Electronic electrician skill assessment optimization management method and system based on artificial intelligence

Abstract

The invention discloses an electronic electrician skill assessment optimization management method and system based on artificial intelligence. The method comprises the steps of system structure design, assessment question generation, examinee distribution management, score management optimization and skill assessment management. The invention relates to the technical field of electrician assessment management, in particular to an electronic electrician skill assessment optimization management method and system based on artificial intelligence. The electronic electrician skill assessment optimization management system based on artificial intelligence is constructed, and three intelligent functions are achieved through an intelligent management module; the automation and the intelligent value of the whole system are improved; assessment questions are generated by adopting a reinforcement learning method combined with project reflection factor analysis, and good intelligent scheme enlightenment is provided for the whole assessment process; and examinee distribution management is carried out by adopting a long-short-term memory neural network combining ant colony optimization and reinforcement learning, so that a guarantee is provided for high intelligence of the system.

Description

Technical Field

[0001] The present invention relates to the technical field of electrician assessment management, and specifically to an electronic and electrician skill assessment optimization management method and system based on artificial intelligence. Background Art

[0002] The AI-based electronic and electrical skills assessment optimization management method is a system that integrates AI technology and aims to improve the efficiency and accuracy of electronic and electrical skills assessment. This method optimizes the assessment process by using AI to generate assessment questions, assign candidates, and record and analyze scores. Specifically, it can intelligently generate targeted test questions, reasonably assign candidates, and record and analyze scores in real time, thereby improving the fairness, scientificity, and effectiveness of the assessment. Through this method, managers can better monitor assessment progress, identify skill training needs, and optimize training and assessment strategies based on data-driven insights.

[0003] However, in the existing electronic and electrician skills assessment optimization management system, there are technical problems that the electronic and electrician skills assessment is not intelligent enough, and the combination of electronic and electrician assessment and intelligent system is difficult to implement and apply; in the assessment question generation of the existing electronic and electrician skills assessment optimization management method, there is a technical problem that the effectiveness of the electronic and electrician assessment depends on the effectiveness of the question setting, while the traditional manual setting is not adaptable and targeted enough, and the application of a single model to analyze electronic and electrician skills test questions cannot handle complex assessment processes and difficult test questions; in the candidate allocation management of the existing electronic and electrician skills assessment optimization management method, there is a technical problem that the examination room resources for the electronic and electrician assessment are relatively limited, and how to optimize the time and candidate allocation of the specific assessment process is a complex issue involving the time and space resources of the candidates and the examination room, and the traditional intelligent method cannot take into account multiple factors, resulting in the intelligence of the candidate allocation management cannot meet the system requirements. Summary of the invention

[0004] In view of the above situation, in order to overcome the defects of the prior art, the present invention provides an electronic and electrician skill assessment optimization management method and system based on artificial intelligence. In view of the technical problems that the electronic and electrician skill assessment is not intelligent enough in the existing electronic and electrician skill assessment optimization management system, and the combination of electronic and electrician assessment and intelligent system is difficult to implement and apply, this solution creatively constructs an electronic and electrician skill assessment optimization management system based on artificial intelligence, and carries out intelligent layout and implementation of specific electronic and electrician skill assessment through the host computer control module and the evaluation and assessment module, and realizes three intelligent functions through the intelligent management module, thereby improving the overall automation and intelligent value of the system; in view of the fact that in the generation of assessment questions of the existing electronic and electrician skill assessment optimization management method, the effectiveness of the electronic and electrician assessment depends on the effectiveness of the question setting, while the traditional manual setting is insufficient in adaptability and pertinence, and the application of a single model to analyze electronic and electrician skill test questions cannot handle complex assessment processes and difficult test questions, this solution creatively adopts the enhanced The learning method integrates multi-dimensional characteristics of candidates by introducing molecular analysis, and optimizes the generation process of test questions through reinforcement learning and adaptive strategies, which provides a good intelligent solution inspiration for the entire assessment process and improves the pertinence and adaptability of electronic and electrical assessments. In the candidate allocation management of the existing electronic and electrical skill assessment optimization management method, there is a relatively limited examination room resources for electronic and electrical assessments. How to optimize the time and candidate allocation of the specific assessment process is a complex problem involving the time and space resources of candidates and examination rooms. The traditional intelligent method cannot take into account multiple factors, resulting in the inability of the intelligence of candidate allocation management to meet the technical requirements of the system. This solution creatively uses a long short-term memory neural network combined with ant colony optimization and reinforcement learning for candidate allocation management. The ant colony optimization algorithm is used to strengthen the collaborative optimal solution of candidate and examination room resource allocation, and reinforcement learning is used to make intelligent decisions on candidate allocation. Finally, the long short-term memory network is used to improve the timing prediction and processing capabilities, and the time and space management capabilities of examination room resource allocation are comprehensively improved, providing a guarantee for the high intelligence of the system.

[0005] The technical solution adopted by the present invention is as follows: The electronic and electrician skill assessment optimization management method based on artificial intelligence provided by the present invention comprises the following steps:

[0006] Step S1: system structure design;

[0007] Step S2: generating test questions;

[0008] Step S3: candidate allocation management;

[0009] Step S4: Optimization of performance management;

[0010] Step S5: Skill assessment management.

