Post target-based learning map generation method and system

Through the learning map generation method based on job goals, convolutional neural networks and multi-layer neural networks, combined with knowledge graphs and personal ability portraits, the problem of inaccurate matching between jobs and learners in the existing technology is solved, and more accurate Job recommendation and ability development.

CN119027075APending Publication Date: 2024-11-26GUANGZHOU ZHIXIN CLOUD TECH CO LTD

Patent Information

Application Number
CN202410867027.4
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-01
Publication Date
2024-11-26

AI Technical Summary

Technical Problem

There are inaccurate problems in the matching of jobs and learning in the existing technology, which leads to high trial and error costs for employment and makes it difficult to achieve accurate matching between each employer and job.

Method used

The learning map generation method based on job goals is adopted, and the convolutional neural network and multi-layer neural network are combined with knowledge graphs and personal ability portraits to update job matching and employment competitiveness in real time, and automatically push related job information.

Benefits of technology

It improves the accuracy and simplicity of the learning map, reduces the amount of calculation, and achieves more accurate matching and recommendations between jobs and learners' abilities.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119027075A_ABST
    Figure CN119027075A_ABST
Patent Text Reader

Abstract

The invention relates to the technical field of human resource allocation, in particular to a post-target-based learning map generation method and system, which is characterized in that a post-oriented learning map is generated based on post ability standards and learner characteristics, related courses are learned and completed according to the learning map, and the system automatically generates personal ability portraits, so that the personal ability portraits can be acquired. And according to convolutional neural network learning, the post-oriented matching degree and employment competitiveness are updated in real time, and the system automatically pushes related post information according to the personal ability portrait and the employment competitiveness index. The post target and the learning map are processed through the artificial neural network, the effect of the learning map can be more concise, and post recommendation and post ability cultivation of the learning map are more accurate according to calculation of the fitness degree.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of human resource deployment, and in particular to a method and system for generating a learning map based on job objectives. Background Art

[0002] There is a certain matching phenomenon between the matching degree between positions and learning and the job skills training and job certification courses. In addition, since some jobs have rigid requirements that the employees of the jobs have corresponding skills certificates, there are certain differences between skills certificates, jobs and job learning. In addition, the trial and error cost of job employment is relatively high for most workers. According to the public technology of the Chinese invention patent with the public application number of 2021102952019 (Intelligent talent recommendation and training platform and method based on learning map and job portrait), it is indicated that through the big data analysis of members' learning, interaction and sharing, the weighted value is designed by referring to the memory curve to obtain the long-term and short-term knowledge and ability of big data intelligent analysis, and the algorithm is designed by combining the work background and the ability reference of the upper and lower positions, and the needs of past and future positions are considered. Through the precise matching of big data, the talent recommendation and conscious training of the position are realized, and the long-term development of the enterprise is realized. However, in learning, only through big data analysis, it is not accurate for every employee to enjoy a more accurate matching of job goals, so a learning map generation method and system based on job goals is urgently needed to solve the above problems. Summary of the invention

[0003] The purpose of the present invention is to propose a method and system for generating a learning map based on job objectives, so as to solve one or more technical problems existing in the prior art and at least provide a beneficial choice or create conditions.

[0004] A method for generating a learning map based on a job objective, the method comprising the following steps:

[0005] S100: Generate a job-oriented learning map based on job competency standards and learner characteristics;

[0006] S200: After completing relevant courses based on the learning map, the system automatically generates a personal ability profile;

[0007] S300: Based on convolutional neural network learning, it updates job matching and employment competitiveness in real time;

[0008] S400: The system automatically pushes relevant job information based on personal ability profiles and employment competitiveness indicators.

[0009] Further, in step S100, the job capability labeling includes: according to the general job requirements with the job characteristics in the job library, the learner characteristics include: the learner's job selection, the learner's ability corresponding to the job, the match between the learner and the job and the learner's employment competitiveness in the job.

[0010] Furthermore, in step S200, according to the guidance of the learning map, the learner's personal ability is counted after learning the relevant job courses, and a personal ability portrait is generated. The personal ability portrait includes: learning course progress, learning course scores, and simulated employment scores.

