CN113239034A - Big data resource integration method and system based on artificial intelligence and cloud platform - Google Patents
Big data resource integration method and system based on artificial intelligence and cloud platform Download PDFInfo
- Publication number
- CN113239034A CN113239034A CN202110666255.1A CN202110666255A CN113239034A CN 113239034 A CN113239034 A CN 113239034A CN 202110666255 A CN202110666255 A CN 202110666255A CN 113239034 A CN113239034 A CN 113239034A
- Authority
- CN
- China
- Prior art keywords
- data
- track
- integration
- data integration
- list
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Withdrawn
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/21—Design, administration or maintenance of databases
- G06F16/215—Improving data quality; Data cleansing, e.g. de-duplication, removing invalid entries or correcting typographical errors
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- Biophysics (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Biomedical Technology (AREA)
- Computational Linguistics (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Life Sciences & Earth Sciences (AREA)
- Quality & Reliability (AREA)
- Health & Medical Sciences (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention relates to the technical field of big data and artificial intelligence, in particular to a big data resource integration method and system based on artificial intelligence and a cloud platform. The method comprises the steps of obtaining a track user data set of a first track and customer description data on the first track, inputting the track user data set of the first track and the customer description data on the first track into a first target user data set neural model to determine a preset resource data integration list from a first initial calculation point to a first termination calculation point in a data integration model, obtaining the track user data set of the first track and the customer description data of the first track from the data integration model, inputting the neural model to further determine the preset resource data integration list, achieving the purpose of improving the accuracy of determining the preset resource data integration list, improving the integration accuracy of the preset resource data integration list, and providing the technical effect of optimal track selection for a user.
Description
Technical Field
The disclosure relates to the technical field of big data and artificial intelligence, in particular to a big data resource integration method and system based on artificial intelligence and a cloud platform.
Background
Cloud platforms used by current related data designers need to count a wide variety of data resources, for example, when creating a piece of related data, a large number of related data elements are required to be combined. However, some flaws exist when some of the related data elements are integrated, but the completed related data elements cannot be managed by the cloud platform corresponding to the current related data, so that related data resources are wasted.
Disclosure of Invention
In order to solve the technical problems existing in the background technology in the related art, the disclosure provides a big data resource integration method and system based on artificial intelligence and a cloud platform.
The application provides a big data resource integration method based on artificial intelligence, which comprises the following steps:
acquiring a track user data set of a first track and customer description class data on the first track, wherein the first track has a one-to-one correspondence relationship with a first target user data set neural model, the first track is a track in a data integration model, and the track user data set of the first track is used for representing an association relationship between a first group of data integration classifications included in the first track and a first data integration classification in the first group of data integration classifications;
inputting the track user data set of the first track and the customer description class data on the first track into the first target user data set neural model to obtain a first resource data integration list output by the first target user data set neural model;
and determining a preset resource data integration list from a first initial calculation point to a first termination calculation point in the data integration model according to the first resource data integration list, wherein the first track is passed from the first initial calculation point to the first termination calculation point.
Further, the inputting the track user data set of the first track and the customer description class data on the first track into the first target user data set neural model to obtain a first resource data integration list output by the first target user data set neural model includes:
according to an input track user data set of the first track and customer description class data on the first track, determining a description strategy of a first data type included in the track user data set and a description strategy of first key content included in the track user data set, wherein the first data type and a first data integration class have a one-to-one correspondence relationship, the first data type represents a corresponding data integration class in the first group of data integration classes, the first key content connects two first data types and represents that the two first data integration classes corresponding to the two first data types are matched, and the first data integration class is a data integration class in the first group of data integration classes and is used for representing a local track in the first track;
determining a reference coefficient of the first data type according to the description strategy of the first data type and the description strategy of the first key content;
and determining the first resource data integration list according to the reference coefficient of the first data type.
Further, the determining the reference coefficient of the first data type according to the description policy of the first data type and the description policy of the first key content includes:
determining a reference coefficient of a current data type on a next trade line according to a description policy of the current data type, a description policy of key content associated with the current data type, a description policy of an adapted data type of the current data type, and a reference coefficient of the adapted data type of the current data type on the current trade line, wherein the first data type comprises the current data type and the adapted data type of the current data type.