[0011] Further, in step S1, the system structure design is used to construct the basic structure of the electronic and electrician skill assessment optimization management system, specifically to construct a host computer control module, an evaluation and assessment module, an intelligent management module and a data transmission module;

[0012] The host computer control module is used to set the assessment class, set the assessment test content, assign the assessment number, record and store the assessment score and output the assessment result, which is specifically implemented by the host computer control software;

[0013] The evaluation and assessment module is used to construct an embedded assessment device, which specifically includes a single-chip microcomputer, a serial port LCD screen, a measured component output circuit, and a measured waveform output circuit;

[0014] The intelligent management module specifically includes an examination question generation unit, a candidate allocation management unit and a score management optimization unit;

[0015] The data transmission module is used to collect original data and realize data transmission between functional modules, specifically, to obtain the original data set of assessment management by collecting historical electronic and electrician skill assessment records and basic information of examinees;

[0016] The assessment management original data set specifically includes assessment question generation original data, examinee allocation management original data and score management optimization original data.

[0017] Further, in step S2, the assessment question generation is used to generate questions and adaptively adjust the difficulty of the questions in combination with the personalized skill level of the School of Electronic and Electrical Engineering, specifically generating original data based on the assessment questions in the assessment management original data set, using a reinforcement learning method combined with item reflection factor analysis to generate assessment questions, and obtaining assessment question generation output data;

[0018] The test question generates output data, specifically personalized electronic and electrician test question data;

[0019] The step of using the reinforcement learning method combined with item reflection factor analysis to generate test questions and obtain test question generation output data includes:

[0020] Step S21: data preparation, specifically generating original data according to the test questions, and obtaining test question generation optimization data through data extraction and manual feature extraction, specifically including question data, examinee data and question label data;

[0021] Step S22: Test question difficulty analysis, specifically, by constructing a project response theoretical model based on factor analysis, performing a dual-parameter test question difficulty analysis, obtaining a reference probability of correct answer rate of the test question, and optimizing the parameters of the project response theoretical model by constructing a maximized log-likelihood function, thereby constructing an optimized project response factor analysis model;

[0022] Step S23: Dynamic test question generation, specifically, using a reinforcement learning method, by constructing a deep Q model, and constructing state parameters, action parameters, reward parameters and strategy parameters, performing deep Q learning model training, and constructing a reinforcement learning question generation model;

[0023] Step S24: Training the assessment question generation model, specifically, by obtaining the optimization project reflection factor analysis model and the reinforcement learning question generation model, performing sub-model integration training to obtain the assessment question comprehensive generation model Model TG ;

[0024] Step S25: Generating test questions, specifically using the test questions to comprehensively generate a model Model TG , generate test questions, and optimize the test questions adaptively by building real-time adaptive feedback rules to obtain test question generation output data;

[0025] The real-time adaptive feedback rule specifically means that if the examinee continuously answers questions that are difficult and the difficulty analysis results show that they are difficult, the comprehensive generation model for test questions will increase the difficulty of the generated questions; if the examinee continuously answers the current questions incorrectly, the comprehensive generation model for test questions will reduce the difficulty of the generated questions.

[0026] Further, in step S3, the candidate allocation management is used to perform time management of specific assessments in combination with the results of assessment question generation during the actual assessment process, and further optimize the scheduling of candidates and the comprehensive assessment time arrangement, specifically, based on the candidate allocation management original data in the assessment management original data set, a long short-term memory neural network combining ant colony optimization and reinforcement learning is used to perform candidate allocation management, and candidate allocation management is performed to obtain candidate allocation management data;

[0027] The candidate allocation management data specifically includes examination time management reference data and candidate scheduling reference data;

[0028] The long short-term memory neural network combining ant colony optimization and reinforcement learning specifically includes an ant colony optimization algorithm, a deep Q subnet and a long short-term memory subnet;

[0029] The ant colony optimization algorithm is used to globally search for the optimal allocation scheme for examinees and examination room resources;

[0030] The deep Q subnetwork is used to allocate the optimal candidate scheduling allocation plan in a complex examination environment;

[0031] The long short-term memory subnet is used to optimize the time management plan during the examination process by predicting the temporal changes of the examination time and examination room resources of the examinees;

[0032] The steps of using a long short-term memory neural network combining ant colony optimization and reinforcement learning to perform candidate allocation management, performing candidate allocation management, and obtaining candidate allocation management data include:

[0033] Step S31: optimizing the examination room resource allocation scheme, specifically constructing an ant colony optimization algorithm, initializing the pheromone matrix, calculating the path selection probability and pheromone update, performing ant colony optimization modeling, and obtaining an examination room resource allocation optimization decision-making scheme;

[0034] Step S32: optimizing the candidate scheduling allocation scheme, specifically constructing a deep Q subnet, using the reinforcement learning method in step S23, through state definition and action space definition, to construct a deep Q subnet, and obtain an optimized candidate scheduling allocation scheme;

[0035] Step S33: optimizing the assessment time management plan, specifically constructing a long short-term memory subnet, by constructing a standard long short-term memory neural network including an input gate, a forget gate, and an output gate as a long short-term memory subnet, and performing time series data prediction through the long short-term memory subnet to obtain an assessment time management optimization plan;

[0036] Step S34: Train the candidate allocation management model, specifically, train the candidate allocation management model through the ant colony optimization algorithm, the deep Q subnet and the long short-term memory subnet to obtain the candidate allocation management model Model DM ;

[0037] Step S35: candidate allocation management, specifically using the candidate allocation management model Model DM , perform candidate allocation management based on the candidate allocation management original data to obtain candidate allocation management data.