[0011] Further, in step S300, the specific method of updating the job-oriented matching degree and employment competitiveness in real time according to the convolutional neural network learning includes the following steps:

[0012] S301: Define the job objective as Or, define the job information as In, and define the job requirement as Y. Store the job information according to the job information In as the main node, build a knowledge graph according to the node storage, and perform distributed storage of the job objective Or, job information In, and job requirement Y in the knowledge graph. Perform data transmission by storing different nodes and connecting the storage nodes.

[0013] The current job target state is defined, and the job target state includes the job target Or, job information In, job requirement Y and the connection of their storage nodes. The current job target state is defined as G(Or, In, Y), wherein G represents the current job target state. According to the current state G, it is judged whether the connection between the job target and the knowledge graph meets the standard. The judgment method is: setting a hit rule, and scoring if the hit rule is triggered during monitoring, and not scoring if the hit rule is not triggered. By judging whether the score value reaches the qualified value, it is judged whether the maintenance of the area meets the standard, and whether it meets the standard and the score are recorded as the maintenance state. The job target state G(Or, In, Y) is stored in the storage, and the knowledge graph is optimized according to the judgment rule.

[0014] Furthermore, the matching degree of the position is calculated by connecting the position objectives with the position requirements, and the employment competitiveness is analyzed;

[0015] Calculate the ratio of passed courses to failed courses during the learning process within the cycle time, obtain the time when the courses meet the standard in different time periods, and calculate the job matching degree Zi, Zi = |G(Or, In, Y)| / t, where t is the cycle time period, Zi is the matching degree of the job from the unqualified to the qualified in the job target state at the i-th moment in the cycle, construct the set [a] with the matching degree Zi, obtain the maximum conversion value max(a) in the set [a], and the maximum conversion value max(b) in the set [b];

[0016] S302: Calculate and obtain the job matching conversion rate D for the job target status in the region at the current moment.

[0017]

[0018] Where n is the total number of elements in set [a], and mean() is the function for obtaining the average value;

[0019] S303: Obtain the corresponding fitness degree and storage node position in the job target state;

[0020] S304: Match the conversion rate of the corresponding position in the position target state with the storage node position and the calculated conversion rate D to construct and update the learning map.

[0021] Furthermore, in step S400, continuously updating the job-oriented matching degree and employment competitiveness includes the following steps:

[0022] S401: Obtain the job target state G (Or, In, Y) in the monitoring network model and obtain the average conversion rate of all regions at the current moment The e represents the current moment e, and according to the monitoring network model, the conversion rate D is set to be greater than the average conversion rate. , it is recorded as abnormal conversion state and marked as false. If the conversion rate D is less than or equal to the average conversion rate When , mark true, and record the mark updated in real time in the regional node corresponding to the monitoring network model;

[0023] S402: training a multi-layer neural network according to the mark, inputting the job target state G (Or, In, Y) and the corresponding mark, and putting the output into the neural network model for training. After the training is completed, a quick judgment index X is output for the input maintenance status data. The judgment index X is used to judge whether the area meets the qualified standard at this point in the operation cycle. Each judgment index X needs to be trained through a separate neural network model to obtain the mapping relationship between the input data and the performance index;

[0024] Among them, in the neural network model, the number of neurons in the network input layer is equal to the number of input data types, the number of neurons in the output layer is equal to the number of output data types, and the number of intermediate layers and the number of neurons in each layer are freely set. The setting of the intermediate layer will directly affect the accuracy of the approximate data of the neural network. According to the training results, it is judged whether the number of intermediate layers or neurons needs to be increased or decreased. The calculation method for judging the number of intermediate layers and neurons is:

[0025] The intermediate layers in the neural network model are added layer by layer, and the judgment index X corresponding to the output of the intermediate layer number is named. When the intermediate layer number is 1, the judgment index X1 is output, and so on. According to the judgment index X, a nonlinear regression function is constructed. a is a constant value, and the variance E of the function's independent variable is calculated. Where L is the number of intermediate layers at output, X i is the i-th output

[0026] The bit judgment index X, the X i is the input position target state data corresponding to the output i-th judgment index X. When A takes the minimum value, To obtain the partial derivative of the function, A is a parameter for determining the number of intermediate layers. The value of A is compared with the variance E. If E>A, the output judgment index X is input into the neural network model, and the number of intermediate layers is increased by 1. If E≤A, the judgment index X is output;

[0027] Update the weight coefficients of neurons by gradient descent:

[0028]

[0029] in, To obtain the partial derivative of the function, P and Q are the system output error and the neuron weight increment respectively, both of which are constant values, ω(l) is the neuron weight coefficient, Δω(l) is the updated neuron weight coefficient, δ represents the neuron learning rate, and the number of neurons set in the middle layer is determined by the Δω(l).