Further, the determining a preset resource data integration list from a first initial computation point to a first termination computation point in the data integration model according to the first resource data integration list comprises:
determining a preset resource data integration list from the initial calculation point of the first track to the termination calculation point of the first track as the first resource data integration list when the first initial calculation point is the initial calculation point of the first track and the first termination calculation point is the termination calculation point of the first track;
and/or determining a preset resource data integration list from the initial calculation point of the first data integration classification to the termination calculation point of the first data integration classification as a sample list of the first data integration classification if the first initial calculation point is the initial calculation point of the first data integration classification and the first termination calculation point is the termination calculation point of the first data integration classification;
and/or in the case that the first initial calculation point is an initial calculation point of the plurality of data integration classifications and the first termination calculation point is a termination calculation point of the plurality of data integration classifications, determining a preset resource data integration list from the initial calculation point of the plurality of data integration classifications to the termination calculation point of the plurality of data integration classifications as an intersection of the sample lists of the plurality of data integration classifications.
Further, the method further comprises:
training a first training data neural model according to a sample track user data set of a sample track, sample client description class data on the sample track and a real-time sample list of the sample track to obtain a first target user data set neural model, wherein under the condition that a value of an error range between the sample list output by the first training data neural model and the real-time sample list meets a preset condition, the training of the first training data neural model is finished, and the first training data neural model when the training is finished is determined as the first target user data set neural model.
Further, the method further comprises:
and comparing the first resource data integration list with the real-time sample list of the first track to update the first target user data set neural model, wherein the first target user data set neural model is updated under the condition that the value of the error range between the first resource data integration list and the real-time sample list meets a preset condition.
Further, the method further comprises:
acquiring a track user data set of a second track and customer description class data on the second track, wherein the second track has a one-to-one correspondence relationship with a second target user data set neural model, and the track user data set of the second track is used for representing an association relationship between a second group of data integration classification and a data integration classification in the second group of data integration classification included in the second track;
inputting the track user data set of the second track and the customer description class data on the second track into the second target user data set neural model to obtain a second resource data integration list output by the second target user data set neural model;
determining a preset resource data integration list from a first initial computation point to a first termination computation point in the data integration model according to the first resource data integration list, including: and determining a preset resource data integration list from a first initial calculation point to a first termination calculation point in the data integration model according to the first resource data integration list and the second resource data integration list, wherein the first track and the second track are passed from the first initial calculation point to the first termination calculation point.
The application provides big data resource integration system based on artificial intelligence, including data collection equipment and cloud platform, data collection equipment with cloud platform communication connection, the cloud platform includes:
the data description acquisition module is used for acquiring a track user data set of a first track and customer description class data on the first track, wherein the first track and a first target user data set neural model have a one-to-one correspondence relationship, the first track is a track in a data integration model, and the track user data set of the first track is used for representing an association relationship between a first group of data integration classifications included in the first track and a first data integration classification in the first group of data integration classifications;
a data integration determining module, configured to input the track user data set of the first track and the customer description class data on the first track into the first target user data set neural model, so as to obtain a first resource data integration list output by the first target user data set neural model;
and the integrated data calculation module is used for determining a preset resource data integration list from a first initial calculation point to a first termination calculation point in the data integration model according to the first resource data integration list, wherein the first track is passed from the first initial calculation point to the first termination calculation point.
The application provides a cloud platform, which comprises a processor and a memory which are communicated with each other, wherein the processor is used for calling a computer program from the memory and realizing the method of any one of the above items by running the computer program.
The present application provides a computer-readable storage medium having stored thereon a computer program which, when executed, implements the method of any of the above.
The technical scheme provided by the embodiment of the disclosure can have the following beneficial effects.
A big data resource integration method, a system and a cloud platform based on artificial intelligence are disclosed, a track user data set of a first track and customer description class data on the first track are obtained, the track user data set of the first track and the customer description class data on the first track are input into a first target user data set neural model to obtain a first resource data integration list output by the first target user data set neural model, a preset resource data integration list from a first initial calculation point to a first termination calculation point in the data integration model is determined according to the first resource data integration list, the track user data set of the first track and the customer description class data of the first track are obtained from the data integration model to input the neural model, and then the preset resource data integration list is determined, so that the purpose of improving the accuracy of determining the preset resource data integration list is achieved, therefore, the technical effects of optimizing the rate of resource integration, improving the integration accuracy of the preset resource data integration list and providing the optimal track selection for the user are achieved.
It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the disclosure.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and together with the description, serve to explain the principles of the application.
FIG. 1 is a schematic diagram of an architecture of a big data resource integration system based on artificial intelligence according to an embodiment of the present application;
FIG. 2 is a flowchart of a big data resource integration method based on artificial intelligence according to an embodiment of the present disclosure;
fig. 3 is a functional block diagram of a big data resource integration apparatus based on artificial intelligence according to an embodiment of the present disclosure.
Detailed Description
Reference will now be made in detail to the exemplary embodiments, examples of which are illustrated in the accompanying drawings. When the following description refers to the accompanying drawings, like numbers in different drawings represent the same or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present application. Rather, they are merely examples of apparatus and methods consistent with certain aspects of the present application, as detailed in the appended claims.