[0038] Further, in step S4, the performance management optimization is used to record the electronic and electrician skill assessment results and perform performance clustering and change trend analysis, specifically, based on the performance management optimization original data in the assessment management original data set, the performance management optimization is performed using a long short-term memory neural network combined with mean clustering to obtain performance management optimization data;

[0039] The said score management optimization specifically includes candidate cluster analysis reference data and candidate score trend prediction data;

[0040] The long short-term memory neural network combined with mean clustering specifically clusters the examinees' scores by using the K-means clustering method, and uses the standard long short-term memory neural network to analyze the trend of changes in the examinees' scores.

[0041] Further, in step S5, the skill assessment management is used to comprehensively optimize the electronic and electrician skill assessment management by combining assessment question generation, candidate allocation management and score management optimization. Specifically, before the start of the electronic and electrician assessment, the assessment question generation is applied to obtain the assessment question generation output data, and the questions are adaptively adjusted to set specific assessment questions;

[0042] In the process of electronic and electrician assessment, by applying candidate allocation management, the candidate allocation management data is obtained, and specific examination room time management and candidate scheduling are carried out;

[0043] After the electronic and electrician assessment is completed, score management optimization data is obtained by applying score management optimization, and the candidate cluster analysis reference data and candidate score trend prediction data are used as the data source for adjusting the electronic and electrician assessment questions. Manual score management and analysis are then performed to obtain reference information for electronic and electrician assessment teaching and training.

[0044] The electronic and electrician skill assessment optimization management system based on artificial intelligence provided by the present invention includes a host computer control module, an evaluation and assessment module, an intelligent management module and a data transmission module;

[0045] The data transmission module is used for data transmission between modules and collection of original data. Through the collection of original data, an original data set for assessment management is obtained, and the original data set for assessment management is sent to the intelligent management module through data transmission;

[0046] The intelligent management module is used to apply assessment question generation, candidate allocation management and score management optimization. Through the assessment question generation, candidate allocation management and score management optimization, assessment question generation output data, candidate allocation management data, score management optimization data and electronic and electrician assessment teaching and training reference information are obtained, and the assessment question generation output data is sent to the upper computer control module through the data transmission module, and the assessment result data is received from the evaluation and assessment module and used for intelligent management, and the electronic and electrician assessment teaching and training reference information is used as the comprehensive analysis result of intelligent management;

[0047] The host computer control module is used to set and implement specific electronic and electrical skills assessment;

[0048] The evaluation and assessment module is used to construct an embedded assessment device and perform specific assessment tasks.

[0049] The beneficial effects achieved by the present invention using the above scheme are as follows:

[0050] (1) In view of the technical problems that the electronic and electrician skill assessment in the existing electronic and electrician skill assessment optimization management system is not intelligent enough and the combination of electronic and electrician skill assessment and intelligent system is difficult to implement and apply, this solution creatively constructs an electronic and electrician skill assessment optimization management system based on artificial intelligence, and uses the host computer control module and the evaluation and assessment module to carry out the intelligent layout and implementation of specific electronic and electrician skill assessment, and realizes three intelligent functions through the intelligent management module, thereby improving the overall automation and intelligent value of the system;

[0051] (2) In the generation of test questions in the existing electronic and electrical skills assessment optimization management method, there is a problem that the effectiveness of the electronic and electrical assessment depends on the effectiveness of the question setting, while the traditional manual setting is not adaptable and targeted enough, and the application of a single model to analyze the electronic and electrical skills test questions cannot handle the complex assessment process and technical problems of high-difficulty test questions. This solution creatively adopts a reinforcement learning method combined with item reflection factor analysis, integrates multi-dimensional candidate characteristics by introducing molecular analysis, and optimizes the test question generation process through reinforcement learning and adaptive strategies, providing a good intelligent solution inspiration for the entire assessment process, and also improving the pertinence and adaptability of the electronic and electrical assessment;

[0052] (3) In the existing electronic and electrical skills assessment optimization management method, the examination room resources for electronic and electrical skills assessment are relatively limited. How to optimize the time and candidate allocation of the specific assessment process is a complex problem involving the temporal and spatial resources of candidates and examination rooms. The traditional intelligent method cannot take into account multiple factors, resulting in the inability of the intelligence of candidate allocation management to meet the technical requirements of the system. This solution creatively uses a long short-term memory neural network that combines ant colony optimization and reinforcement learning to manage candidate allocation. The ant colony optimization algorithm is used to strengthen the collaborative optimal solution of candidate and examination room resource allocation, and reinforcement learning is used to make intelligent decisions on candidate allocation. Finally, the long short-term memory network is used to improve the time series prediction and processing capabilities, and the temporal and spatial management capabilities of examination room resource allocation are comprehensively improved, providing a guarantee for the high intelligence of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] Figure 1 A flow chart of the electronic and electrician skill assessment optimization management method based on artificial intelligence provided by the present invention;