[0030] S403: Obtain the storage node coordinates of the conversion rate marked as False, obtain the Euclidean distances Ek` of all adjacent coordinates of all corresponding marked points marked as False, construct a distance sequence [Ek`], reorder the distance sequence [Ek`] according to size, update the distance sequence [Ek`], obtain the node end number corresponding to the updated distance sequence [Ek`], and construct a node end sequence [m], determine the node end in the data abnormal area, narrow the scope of the abnormal area, construct a set [m] of node ends in the abnormal area, determine the node ends in the set as abnormal node ends, and feed back the maintenance status of the abnormal node ends [m] in the abnormal area to the learning map.

[0031] Among them, in the neural network model, the number of neurons in the network input layer is equal to the number of input data types, the number of neurons in the output layer is equal to the number of output data types, and the number of intermediate layers and the number of neurons in each layer are freely set. The setting of the intermediate layer will directly affect the accuracy of the approximate data of the neural network. According to the training results, it is judged whether the number of intermediate layers or neurons needs to be increased or decreased. The calculation method for judging the number of intermediate layers and neurons is:

[0032] The intermediate layers in the neural network model are added layer by layer, and the judgment index X corresponding to the output of the intermediate layer number is named. When the intermediate layer number is 1, the judgment index X1 is output, and so on. According to the judgment index X, a nonlinear regression function is constructed. a is a constant value, and the variance E of the function's independent variable is calculated. Where L is the number of intermediate layers at output, X i is the i-th output

[0033] `

[0034] The bit judgment index X, the X i is the input position target state data corresponding to the output i-th judgment index X. When A takes the minimum value, To obtain the partial derivative of the function, A is a parameter for determining the number of intermediate layers. The value of A is compared with the variance E. If E>A, the output judgment index X is input into the neural network model, and the number of intermediate layers is increased by 1. If E≤A, the judgment index X is output;

[0035] Update the weight coefficients of neurons by gradient descent:

[0036]

[0037] in, To obtain the partial derivative of the function, P and Q are the system output error and the neuron weight increment respectively, both of which are constant values, ω(l) is the neuron weight coefficient, Δω(l) is the updated neuron weight coefficient, δ represents the neuron learning rate, and the number of neurons set in the middle layer is determined by the Δω(l).

[0038] S403: Obtain the storage node coordinates of the conversion rate marked as False, obtain the Euclidean distances Ek` of all adjacent coordinates of all corresponding marked points marked as False, construct a distance sequence [Ek`], reorder the distance sequence [Ek`] according to size, update the distance sequence [Ek`], obtain the node end number corresponding to the updated distance sequence [Ek`], and construct a node end sequence [m], determine the node end in the data abnormal area, narrow the scope of the abnormal area, construct a set [m] of node ends in the abnormal area, determine the node ends in the set as abnormal node ends, and feed back the maintenance status of the abnormal node ends [m] in the abnormal area to the learning map.

[0039] A learning map generation system based on job objectives, the system comprising: a data acquisition module, a data analysis module and a data storage module, the data transmission module and the data analysis module implementing any one of the steps of the learning map generation method based on job objectives in the above method when executing the computer program.

[0040] The beneficial effects of the present invention are as follows: by processing the job objectives and learning maps through artificial neural networks, the learning map effect can be made more concise, and based on the fitness calculation, the job recommendations and job capability training of the rest map can be made more accurate, and by improving the middle layer of the neural network, the amount of calculation can be greatly reduced and the data can be made more accurate. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] The above and other features of the present invention will become more obvious by describing in detail the embodiments shown in the accompanying drawings. The same reference numerals in the accompanying drawings of the present invention represent the same or similar elements. It is obvious that the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative work.

[0042] In the figure:

[0043] Figure 1 Flowchart of a method for generating a learning map based on job objectives. DETAILED DESCRIPTION

[0044] The following will be combined with the embodiments and drawings to clearly and completely describe the concept, specific structure and technical effects of the present invention, so as to fully understand the purpose, scheme and effect of the present invention. It should be noted that the embodiments and features in the embodiments of this application can be combined with each other without conflict.