In order to facilitate the description of the above method, system and cloud platform for integrating big data resources based on artificial intelligence, please refer to fig. 1, which provides a schematic diagram of a communication architecture of a system 100 for integrating big data resources based on artificial intelligence disclosed in the embodiments of the present application. The artificial intelligence based big data resource integration system 100 can comprise a data collection device 200 and a cloud platform 300, wherein the data collection device 200 is connected with the cloud platform 300 in a communication mode.
In a specific embodiment, the cloud platform 300 may be a desktop computer, a tablet computer, a notebook computer, a mobile phone, or another cloud platform capable of implementing data processing and data communication, which is not limited herein.
On the basis, please refer to fig. 2 in combination, which is a schematic flow chart of an artificial intelligence based big data resource integration method according to an embodiment of the present application, where the artificial intelligence based big data resource integration method may be applied to the cloud platform 300 in fig. 1, and further, the artificial intelligence based big data resource integration method may specifically include the contents described in the following steps S21 to S23.
Step S21, a track user data set of a first track and customer description class data on the first track are obtained.
Illustratively, the first trajectory has a one-to-one correspondence relationship with a neural model of a first target user data set, the first trajectory is a trajectory in a data integration model, and a trajectory user data set of the first trajectory is used for representing an association relationship between a first set of data integration classifications included in the first trajectory and a first data integration classification in the first set of data integration classifications.
Step S22, inputting the trajectory user data set of the first trajectory and the customer description class data on the first trajectory into the first target user data set neural model, and obtaining a first resource data integration list output by the first target user data set neural model.
Illustratively, the first resource data integration list represents a distribution set formed by classifying the related data into categories.
Step S23, determining a preset resource data integration list from a first initial computation point to a first termination computation point in the data integration model according to the first resource data integration list.
Illustratively, the first trajectory is traversed from the first initial computation point to the first termination computation point.
It can be understood that, when the contents described in the above steps S21-S23 are executed, the track user data set of the first track and the customer description class data on the first track are obtained, the track user data set of the first track and the customer description class data on the first track are input into the first target user data set neural model, the first resource data integration list output by the first target user data set neural model is obtained, according to the first resource data integration list, the preset resource data integration list from the first initial computation point to the first termination computation point in the data integration model is determined, the track user data set of the first track and the customer description class data of the first track are obtained from the data integration model to input the neural model, and then the preset resource data integration list is determined, so as to achieve the purpose of improving the accuracy of determining the preset resource data integration list, therefore, the technical effects of optimizing the rate of resource integration, improving the integration accuracy of the preset resource data integration list and providing the optimal track selection for the user are achieved.
In another alternative embodiment, the input of the track user data set of the first track and the customer description class data on the first track into the first target user data set neural model has a problem of related model calculation error, so that it is difficult to accurately obtain the first resource data integration list output by the first target user data set neural model, and in order to improve the above technical problem, the step of inputting the track user data set of the first track and the customer description class data on the first track into the first target user data set neural model to obtain the first resource data integration list output by the first target user data set neural model, which is described in step S22, may specifically include the following steps S221 to S223.
Step S221, determining a description policy of a first data type included in the track user data set and a description policy of first key content included in the track user data set according to the input track user data set of the first track and the client description class data on the first track.
Illustratively, the first data type has a one-to-one correspondence relationship with a first data integration classification, the first data type represents a corresponding one of the first group of data integration classifications, the first key content connects two of the first data types, and represents that the two first data integration classifications corresponding to the two first data types match, and the first data integration classification is a data integration classification in the first group of data integration classifications and is used for representing a local trajectory in the first trajectory.
Step S222, determining a reference coefficient of the first data type according to the description policy of the first data type and the description policy of the first key content.
Step S223, determining the first resource data integration list according to the reference coefficient of the first data type.
It can be understood that, when the contents described in steps S221 to S223 are executed, the track user data set of the first track and the client description class data on the first track are input to the first target user data set neural model, so as to avoid the problem of calculation error of the correlation model, and thus the first resource data integration list output by the first target user data set neural model can be accurately obtained.
In another alternative embodiment, there is a problem that the current data type is inaccurate according to the description policy of the first data type and the description policy of the first key content, so that it is difficult to accurately determine the reference coefficient of the first data type, and in order to improve the above technical problem, the step of determining the reference coefficient of the first data type according to the description policy of the first data type and the description policy of the first key content described in step S222 may specifically include the following step q 1.
Step q1, determining a reference coefficient of the current data type on a next trade boundary according to a description policy of the current data type, a description policy of key content associated with the current data type, a description policy of an adapted data type of the current data type, and a reference coefficient of the adapted data type of the current data type on a current trade boundary, wherein the first data type includes the current data type and the adapted data type of the current data type.