[0054] Figure 2 A schematic diagram of the electronic and electrician skill assessment optimization management system based on artificial intelligence provided by the present invention;

[0055] Figure 3 A schematic diagram of the process of generating the assessment questions in step S2;

[0056] Figure 4A schematic diagram of the process of candidate allocation management in step S3;

[0057] Figure 5 This is a schematic diagram of the interface of the host computer control software in the host computer control module of this solution.

[0058] The accompanying drawings are used to provide further understanding of the present invention and constitute a part of the specification. They are used to explain the present invention together with the embodiments of the present invention and do not constitute a limitation of the present invention. DETAILED DESCRIPTION

[0059] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments; based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0060] In the description of the present invention, it should be understood that terms such as “upper”, “lower”, “front”, “back”, “left”, “right”, “top”, “bottom”, “inside” and “outside” indicating directions or positional relationships are based on the directions or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific direction, be constructed and operated in a specific direction, and therefore should not be understood as limiting the present invention.

[0061] Example 1, see Figure 1 The electronic and electrician skill assessment optimization management method based on artificial intelligence provided by the present invention comprises the following steps:

[0062] Step S1: system structure design;

[0063] Step S2: generating test questions;

[0064] Step S3: candidate allocation management;

[0065] Step S4: Optimization of performance management;

[0066] Step S5: Skill assessment management.

[0067] Example 2, see Figure 1 , Figure 2 and Figure 5 In step S1, the system structure design is used to construct the basic structure of the electronic and electrician skill assessment optimization management system, specifically to construct a host computer control module, an evaluation and assessment module, an intelligent management module and a data transmission module;

[0068] The host computer control module is used to set the assessment class, set the assessment test content, assign the assessment number, record and store the assessment score and output the assessment result, which is specifically implemented by the host computer control software;

[0069] The host computer control software is connected to the evaluation and assessment module through a gateway;

[0070] See also Figure 5 , the host computer control software is used to implement the assessment function of the host computer control module;

[0071] The evaluation and assessment module is used to construct an embedded assessment device, which specifically includes a single-chip microcomputer, a serial port LCD screen, a measured component output circuit, and a measured waveform output circuit;

[0072] The functions of the single chip microcomputer include testing communication and input and output control;

[0073] The assessment communication specifically uses Zigbee wireless communication technology to communicate with the host computer, obtain the assessment question content, and upload the assessment score to the host computer;

[0074] The input and output control specifically refers to controlling the output circuit of the measured component through the IO port to output the measured resistance, capacitance and diode, and controlling the measured waveform output circuit to generate different waveforms through the IO port, and the generated waveforms have different periods, frequencies, amplitudes and different DC components;

[0075] The functions of the serial port LCD screen include question type display, obtaining single chip control commands, displaying test content and screen touch-screen human-computer interaction;

[0076] The functions of the output circuit of the measured element include control signal reception and circuit information output;

[0077] The control signal receiving specifically refers to receiving the control signal of the single chip microcomputer through 74HC595D, and selecting different relays according to the control signal;

[0078] The circuit information output specifically refers to the resistance, capacitance or diode connected to the output relay, which is provided to the person being assessed for measurement.

[0079] The functions of the measured waveform output circuit include receiving a waveform control signal, controlling waveform generation, and controlling waveform output;

[0080] The waveform control signal reception specifically refers to receiving the control signal of the single chip microcomputer through AD9833, and generating different waveforms S according to the control signal;

[0081] The control waveform generation specifically refers to setting the period and frequency of the waveform S;

[0082] The control waveform output specifically refers to compressing the waveform S through MCP410101 and changing the amplitude of the signal to obtain a waveform S', and outputting a 0-5V DC voltage D through another MCP410101, and finally synthesizing the waveform S' and the DC voltage D through SGM8270 to output the measured signal;

[0083] The intelligent management module specifically includes an examination question generation unit, a candidate allocation management unit and a score management optimization unit;

[0084] The test question generating unit is used to generate questions and adaptively adjust the difficulty of the test questions;

[0085] The candidate allocation management unit is used to manage the examination time and candidate scheduling;

[0086] The performance management optimization unit is used to record assessment results and perform clustering and performance change trend analysis;

[0087] The data transmission module is used to collect original data and realize data transmission between functional modules, specifically, to obtain the original data set of assessment management by collecting historical electronic and electrician skill assessment records and basic information of examinees;

[0088] The assessment management original data set specifically includes assessment question generation original data, examinee allocation management original data and score management optimization original data;

[0089] The test questions generate raw data, specifically including test question bank data, examinee historical score data, question historical answer data and question usage frequency data;

[0090] The candidate allocation management raw data specifically includes candidate basic information data, examination room resource situation data, assessment index task data, candidate time arrangement history data and hardware equipment status data;

[0091] The score management optimization raw data specifically includes examinee historical score data, examinee operation record data and examinee score evaluation label data.