[0045] like Figure 1As shown, the purpose of the present invention is to propose a method and system for generating a learning map based on job objectives to solve one or more technical problems existing in the prior art and at least provide a beneficial choice or create conditions.

[0046] A method for generating a learning map based on a job objective, the method comprising the following steps:

[0047] S100: Generate a job-oriented learning map based on job competency standards and learner characteristics;

[0048] S200: After completing relevant courses based on the learning map, the system automatically generates a personal ability profile;

[0049] S300: Based on convolutional neural network learning, it updates job matching and employment competitiveness in real time;

[0050] S400: The system automatically pushes relevant job information based on personal ability profiles and employment competitiveness indicators.

[0051] Further, in step S100, the job capability labeling includes: according to the general job requirements with the job characteristics in the job library, the learner characteristics include: the learner's job selection, the learner's ability corresponding to the job, the match between the learner and the job and the learner's employment competitiveness in the job.

[0052] Furthermore, in step S200, according to the guidance of the learning map, the learner's personal ability is counted after learning the relevant job courses, and a personal ability portrait is generated. The personal ability portrait includes: learning course progress, learning course scores, and simulated employment scores.

[0053] Further, in step S300, the specific method of updating the job-oriented matching degree and employment competitiveness in real time according to the convolutional neural network learning includes the following steps:

[0054] S301: Define the job objective as Or, define the job information as In, and define the job requirement as Y. Store the job information according to the job information In as the main node, build a knowledge graph according to the node storage, and perform distributed storage of the job objective Or, job information In, and job requirement Y in the knowledge graph. Perform data transmission by storing different nodes and connecting the storage nodes.

[0055] The current job target state is defined, and the job target state includes the job target Or, job information In, job requirement Y and the connection of their storage nodes. The current job target state is defined as G(Or, In, Y), wherein G represents the current job target state. According to the current state G, it is judged whether the connection between the job target and the knowledge graph meets the standard. The judgment method is: setting a hit rule, and scoring if the hit rule is triggered during monitoring, and not scoring if the hit rule is not triggered. By judging whether the score value reaches the qualified value, it is judged whether the maintenance of the area meets the standard, and whether it meets the standard and the score are recorded as the maintenance state. The job target state G(Or, In, Y) is stored in the storage, and the knowledge graph is optimized according to the judgment rule.

[0056] Furthermore, the matching degree of the position is calculated by connecting the position objectives with the position requirements, and the employment competitiveness is analyzed;

[0057] Calculate the ratio of passed courses to failed courses during the learning process within the cycle time, obtain the time when the courses meet the standard in different time periods, and calculate the job matching degree Zi, Zi = |G(Or, In, Y)| / t, where t is the cycle time period, Zi is the matching degree of the job from the unqualified to the qualified in the job target state at the i-th moment in the cycle, construct the set [a] with the matching degree Zi, obtain the maximum conversion value max(a) in the set [a], and the maximum conversion value max(b) in the set [b];

[0058] S302: Calculate and obtain the job matching conversion rate D for the job target status in the region at the current moment.

[0059]

[0060] Where n is the total number of elements in set [a], and mean() is the function for obtaining the average value;

[0061] S303: Obtain the corresponding fitness degree and storage node position in the job target state;

[0062] S304: Match the conversion rate of the corresponding position in the position target state with the storage node position and the calculated conversion rate D to construct and update the learning map.

[0063] Furthermore, in step S400, continuously updating the job-oriented matching degree and employment competitiveness includes the following steps:

[0064] S401: Obtain the job target state G (Or, In, Y) in the monitoring network model and obtain the average conversion rate of all regions at the current moment The e represents the current moment e, and according to the monitoring network model, the conversion rate D is set to be greater than the average conversion rate. , it is recorded as abnormal conversion state and marked as false. If the conversion rate D is less than or equal to the average conversion rate When , mark true, and record the mark updated in real time in the regional node corresponding to the monitoring network model;

[0065] S402: training a multi-layer neural network according to the mark, inputting the job target state G (Or, In, Y) and the corresponding mark, and putting the output into the neural network model for training. After the training is completed, a quick judgment index X is output for the input maintenance status data. The judgment index X is used to judge whether the area meets the qualified standard at this point in the operation cycle. Each judgment index X needs to be trained through a separate neural network model to obtain the mapping relationship between the input data and the performance index;