It can be understood that, when the content described in the above step q1 is executed, according to the description policy of the first data type and the description policy of the first key content, the problem of inaccurate current data type is avoided, so that the reference coefficient of the first data type can be accurately determined.
In a specific implementation process, the inventors found that, according to the first resource data integration list, there is a problem that a starting point and an ending point of the first resource data are inaccurate, so that it is difficult to accurately determine a preset resource data integration list from a first initial calculation point to a first ending calculation point in the data integration model, and in order to improve the above technical problem, the step of determining the preset resource data integration list from the first initial calculation point to the first ending calculation point in the data integration model according to the first resource data integration list described in step S23 may specifically include the following steps S231-S233.
Step S231, if the first initial computing point is an initial computing point of the first track and the first termination computing point is a termination computing point of the first track, determining a preset resource data integration list from the initial computing point of the first track to the termination computing point of the first track as the first resource data integration list.
Step S232, and/or in a case that the first initial calculation point is an initial calculation point of the first data integration classification, and the first termination calculation point is a termination calculation point of the first data integration classification, determining a preset resource data integration list from the initial calculation point of the first data integration classification to the termination calculation point of the first data integration classification as a sample list of the first data integration classification.
Step S233, and/or in a case that the first initial calculation point is an initial calculation point of the multiple data consolidation classifications, and the first termination calculation point is a termination calculation point of the multiple data consolidation classifications, determining a preset resource data consolidation list from the initial calculation point of the multiple data consolidation classifications to the termination calculation point of the multiple data consolidation classifications as an intersection of sample lists of the multiple data consolidation classifications.
It can be understood that, when the contents described in steps S231 to S233 are executed, the problem that the starting point and the ending point of the first resource data are inaccurate is avoided according to the first resource data integration list, so that the preset resource data integration list from the first initial computing point to the first ending computing point in the data integration model can be accurately determined.
Based on the above basis, the following description of step q1 is also included.
Step q1, training a first training data neural model according to a sample track user data set of a sample track, sample client description class data on the sample track, and a real-time sample list of the sample track to obtain the first target user data set neural model, wherein when a value of an error range between the sample list output by the first training data neural model and the real-time sample list meets a preset condition, training of the first training data neural model is finished, and the first training data neural model when training is finished is determined as the first target user data set neural model.
It can be understood that, when the content described in the above step 1 is executed, the calculation accuracy of the neural model of the first target user data set is improved by improving the correctness of the real-time sample list.
Based on the above basis, the method also comprises the following steps w 1.
Step w1, comparing the first resource data integration list with the real-time sample list of the first trajectory to update the first target user data set neural model, wherein the first target user data set neural model is updated when a value of an error range between the first resource data integration list and the real-time sample list meets a preset condition.
It can be understood that when the content described in the above step w1 is executed, the neural model of the first target user data set can be accurately updated by improving the comparison accuracy between the first resource data integration list and the real-time sample list of the first track.
Based on the above basis, the method also comprises the following contents described in the steps e 1-e 2.
Step e1, a track user data set of a second track and customer description class data on the second track are obtained.
For example, the second trajectory has a one-to-one correspondence with a second target user data set neural model, and the trajectory user data set of the second trajectory is used to represent an association between a second set of data integration classifications included in the second trajectory and data integration classifications in the second set of data integration classifications.
And e2, inputting the track user data set of the second track and the customer description class data on the second track into the second target user data set neural model, and obtaining a second resource data integration list output by the second target user data set neural model.
Step e3, said determining a preset integration list of resource data from a first initial computation point to a first termination computation point in said data integration model according to said first integration list of resource data, comprising: and determining a preset resource data integration list from a first initial calculation point to a first termination calculation point in the data integration model according to the first resource data integration list and the second resource data integration list.
For example, the first trajectory and the second trajectory are traversed from the first initial computation point to the first termination computation point.
It can be understood that, when the contents described in the above-mentioned steps e 1-e 2 are executed, the accuracy of the track user data set of the second track and the customer description class data on the second track is improved, so that the preset resource data integration list from the first initial calculation point to the first termination calculation point in the data integration model can be accurately determined.
Based on the above basis, what is described in the following step t1 is also included.
Step t1, in a case that the preset resource data integration list from the first initial calculation point to the first termination calculation point satisfies a preset statistical condition, determining the first trajectory as a trajectory through which a statistically planned trajectory from the first initial calculation point to the first termination calculation point passes.
It can be understood that, when the content described in the above step t1 is executed, the track passed by the statistically planned track can be accurately determined in the case that the preset resource data integration list satisfies the preset statistical condition.