[0092] By executing the above operations, in order to address the technical problems in the existing electronic and electrician skills assessment optimization management system, such as the insufficient intelligence level of electronic and electrician skills assessment, and the difficulty in implementing and applying the combination of electronic and electrician assessment and intelligent systems, this solution creatively constructs an electronic and electrician skills assessment optimization management system based on artificial intelligence, and uses the host computer control module and the evaluation and assessment module to carry out the intelligent layout and implementation of specific electronic and electrician skills assessment, and realizes three intelligent functions through the intelligent management module, thereby improving the overall automation and intelligent value of the system.

[0093] Example 3, see Figure 1 , Figure 2 and Figure 3 , This embodiment is based on the above embodiment. In step S2, the assessment question generation is used to generate questions and adaptively adjust the difficulty of the questions in combination with the personalized skill level of the School of Electronic and Electrical Engineering. Specifically, the original data is generated based on the assessment questions in the assessment management original data set, and the reinforcement learning method combined with the project reflection factor analysis is used to generate the assessment questions to obtain the assessment question generation output data;

[0094] The test question generates output data, specifically personalized electronic and electrician test question data;

[0095] The step of using the reinforcement learning method combined with item reflection factor analysis to generate test questions and obtain test question generation output data includes:

[0096] Step S21: data preparation, specifically generating original data according to the test questions, and obtaining test question generation optimization data through data extraction and manual feature extraction, specifically including question data, examinee data and question label data;

[0097] The question data specifically includes question number data, question difficulty data and question skill requirement data, and the calculation formula is:

[0098] ;

[0099] In the formula, TA is the question data, q i is the topic number data, i is the topic index, d i is the difficulty data of the question, S i It is the data of the skill requirements of the question;

[0100] The candidate data specifically includes candidate number data, candidate skill data and candidate historical accuracy data, and the calculation formula is:

[0101] ;

[0102] In the formula, SA is the candidate data, p j is the candidate number data, j is the candidate index, is the candidate skill data, which is used to represent the skill level of the jth candidate, and is obtained by manual evaluation based on the candidate's historical accuracy rate. is the historical correct rate data of the examinee, which is used to indicate the historical correct rate of the jth examinee’s answer to question i;

[0103] The question label data specifically includes skill labels, difficulty labels, and historical answer labels;

[0104] Step S22: Test question difficulty analysis, specifically, by constructing a project response theoretical model based on factor analysis, performing a dual-parameter test question difficulty analysis, obtaining a reference probability of correct answer rate of the test question, and optimizing the parameters of the project response theoretical model by constructing a maximized log-likelihood function, thereby constructing an optimized project response factor analysis model;

[0105] The calculation formula of the theoretical model based on factor analysis is:

[0106] ;

[0107] Where P(·) is the probability that candidate j answers question i correctly, R ij is the answer of candidate j on question i, and its specific value range is {0,1}, where a value of 0 indicates an incorrect answer, and a value of 1 indicates a correct answer. is the candidate's skill data, d i is the difficulty data of the question, a i It is the question discrimination data, which is used to indicate the difference between question i and different candidate skill data. The degree of differentiation of the examinees, exp(·) is the natural exponential function;

[0108] The calculation formula for maximizing the log-likelihood function is:

[0109] ;

[0110] Where L is the log-likelihood function. The parameters of the item reflection theory model are optimized by maximizing the log-likelihood function. n is the total number of questions, i is the question index, m is the total number of examinees, j is the examinee index, P(·) is the probability that examinee j correctly answers question i, and R ij is the answer of candidate j on question i;

[0111] Step S23: Dynamic test question generation, specifically, using a reinforcement learning method, by constructing a deep Q model, and constructing state parameters, action parameters, reward parameters and strategy parameters, performing deep Q learning model training, and constructing a reinforcement learning question generation model;

[0112] In constructing the deep Q model, the state parameter s t , specifically including candidate skill data for the current candidate and the difficulty of the current test questions t , t is the assessment process index;

[0113] The action parameter a t , specifically refers to the difficulty of the next test question dt+1 Make a selection;

[0114] The reward parameter R t , specifically refers to the reward judgment of the current candidate's answer result. If the answer is correct, the reward parameter value is positive, and if the answer is wrong, the reward parameter value is negative;

[0115] The strategy parameters specifically refer to the current state parameters s t Select action parameter a t strategy;

[0116] The deep Q learning model training specifically refers to updating and training the strategy through the Q learning update rule, and the calculation formula is:

[0117] ;

[0118] In the formula, is the update operator, The Q(·) on the left is the updated Q value, which is used to represent the current state parameter s t Select action parameter a t The probability of The Q(·) on the right is the Q value before the update. is the learning rate parameter, which is used to control the update step size, R t is the reward parameter, is the discount factor used as the weight of the updated reward parameter, is the next state parameter s t+1 When you select Update Action Parameters The maximum probability value of the Q value;

