimation | |
CN110309119B (en) | System, method and apparatus for implementing data upload, processing and predictive query API disclosure | |
US20200334293A1 (en) | Computation platform agnostic data classification workflows | |
US10459979B2 (en) | Graphically managing data classification workflows in a social networking system with directed graphs | |
CN102708130B (en) | Calculate the easily extensible engine that fine point of user is mated for offer | |
CN108701258A (en) | For by counting the system and method for dissecting and carrying out ontology conclusion with reference model matching | |
US11704340B2 (en) | Linking discrete dimensions to enhance dimensional analysis | |
US20180004835A1 (en) | Data classification workflows implemented with dynamically modifiable directed graphs | |
CN106663038A (en) | Feature processing recipes for machine learning | |
US20080288266A1 (en) | System and method for service offering for feedback controller design and implementation for performance management in information technology systems | |
CN103646070A (en) | Data processing method and device for search engine | |
CN112559896B (en) | Information recommendation method, device, equipment and computer readable storage medium | |
CN110249356A (en) | A kind of sharing method and system of user-defined ERP function | |
Feng et al. | [Retracted] Design and Simulation of Human Resource Allocation Model Based on Double‐Cycle Neural Network | |
US20220058499A1 (en) | Multidimensional hierarchy level recommendation for forecasting models | |
CN110245820A (en) | Conversion platform | |
CN105940636A (en) | Technologies for cloud data center analytics | |
KR20200021843A (en) | Method and system for printing service | |
CN105138610A (en) | Image element based image feature value prediction method and apparatus | |
CN106575296A (en) | Dynamic N-dimensional cubes for hosted analytics | |
CN109242141A (en) | A kind of prediction technique and device of commodity stocks quantity | |
CN110262973B (en) | Data maintenance method, device, equipment and computer storage medium | |
JP2022191637A (en) | AI platform system |
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 | ||
GR01 | Patent grant | ||
GR01 | Patent grant |