In an alternative embodiment, in the case that the preset integration list of resource data from the first initial computation point to the first termination computation point satisfies a preset statistical condition, there is a problem that the first trajectory does not satisfy the preset statistical condition, it is difficult to accurately determine the trajectory traversed by the statistically planned trajectory from the first initial calculation point to the first termination calculation point, in order to improve the above technical problem, in the case that the preset integration list of resource data from the first initial computing point to the first termination computing point satisfies the preset statistical condition as described in step t1, the step of determining the first trajectory as the trajectory traversed by the statistically planned trajectory from the first initial calculation point to the first termination calculation point may specifically include the following steps y1 and y 2.
Step y1, determining the first trajectory as the trajectory traversed by the statistically planned trajectory from the first initial computation point to the first termination computation point if the preset integration list of resource data from the first initial computation point to the first termination computation point is less than or equal to a preset list identification result.
Step y2, or in case that the preset integration list of resource data from the first initial computation point to the first termination computation point is the minimum value of a plurality of preset integration lists of resource data, determining the first trajectory as the trajectory through which a statistically planned trajectory from the first initial computation point to the first termination computation point passes, wherein the preset resource data integration lists and the planning tracks have mapping relations, each preset resource data integration list in the plurality of preset resource data integration lists is a preset resource data integration list of a corresponding one of the plurality of planned trajectories, the plurality of planned trajectories include the first trajectory, and the plurality of planned trajectories are trajectories respectively traversed by different planned trajectories from the first initial calculation point to the first termination calculation point.
It is understood that, when the contents described in the above steps y1 and y2 are executed, in the case that the preset resource data integration list from the first initial calculation point to the first termination calculation point satisfies the preset statistical condition, the problem that the first trajectory does not satisfy the preset statistical condition is avoided, so that the trajectory through which the statistically planned trajectory from the first initial calculation point to the first termination calculation point passes can be accurately determined.
Based on the same inventive concept, the system for integrating the big data resources based on artificial intelligence is further provided, and the system comprises data collection equipment and a cloud platform, wherein the data collection equipment is in communication connection with the cloud platform, and the cloud platform is specifically used for:
acquiring a track user data set of a first track and customer description class data on the first track, wherein the first track has a one-to-one correspondence relationship with a first target user data set neural model, the first track is a track in a data integration model, and the track user data set of the first track is used for representing an association relationship between a first group of data integration classifications included in the first track and a first data integration classification in the first group of data integration classifications;
inputting the track user data set of the first track and the customer description class data on the first track into the first target user data set neural model to obtain a first resource data integration list output by the first target user data set neural model;
and determining a preset resource data integration list from a first initial calculation point to a first termination calculation point in the data integration model according to the first resource data integration list, wherein the first track is passed from the first initial calculation point to the first termination calculation point.
Further, the cloud platform is specifically configured to:
according to an input track user data set of the first track and customer description class data on the first track, determining a description strategy of a first data type included in the track user data set and a description strategy of first key content included in the track user data set, wherein the first data type and a first data integration class have a one-to-one correspondence relationship, the first data type represents a corresponding data integration class in the first group of data integration classes, the first key content connects two first data types and represents that the two first data integration classes corresponding to the two first data types are matched, and the first data integration class is a data integration class in the first group of data integration classes and is used for representing a local track in the first track;
determining a reference coefficient of the first data type according to the description strategy of the first data type and the description strategy of the first key content;
and determining the first resource data integration list according to the reference coefficient of the first data type.
Further, the cloud platform is specifically configured to:
determining a reference coefficient of a current data type on a next trade line according to a description policy of the current data type, a description policy of key content associated with the current data type, a description policy of an adapted data type of the current data type, and a reference coefficient of the adapted data type of the current data type on the current trade line, wherein the first data type comprises the current data type and the adapted data type of the current data type.
Further, the cloud platform is specifically configured to:
determining a preset resource data integration list from the initial calculation point of the first track to the termination calculation point of the first track as the first resource data integration list when the first initial calculation point is the initial calculation point of the first track and the first termination calculation point is the termination calculation point of the first track;
and/or determining a preset resource data integration list from the initial calculation point of the first data integration classification to the termination calculation point of the first data integration classification as a sample list of the first data integration classification if the first initial calculation point is the initial calculation point of the first data integration classification and the first termination calculation point is the termination calculation point of the first data integration classification;
and/or in the case that the first initial calculation point is an initial calculation point of the plurality of data integration classifications and the first termination calculation point is a termination calculation point of the plurality of data integration classifications, determining a preset resource data integration list from the initial calculation point of the plurality of data integration classifications to the termination calculation point of the plurality of data integration classifications as an intersection of the sample lists of the plurality of data integration classifications.