[0066] Among them, in the neural network model, the number of neurons in the network input layer is equal to the number of input data types, the number of neurons in the output layer is equal to the number of output data types, and the number of intermediate layers and the number of neurons in each layer are freely set. The setting of the intermediate layer will directly affect the accuracy of the approximate data of the neural network. According to the training results, it is judged whether the number of intermediate layers or neurons needs to be increased or decreased. The calculation method for judging the number of intermediate layers and neurons is:

[0067] The intermediate layers in the neural network model are added layer by layer, and the judgment index X corresponding to the output of the intermediate layer number is named. When the intermediate layer number is 1, the judgment index X1 is output, and so on. According to the judgment index X, a nonlinear regression function is constructed. a is a constant value, and the variance E of the function's independent variable is calculated. Where L is the number of intermediate layers at output, X i is the i-th output

[0068] `

[0069] The bit judgment index X, the X i is the input position target state data corresponding to the output i-th judgment index X. When A takes the minimum value, To obtain the partial derivative of the function, A is a parameter for determining the number of intermediate layers. The value of A is compared with the variance E. If E>A, the output judgment index X is input into the neural network model, and the number of intermediate layers is increased by 1. If E≤A, the judgment index X is output;

[0070] Update the weight coefficients of neurons by gradient descent:

[0071]

[0072] in, To obtain the partial derivative of the function, P and Q are the system output error and the neuron weight increment respectively, both of which are constant values, ω(l) is the neuron weight coefficient, Δω(l) is the updated neuron weight coefficient, δ represents the neuron learning rate, and the number of neurons set in the middle layer is determined by the Δω(l).

[0073] S403: Obtain the storage node coordinates of the conversion rate marked as False, obtain the Euclidean distances Ek` of all adjacent coordinates of all corresponding marked points marked as False, construct a distance sequence [Ek`], reorder the distance sequence [Ek`] according to size, update the distance sequence [Ek`], obtain the node end number corresponding to the updated distance sequence [Ek`], and construct a node end sequence [m], determine the node end in the data abnormal area, narrow the scope of the abnormal area, construct a set [m] of node ends in the abnormal area, determine the node ends in the set as abnormal node ends, and feed back the maintenance status of the abnormal node ends [m] in the abnormal area to the learning map.

[0074] Among them, in the neural network model, the number of neurons in the network input layer is equal to the number of input data types, the number of neurons in the output layer is equal to the number of output data types, and the number of intermediate layers and the number of neurons in each layer are freely set. The setting of the intermediate layer will directly affect the accuracy of the approximate data of the neural network. According to the training results, it is judged whether the number of intermediate layers or neurons needs to be increased or decreased. The calculation method for judging the number of intermediate layers and neurons is:

[0075] The intermediate layers in the neural network model are added layer by layer, and the judgment index X corresponding to the output of the intermediate layer number is named. When the intermediate layer number is 1, the judgment index X1 is output, and so on. According to the judgment index X, a nonlinear regression function is constructed. a is a constant value, and the variance E of the function's independent variable is calculated. Where L is the number of intermediate layers at output, X i is the i-th output

[0076] `

[0077] The bit judgment index X, the X i is the input position target state data corresponding to the output i-th judgment index X. When A takes the minimum value, To obtain the partial derivative of the function, A is a parameter for determining the number of intermediate layers. The value of A is compared with the variance E. If E>A, the output judgment index X is input into the neural network model, and the number of intermediate layers is increased by 1. If E≤A, the judgment index X is output;

[0078] Update the weight coefficients of neurons by gradient descent:

[0079]

[0080] in, To obtain the partial derivative of the function, P and Q are the system output error and the neuron weight increment respectively, both of which are constant values, ω(l) is the neuron weight coefficient, Δω(l) is the updated neuron weight coefficient, δ represents the neuron learning rate, and the number of neurons set in the middle layer is determined by the Δω(l).