[0119] Step S24: Training the assessment question generation model, specifically, by obtaining the optimization project reflection factor analysis model and the reinforcement learning question generation model, performing sub-model integration training to obtain the assessment question comprehensive generation model Model TG ;

[0120] Step S25: Generating test questions, specifically using the test questions to comprehensively generate a model Model TG , generate test questions, and optimize the test questions adaptively by building real-time adaptive feedback rules to obtain test question generation output data;

[0121] The real-time adaptive feedback rule specifically means that if the examinee continuously answers questions that are difficult and the difficulty analysis results show that they are difficult, the comprehensive generation model for test questions will increase the difficulty of the generated questions; if the examinee continuously answers the current questions incorrectly, the comprehensive generation model for test questions will reduce the difficulty of the generated questions.

[0122] By executing the above operations, in the generation of assessment questions in the existing electronic and electrical skills assessment optimization management method, there is a problem that the effectiveness of the electronic and electrical assessment depends on the effectiveness of the question setting, while the traditional manual setting is not adaptable and targeted enough, and the application of a single model to analyze electronic and electrical skills test questions cannot handle the complex assessment process and technical problems of high-difficulty test questions. This solution creatively adopts a reinforcement learning method combined with item reflection factor analysis, integrates multi-dimensional candidate characteristics by introducing molecular analysis, and optimizes the generation process of test questions through reinforcement learning and adaptive strategies, which provides a good intelligent solution inspiration for the entire assessment process and improves the targeting and adaptability of electronic and electrical assessments.

[0123] Example 4, see Figure 1 , Figure 2 and Figure 4 , this embodiment is based on the above embodiment. In step S3, the candidate allocation management is used to perform time management of specific assessments in combination with the results of assessment question generation during the actual assessment process, and further optimize the scheduling of candidates and the comprehensive assessment time arrangement. Specifically, the candidate allocation management is performed based on the candidate allocation management original data in the assessment management original data set, and a long short-term memory neural network combining ant colony optimization and reinforcement learning is used to perform candidate allocation management, and candidate allocation management is performed to obtain candidate allocation management data;

[0124] The candidate allocation management data specifically includes examination time management reference data and candidate scheduling reference data;

[0125] The long short-term memory neural network combining ant colony optimization and reinforcement learning specifically includes an ant colony optimization algorithm, a deep Q subnet and a long short-term memory subnet;

[0126] The ant colony optimization algorithm is used to globally search for the optimal allocation scheme for examinees and examination room resources;

[0127] The deep Q subnetwork is used to allocate the optimal candidate scheduling allocation plan in a complex examination environment;

[0128] The long short-term memory subnet is used to optimize the time management plan during the examination process by predicting the temporal changes of the examination time and examination room resources of the examinees;

[0129] The steps of using a long short-term memory neural network combining ant colony optimization and reinforcement learning to perform candidate allocation management, performing candidate allocation management, and obtaining candidate allocation management data include:

[0130] Step S31: optimizing the examination room resource allocation plan, specifically constructing an ant colony optimization algorithm, initializing the pheromone matrix, calculating the path selection probability and pheromone update, performing ant colony optimization modeling, and obtaining an examination room resource allocation optimization decision plan;

[0131] The ant colony optimization modeling specifically includes the following steps:

[0132] Step S311: Initialize the pheromone matrix, specifically initializing the connection pheromone matrix between the examinee and the examination room resources, and the calculation formula is:

[0133] ;

[0134] In the formula, is the pheromone matrix, It is to initialize the examination room resource data. is the examination room index, j is the examinee index;

[0135] Step S312: Calculate the path selection probability, specifically, use a single ant element in the ant colony algorithm to represent the examinee, and calculate the examination room path selection probability by simulating the process of the examinee selecting the examination room. The calculation formula is:

[0136] ;

[0137] In the formula, It is the candidate who chooses the examination room The probability of is the pheromone weighting matrix, is the pheromone weight, is the heuristic weighting factor, is the heuristic weight, which is used to represent the status information of the current examination room, K is the total number of candidate examination rooms, k is the index of the candidate examination room, is the pheromone weighted matrix for candidate j to choose alternative examination room k, is the heuristic weighting factor for candidate j to choose alternative examination center k;

[0138] Step S313: Pheromone update, specifically refers to updating the pheromone matrix according to the real-time data in the examination room, and the calculation formula is:

[0139] ;

[0140] In the formula, is the updated pheromone matrix, is the pheromone evaporation rate parameter, is the pheromone matrix before updating, is the pheromone increment, which is used to indicate the quality of the current examination room resource allocation scheme;

[0141] Step S32: optimizing the candidate scheduling allocation scheme, specifically constructing a deep Q subnet, using the reinforcement learning method in step S23, through state definition and action space definition, to construct a deep Q subnet, and obtain an optimized candidate scheduling allocation scheme;

[0142] The state definition and action space definition specifically refer to the definition of candidate allocation management state parameters, candidate allocation management action parameters and candidate allocation management reward parameters;

[0143] The candidate allocation management status parameters specifically include candidate information, examination room resource information and examination room resource allocation plan information;