Further, the cloud platform is specifically configured to:
training a first training data neural model according to a sample track user data set of a sample track, sample client description class data on the sample track and a real-time sample list of the sample track to obtain a first target user data set neural model, wherein under the condition that a value of an error range between the sample list output by the first training data neural model and the real-time sample list meets a preset condition, the training of the first training data neural model is finished, and the first training data neural model when the training is finished is determined as the first target user data set neural model.
Further, the cloud platform is specifically configured to:
and comparing the first resource data integration list with the real-time sample list of the first track to update the first target user data set neural model, wherein the first target user data set neural model is updated under the condition that the value of the error range between the first resource data integration list and the real-time sample list meets a preset condition.
Further, the cloud platform is specifically configured to:
acquiring a track user data set of a second track and customer description class data on the second track, wherein the second track has a one-to-one correspondence relationship with a second target user data set neural model, and the track user data set of the second track is used for representing an association relationship between a second group of data integration classification and a data integration classification in the second group of data integration classification included in the second track;
inputting the track user data set of the second track and the customer description class data on the second track into the second target user data set neural model to obtain a second resource data integration list output by the second target user data set neural model;
determining a preset resource data integration list from a first initial computation point to a first termination computation point in the data integration model according to the first resource data integration list, including: and determining a preset resource data integration list from a first initial calculation point to a first termination calculation point in the data integration model according to the first resource data integration list and the second resource data integration list, wherein the first track and the second track are passed from the first initial calculation point to the first termination calculation point.
Further, the cloud platform is specifically configured to:
and determining the first track as a track passed by a statistical planning track from the first initial calculation point to the first termination calculation point under the condition that the preset resource data integration list from the first initial calculation point to the first termination calculation point meets a preset statistical condition.
Further, the cloud platform is specifically configured to:
determining the first track as a track passed by a statistical planning track from the first initial calculation point to the first termination calculation point when the preset resource data integration list from the first initial calculation point to the first termination calculation point is less than or equal to a preset list identification result;
or determining the first trajectory as a trajectory through which a statistical planning trajectory from the first initial computation point to the first termination computation point passes when the preset resource data integration list from the first initial computation point to the first termination computation point is a minimum value among a plurality of preset resource data integration lists, where the plurality of preset resource data integration lists have a mapping relationship with a plurality of planning trajectories, each preset resource data integration list in the plurality of preset resource data integration lists is a preset resource data integration list of a corresponding planning trajectory in the plurality of planning trajectories, the plurality of planning trajectories include the first trajectory, and the plurality of planning trajectories are trajectories through which different planning trajectories from the first initial computation point to the first termination computation point pass respectively.
Based on the same inventive concept, please refer to fig. 3, a functional block diagram of an artificial intelligence based big data resource integration apparatus 500 is also provided, and the following describes details of the artificial intelligence based big data resource integration apparatus 500.
An artificial intelligence-based big data resource integration device 500 applied to a cloud platform, the device 500 comprising:
a data description obtaining module 510, configured to obtain a track user data set of a first track and customer description class data on the first track, where the first track has a one-to-one correspondence with a first target user data set neural model, the first track is a track in a data integration model, and the track user data set of the first track is used to represent an association relationship between a first group of data integration classifications included in the first track and a first data integration classification in the first group of data integration classifications;
a data integration determining module 520, configured to input the track user data set of the first track and the customer description class data on the first track into the first target user data set neural model, so as to obtain a first resource data integration list output by the first target user data set neural model;
an integration data calculation module 530, configured to determine, according to the first resource data integration list, a preset resource data integration list from a first initial calculation point to a first termination calculation point in the data integration model, where the first trajectory is traversed from the first initial calculation point to the first termination calculation point.
A cloud platform comprising a processor and a memory in communication with each other, the processor being configured to retrieve a computer program from the memory and to implement the method of any of figure 2 by running the computer program.
A computer-readable storage medium, on which a computer program is stored which, when executed, implements the method of any of fig. 2.
To sum up, a big data resource integration method, a system and a cloud platform based on artificial intelligence are provided, which comprises obtaining a track user data set of a first track and customer description data on the first track, inputting the track user data set of the first track and the customer description data on the first track into a first target user data set neural model to obtain a first resource data integration list output by the first target user data set neural model, determining a preset resource data integration list from a first initial computing point to a first termination computing point in the data integration model according to the first resource data integration list, obtaining the track user data set of the first track and the customer description data of the first track from the data integration model to input the neural model to determine the preset resource data integration list, thereby achieving the purpose of improving the accuracy of determining the preset resource data integration list, therefore, the technical effects of optimizing the rate of resource integration, improving the integration accuracy of the preset resource data integration list and providing the optimal track selection for the user are achieved.