[0081] S403: Obtain the storage node coordinates of the conversion rate marked as False, obtain the Euclidean distances Ek` of all adjacent coordinates of all corresponding marked points marked as False, construct a distance sequence [Ek`], reorder the distance sequence [Ek`] according to size, update the distance sequence [Ek`], obtain the node end number corresponding to the updated distance sequence [Ek`], and construct a node end sequence [m], determine the node end in the data abnormal area, narrow the scope of the abnormal area, construct a set [m] of node ends in the abnormal area, determine the node ends in the set as abnormal node ends, and feed back the maintenance status of the abnormal node ends [m] in the abnormal area to the learning map.

[0082] A learning map generation system based on job objectives, the system comprising: a data acquisition module, a data analysis module and a data storage module, the data transmission module and the data analysis module implementing any one of the steps of the learning map generation method based on job objectives in the above method when executing the computer program.

[0083] The processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete component gate circuits or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor, etc. The processor is the control center of the learning map generation system based on job objectives, and uses various interfaces and lines to connect the various sub-areas of the entire learning map generation system based on job objectives.

[0084] The memory can be used to store the computer program and / or module, and the processor realizes various functions of the learning map generation system based on job objectives by running or executing the computer program and / or module stored in the memory and calling the data stored in the memory. The memory can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, an application required for at least one function (such as a sound playback function, an image playback function, etc.), etc.; the data storage area can store data created according to the use of the mobile phone (such as audio data, a phone book, etc.), etc. In addition, the memory can include a high-speed random access memory, and can also include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), at least one disk storage device, a flash memory device, or other volatile solid-state storage devices.

[0085] Although the description of the present invention has been quite detailed and has been described in particular with respect to several described embodiments, it is not intended to be limited to any of these details or embodiments or any particular embodiment, so as to effectively cover the intended scope of the present invention. In addition, the present invention is described above with the embodiments foreseeable by the inventors, and its purpose is to provide a useful description, and those non-substantial changes to the present invention that are not currently foreseen may still represent equivalent changes of the present invention.

Claims

1. A method for generating a learning map based on job objectives, characterized in that: The method comprises the following steps: S100: Generate a job-oriented learning map based on job competency standards and learner characteristics; S200: After completing relevant courses based on the learning map, the system automatically generates a personal ability profile; S300: Based on convolutional neural network learning, it updates job matching and employment competitiveness in real time; S400: The system automatically pushes relevant job information based on personal ability profiles and employment competitiveness indicators.

2. A method for generating a learning map based on job objectives according to claim 1, characterized in that: In step S100, the job capability labeling includes: according to the general job requirements with the job characteristics in the job database, the learner characteristics include: the learner's job selection, the learner's ability corresponding to the job, the match between the learner and the job and the learner's employment competitiveness in the job.

3. The method for generating a learning map based on job objectives according to claim 1, characterized in that: In step S200, according to the guidance of the learning map, the learner's personal ability is counted after learning the relevant job courses, and a personal ability portrait is generated. The personal ability portrait includes: learning course progress, learning course scores, and simulated employment scores.

4. The method for generating a learning map based on job objectives according to claim 1, characterized in that: In step S300, the specific method of updating the job-oriented matching degree and employment competitiveness in real time according to the convolutional neural network learning includes the following steps: S301: Define the job objective as Or, define the job information as In, and define the job requirement as Y. Store the job information according to the job information In as the main node, build a knowledge graph according to the node storage, and perform distributed storage of the job objective Or, job information In, and job requirement Y in the knowledge graph. Perform data transmission by storing different nodes and connecting the storage nodes. The current job target state is defined, and the job target state includes the job target Or, job information In, job requirement Y and the connection of their storage nodes. The current job target state is defined as G(Or, In, Y), wherein G represents the current job target state. According to the current state G, it is judged whether the connection between the job target and the knowledge graph meets the standard. The judgment method is: setting a hit rule, and scoring if the hit rule is triggered during monitoring, and not scoring if the hit rule is not triggered. By judging whether the score value reaches the qualified value, it is judged whether the maintenance of the area meets the standard, and whether it meets the standard and the score are recorded as the maintenance state. The job target state G(Or, In, Y) is stored in the storage, and the knowledge graph is optimized according to the judgment rule.