[0144] The candidate allocation management action parameter specifically refers to selecting the examination room to which the candidate is allocated;

[0145] The candidate allocation management reward parameters specifically include examination room allocation perspective efficiency parameters and examination room utilization rate parameters;

[0146] Step S33: optimizing the assessment time management plan, specifically constructing a long short-term memory subnet, by constructing a standard long short-term memory neural network including an input gate, a forget gate, and an output gate as a long short-term memory subnet, and performing time series data prediction through the long short-term memory subnet to obtain an assessment time management optimization plan;

[0147] Step S34: Train the candidate allocation management model, specifically, train the candidate allocation management model through the ant colony optimization algorithm, the deep Q subnet and the long short-term memory subnet to obtain the candidate allocation management model Model DM ;

[0148] Step S35: candidate allocation management, specifically using the candidate allocation management model Model DM , perform candidate allocation management based on the candidate allocation management original data to obtain candidate allocation management data.

[0149] By executing the above operations, in the candidate allocation management of the existing electronic and electrical skills assessment optimization management method, there is a problem that the examination room resources for electronic and electrical assessment are relatively limited, and how to optimize the time and candidate allocation of the specific assessment process is a complex problem involving the temporal and spatial resources of candidates and examination rooms. The traditional intelligent method cannot take into account multiple factors, resulting in the technical problem that the intelligence of candidate allocation management cannot meet the system requirements. This solution creatively adopts a long short-term memory neural network that combines ant colony optimization and reinforcement learning to manage candidate allocation. The ant colony optimization algorithm is used to strengthen the collaborative optimal solution of candidate and examination room resource allocation, and reinforcement learning is used to make intelligent decisions on candidate allocation. Finally, the long short-term memory network is used to improve the timing prediction and processing capabilities, and the temporal and spatial management capabilities of examination room resource allocation are comprehensively improved, providing a guarantee for the high intelligence of the system.

[0150] Example 5, see Figure 1 and Figure 2This embodiment is based on the above embodiment. In step S4, the performance management optimization is used to record the electronic and electrical skills assessment results and perform performance clustering and change trend analysis. Specifically, the performance management optimization original data in the assessment management original data set is used to perform performance management optimization using a long short-term memory neural network combined with mean clustering to obtain performance management optimization data;

[0151] The said score management optimization specifically includes candidate cluster analysis reference data and candidate score trend prediction data;

[0152] The long short-term memory neural network combined with mean clustering specifically clusters the examinees' scores by using the K-means clustering method, and uses the standard long short-term memory neural network to analyze the trend of changes in the examinees' scores.

[0153] Example 6, see Figure 1 and Figure 2 , this embodiment is based on the above embodiment. In step S5, the skill assessment management is used to comprehensively optimize the electronic and electrician skill assessment management by combining assessment question generation, candidate allocation management and score management optimization. Specifically, before the start of the electronic and electrician assessment, the assessment question generation is applied to obtain the assessment question generation output data, and the question is adaptively adjusted to set the specific assessment question;

[0154] In the process of electronic and electrician assessment, by applying candidate allocation management, the candidate allocation management data is obtained, and specific examination room time management and candidate scheduling are carried out;

[0155] After the electronic and electrician assessment is completed, score management optimization data is obtained by applying score management optimization, and the candidate cluster analysis reference data and candidate score trend prediction data are used as the data source for adjusting the electronic and electrician assessment questions. Manual score management and analysis are then performed to obtain reference information for electronic and electrician assessment teaching and training.

[0156] Embodiment 7, see Figure 1 and Figure 2 , this embodiment is based on the above embodiment, the electronic and electrician skill assessment optimization management system based on artificial intelligence provided by the present invention includes a host computer control module, an evaluation and assessment module, an intelligent management module and a data transmission module;

[0157] The data transmission module is used for data transmission between modules and collection of original data. Through the collection of original data, an original data set for assessment management is obtained, and the original data set for assessment management is sent to the intelligent management module through data transmission;

[0158] The intelligent management module is used to apply assessment question generation, candidate allocation management and score management optimization. Through the assessment question generation, candidate allocation management and score management optimization, assessment question generation output data, candidate allocation management data, score management optimization data and electronic and electrician assessment teaching and training reference information are obtained, and the assessment question generation output data is sent to the upper computer control module through the data transmission module, and the assessment result data is received from the evaluation and assessment module and used for intelligent management, and the electronic and electrician assessment teaching and training reference information is used as the comprehensive analysis result of intelligent management;

[0159] The host computer control module is used to set and implement specific electronic and electrical skills assessment;

[0160] The evaluation and assessment module is used to construct an embedded assessment device and perform specific assessment tasks.

[0161] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0162] While the embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that many changes, modifications, substitutions and variations can be made to the embodiments without departing from the principles and spirit of the invention.

[0163] The present invention and its embodiments are described above, and such description is not restrictive. The drawings show only one embodiment of the present invention, and the actual structure is not limited thereto. In short, if ordinary technicians in the field are inspired by it, without departing from the purpose of the invention, they can design a structure and embodiment similar to the technical solution without creativity, which should belong to the protection scope of the present invention.