It will be understood that the present application is not limited to the precise arrangements described above and shown in the drawings and that various modifications and changes may be made without departing from the scope thereof. The scope of the application is limited only by the appended claims.
Claims (10)
1. A big data resource integration method based on artificial intelligence is characterized by comprising the following steps:
acquiring a track user data set of a first track and customer description class data on the first track, wherein the first track has a one-to-one correspondence relationship with a first target user data set neural model, the first track is a track in a data integration model, and the track user data set of the first track is used for representing an association relationship between a first group of data integration classifications included in the first track and a first data integration classification in the first group of data integration classifications;
inputting the track user data set of the first track and the customer description class data on the first track into the first target user data set neural model to obtain a first resource data integration list output by the first target user data set neural model;
and determining a preset resource data integration list from a first initial calculation point to a first termination calculation point in the data integration model according to the first resource data integration list, wherein the first track is passed from the first initial calculation point to the first termination calculation point.
2. The method of claim 1, wherein inputting the first trajectory user data set of the first trajectory and the customer description class data on the first trajectory into the first target user data set neural model, resulting in a first integrated list of resource data output by the first target user data set neural model, comprises:
according to an input track user data set of the first track and customer description class data on the first track, determining a description strategy of a first data type included in the track user data set and a description strategy of first key content included in the track user data set, wherein the first data type and a first data integration class have a one-to-one correspondence relationship, the first data type represents a corresponding data integration class in the first group of data integration classes, the first key content connects two first data types and represents that the two first data integration classes corresponding to the two first data types are matched, and the first data integration class is a data integration class in the first group of data integration classes and is used for representing a local track in the first track;
determining a reference coefficient of the first data type according to the description strategy of the first data type and the description strategy of the first key content;
and determining the first resource data integration list according to the reference coefficient of the first data type.
3. The method according to claim 2, wherein the determining the reference coefficient of the first data type according to the description policy of the first data type and the description policy of the first key content comprises:
determining a reference coefficient of a current data type on a next trade line according to a description policy of the current data type, a description policy of key content associated with the current data type, a description policy of an adapted data type of the current data type, and a reference coefficient of the adapted data type of the current data type on the current trade line, wherein the first data type comprises the current data type and the adapted data type of the current data type.
4. The method of claim 1, wherein determining a preset integration list of resource data from a first initial computation point to a first termination computation point in the data integration model according to the first integration list of resource data comprises:
determining a preset resource data integration list from the initial calculation point of the first track to the termination calculation point of the first track as the first resource data integration list when the first initial calculation point is the initial calculation point of the first track and the first termination calculation point is the termination calculation point of the first track;
and/or determining a preset resource data integration list from the initial calculation point of the first data integration classification to the termination calculation point of the first data integration classification as a sample list of the first data integration classification if the first initial calculation point is the initial calculation point of the first data integration classification and the first termination calculation point is the termination calculation point of the first data integration classification;
and/or in the case that the first initial calculation point is an initial calculation point of the plurality of data integration classifications and the first termination calculation point is a termination calculation point of the plurality of data integration classifications, determining a preset resource data integration list from the initial calculation point of the plurality of data integration classifications to the termination calculation point of the plurality of data integration classifications as an intersection of the sample lists of the plurality of data integration classifications.
5. The method of claim 1, further comprising:
training a first training data neural model according to a sample track user data set of a sample track, sample client description class data on the sample track and a real-time sample list of the sample track to obtain a first target user data set neural model, wherein under the condition that a value of an error range between the sample list output by the first training data neural model and the real-time sample list meets a preset condition, the training of the first training data neural model is finished, and the first training data neural model when the training is finished is determined as the first target user data set neural model.
6. The method of claim 1, further comprising:
and comparing the first resource data integration list with the real-time sample list of the first track to update the first target user data set neural model, wherein the first target user data set neural model is updated under the condition that the value of the error range between the first resource data integration list and the real-time sample list meets a preset condition.
7. The method of claim 1, further comprising:
acquiring a track user data set of a second track and customer description class data on the second track, wherein the second track has a one-to-one correspondence relationship with a second target user data set neural model, and the track user data set of the second track is used for representing an association relationship between a second group of data integration classification and a data integration classification in the second group of data integration classification included in the second track;
inputting the track user data set of the second track and the customer description class data on the second track into the second target user data set neural model to obtain a second resource data integration list output by the second target user data set neural model;
determining a preset resource data integration list from a first initial computation point to a first termination computation point in the data integration model according to the first resource data integration list, including: and determining a preset resource data integration list from a first initial calculation point to a first termination calculation point in the data integration model according to the first resource data integration list and the second resource data integration list, wherein the first track and the second track are passed from the first initial calculation point to the first termination calculation point.