5. A method for generating a learning map based on job objectives according to claim 4, characterized in that: The matching degree of the position is calculated based on the connection between the position objectives and the position requirements, and the employment competitiveness is analyzed; Calculate the ratio of passed courses to failed courses during the learning process within the cycle time, obtain the time when the courses meet the standard in different time periods, and calculate the job matching degree Zi, Zi = |G(Or, In, Y)| / t, where t is the cycle time period, Zi is the matching degree of the job from the unqualified to the qualified in the job target state at the i-th moment in the cycle, construct the set [a] with the matching degree Zi, obtain the maximum conversion value max(a) in the set [a], and the maximum conversion value max(b) in the set [b]; S302: Calculate and obtain the job matching conversion rate D for the job target status in the region at the current moment. Where n is the total number of elements in set [a], and mean() is the function for obtaining the average value; S303: Obtain the corresponding fitness degree and storage node position in the job target state; S304: Match the conversion rate of the corresponding position in the position target state with the storage node position and the calculated conversion rate D to construct and update the learning map.

6. The method for generating a learning map based on job objectives according to claim 1, characterized in that: In step S400, continuously updating the job-oriented matching degree and employment competitiveness includes the following steps: S401: Obtain the job target state G (Or, In, Y) in the monitoring network model and obtain the average conversion rate of all regions at the current moment The e represents the current moment e, and according to the monitoring network model, the conversion rate D is set to be greater than the average conversion rate. , it is recorded as abnormal conversion state and marked as false. If the conversion rate D is less than or equal to the average conversion rate When , mark true, and record the mark updated in real time in the regional node corresponding to the monitoring network model; S402: training a multi-layer neural network according to the mark, inputting the job target state G (Or, In, Y) and the corresponding mark, and putting the output into the neural network model for training. After the training is completed, a quick judgment index X is output for the input maintenance status data. The judgment index X is used to judge whether the area meets the qualified standard at this point in the operation cycle. Each judgment index X needs to be trained through a separate neural network model to obtain the mapping relationship between the input data and the performance index; Among them, in the neural network model, the number of neurons in the network input layer is equal to the number of input data types, the number of neurons in the output layer is equal to the number of output data types, and the number of intermediate layers and the number of neurons in each layer are freely set. The setting of the intermediate layer will directly affect the accuracy of the approximate data of the neural network. According to the training results, it is judged whether the number of intermediate layers or neurons needs to be increased or decreased. The calculation method for judging the number of intermediate layers and neurons is: The intermediate layers in the neural network model are added layer by layer, and the judgment index X corresponding to the output of the intermediate layer number is named. When the intermediate layer number is 1, the judgment index X1 is output, and so on. According to the judgment index X, a nonlinear regression function is constructed. a is a constant value, and the variance E of the function's independent variable is calculated. Where L is the number of intermediate layers at output, X i is the i-th judgment index X output, where X i ` is the input position target state data corresponding to the output i-th judgment index X. When A takes the minimum value, To obtain the partial derivative of the function, A is a parameter for determining the number of intermediate layers. The value of A is compared with the variance E. If E>A, the output judgment index X is input into the neural network model, and the number of intermediate layers is increased by 1. If E≤A, the judgment index X is output; Update the weight coefficients of neurons by gradient descent: in, To obtain the partial derivative of the function, P and Q are the system output error and the neuron weight increment respectively, both of which are constant values, ω(l) is the neuron weight coefficient, Δω(l) is the updated neuron weight coefficient, δ represents the neuron learning rate, and the number of neurons set in the middle layer is determined by the Δω(l). S403: Obtain the storage node coordinates of the conversion rate marked as False, obtain the Euclidean distances Ek` of all adjacent coordinates of all corresponding marked points marked as False, construct a distance sequence [Ek`], reorder the distance sequence [Ek`] according to size, update the distance sequence [Ek`], obtain the node end number corresponding to the updated distance sequence [Ek`], and construct a node end sequence [m], determine the node end in the data abnormal area, narrow the scope of the abnormal area, construct a set [m] of node ends in the abnormal area, determine the node ends in the set as abnormal node ends, and feed back the maintenance status of the abnormal node ends [m] in the abnormal area to the learning map.

7. A learning map generation system based on job objectives, characterized in that: The system comprises: a data acquisition module, a data analysis module and a data storage module. When the data transmission module and the data analysis module execute the computer program, the steps of any one of the job objective-based learning map generation methods in claims 1-6 are implemented.

Citation Information

Patent Citations

  • Fire protection and maintenance monitoring method and system based on AI

    CN118447454A

Cited By

  • Method, apparatus, computer readable medium and computer program product for course recommendation based on deep learning

    CN121458506A