8. The big data resource integration system based on artificial intelligence is characterized by comprising a data collection device and a cloud platform, wherein the data collection device is in communication connection with the cloud platform, and the cloud platform comprises:
the data description acquisition module is used for acquiring a track user data set of a first track and customer description class data on the first track, wherein the first track and a first target user data set neural model have a one-to-one correspondence relationship, the first track is a track in a data integration model, and the track user data set of the first track is used for representing an association relationship between a first group of data integration classifications included in the first track and a first data integration classification in the first group of data integration classifications;
a data integration determining module, configured to input the track user data set of the first track and the customer description class data on the first track into the first target user data set neural model, so as to obtain a first resource data integration list output by the first target user data set neural model;
and the integrated data calculation module is used for determining a preset resource data integration list from a first initial calculation point to a first termination calculation point in the data integration model according to the first resource data integration list, wherein the first track is passed from the first initial calculation point to the first termination calculation point.
9. A cloud platform comprising a processor and a memory in communication with each other, the processor being configured to retrieve a computer program from the memory and to implement the method of any one of claims 1 to 7 by running the computer program.
10. A computer-readable storage medium, on which a computer program is stored which, when executed, implements the method of any of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110666255.1A CN113239034A (en) | 2021-06-16 | 2021-06-16 | Big data resource integration method and system based on artificial intelligence and cloud platform |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110666255.1A CN113239034A (en) | 2021-06-16 | 2021-06-16 | Big data resource integration method and system based on artificial intelligence and cloud platform |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113239034A true CN113239034A (en) | 2021-08-10 |
Family
ID=77140090
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110666255.1A Withdrawn CN113239034A (en) | 2021-06-16 | 2021-06-16 | Big data resource integration method and system based on artificial intelligence and cloud platform |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113239034A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114036347A (en) * | 2021-11-18 | 2022-02-11 | 北京中关村软件园发展有限责任公司 | Cloud platform supporting digital fusion service and working method |
-
2021
- 2021-06-16 CN CN202110666255.1A patent/CN113239034A/en not_active Withdrawn
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114036347A (en) * | 2021-11-18 | 2022-02-11 | 北京中关村软件园发展有限责任公司 | Cloud platform supporting digital fusion service and working method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111695697B (en) | Multiparty joint decision tree construction method, equipment and readable storage medium | |
| WO2021147486A1 (en) | Data processing method and apparatus | |
| CN107633257B (en) | Data quality evaluation method and device, computer readable storage medium and terminal | |
| WO2021208535A1 (en) | Recommendation method and device based on automatic feature grouping | |
| CN114418035A (en) | Decision tree model generation method and data recommendation method based on decision tree model | |
| CN108681493B (en) | Data anomaly detection method, device, server and storage medium | |
| CN111352965A (en) | Training method of sequence mining model, and processing method and equipment of sequence data | |
| WO2021035412A1 (en) | Automatic machine learning (automl) system, method and device | |
| CN115130711A (en) | Data processing method and device, computer and readable storage medium | |
| CN111459898A (en) | Machine learning method, computer-readable recording medium, and machine learning apparatus | |
| CN111310918B (en) | Data processing method, device, computer equipment and storage medium | |
| CN108647714A (en) | Acquisition methods, terminal device and the medium of negative label weight | |
| WO2020135054A1 (en) | Method, device and apparatus for video recommendation and storage medium | |
| CN113239034A (en) | Big data resource integration method and system based on artificial intelligence and cloud platform | |
| EP4252151A1 (en) | Data source correlation techniques for machine learning and convolutional neural models | |
| CN113472860A (en) | Service resource allocation method and server under big data and digital environment | |
| CN111353728A (en) | Risk analysis method and system | |
| Vaghela et al. | Boost a weak learner to a strong learner using ensemble system approach | |
| CN117033997A (en) | Data segmentation method, device, electronic equipment and medium | |
| CN114898184A (en) | Model training method, data processing method and device and electronic equipment | |
| CN113837677A (en) | Method, device and equipment for determining logistics line generation strategy | |
| CN112199287A (en) | Cross-project software defect prediction method based on enhanced hybrid expert model | |
| US20240104422A1 (en) | Transfer knowledge from auxiliary data for more inclusive machine learning models | |
| US20220245469A1 (en) | Decision Making Using Integrated Machine Learning Models and Knowledge Graphs | |
| KR102229768B1 (en) | Collective Intelligence Case-Based Reasoning System and its Method for user-customized recommendation |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WW01 | Invention patent application withdrawn after publication |
Application publication date: 20210810 |
|
| WW01 | Invention patent application withdrawn after publication |