Claims

1. An optimization management method for electronic and electrician skill assessment based on artificial intelligence, characterized by: The method comprises the following steps: Step S1: system structure design, specifically building a host computer control module, an evaluation and assessment module, an intelligent management module and a data transmission module; Step S2: Assessment question generation, which is used to generate questions and adaptively adjust the difficulty of the questions in combination with the personalized skill level of the School of Electronic and Electrical Engineering. Specifically, the original data is generated based on the assessment questions in the assessment management original data set, and the assessment questions are generated using a reinforcement learning method combined with item reflection factor analysis to obtain assessment question generation output data. The steps include: data preparation, question difficulty analysis, dynamic question generation, assessment question generation model training, and assessment question generation; Step S3: candidate allocation management, which is used to perform specific assessment time management in combination with the results of assessment question generation during the actual assessment process, and further optimize the scheduling of candidates and the comprehensive assessment time arrangement, specifically, based on the candidate allocation management original data in the assessment management original data set, a long short-term memory neural network combining ant colony optimization and reinforcement learning is used to perform candidate allocation management to obtain candidate allocation management data; Step S4: optimizing performance management, using a long short-term memory neural network combined with mean clustering to optimize performance management, and obtaining performance management optimization data; Step S5: Skill assessment management, which is used to comprehensively optimize the electronic and electrician skill assessment management by combining assessment question generation, candidate allocation management and score management optimization.

2. The method for optimizing and managing electronic and electrician skills assessment based on artificial intelligence according to claim 1 is characterized in that: In step S1, the system structure design is used to construct the basic structure of the electronic and electrician skill assessment optimization management system, specifically to construct a host computer control module, an evaluation and assessment module, an intelligent management module and a data transmission module; The host computer control module is used to set the assessment class, set the assessment test content, assign the assessment number, record and store the assessment score and output the assessment result, which is specifically implemented by the host computer control software; The evaluation and assessment module is used to construct an embedded assessment device, which specifically includes a single-chip microcomputer, a serial port LCD screen, a measured component output circuit, and a measured waveform output circuit; The intelligent management module specifically includes an examination question generation unit, a candidate allocation management unit and a score management optimization unit; The data transmission module is used to collect original data and realize data transmission between functional modules, specifically, to obtain the original data set of assessment management by collecting historical electronic and electrician skill assessment records and basic information of examinees; The assessment management original data set specifically includes assessment question generation original data, examinee allocation management original data and score management optimization original data.

3. The method for optimizing and managing electronic and electrician skills assessment based on artificial intelligence according to claim 2 is characterized in that: In step S2, the test questions generate output data, specifically personalized electronic and electrician test question data; The step of using the reinforcement learning method combined with item reflection factor analysis to generate test questions and obtain test question generation output data includes: Step S21: data preparation, specifically generating original data according to the test questions, and obtaining test question generation optimization data through data extraction and manual feature extraction, specifically including question data, examinee data and question label data; Step S22: Test question difficulty analysis, specifically, by constructing a project response theoretical model based on factor analysis, performing a dual-parameter test question difficulty analysis, obtaining a reference probability of correct answer rate of the test question, and optimizing the parameters of the project response theoretical model by constructing a maximized log-likelihood function, thereby constructing an optimized project response factor analysis model; Step S23: Dynamic test question generation, specifically, using a reinforcement learning method, by constructing a deep Q model, and constructing state parameters, action parameters, reward parameters and strategy parameters, performing deep Q learning model training, and constructing a reinforcement learning question generation model; Step S24: Training the assessment question generation model, specifically, by obtaining the optimization project reflection factor analysis model and the reinforcement learning question generation model, performing sub-model integration training to obtain the assessment question comprehensive generation model Model TG ; Step S25: Generating test questions, specifically using the test questions to comprehensively generate a model Model TG , generate examination questions, and adaptively optimize the examination questions by building real-time adaptive feedback rules to obtain the output data for examination question generation.

4. The method for optimizing and managing electronic and electrician skills assessment based on artificial intelligence according to claim 3 is characterized in that: In step S3, the candidate allocation management data specifically includes examination time management reference data and candidate scheduling reference data; The long short-term memory neural network combining ant colony optimization and reinforcement learning specifically includes an ant colony optimization algorithm, a deep Q subnet and a long short-term memory subnet; The ant colony optimization algorithm is used to globally search for the optimal allocation scheme for examinees and examination room resources; The deep Q subnetwork is used to allocate the optimal candidate scheduling allocation plan in a complex examination environment; The long short-term memory subnet is used to optimize the time management plan during the examination process by predicting the temporal changes of the examinee's examination time and examination room resources.

5. The method for optimizing and managing electronic and electrician skills assessment based on artificial intelligence according to claim 4 is characterized in that: In step S3, the step of using a long short-term memory neural network combining ant colony optimization and reinforcement learning to perform candidate allocation management and obtain candidate allocation management data includes: Step S31: optimizing the examination room resource allocation scheme, specifically constructing an ant colony optimization algorithm, initializing the pheromone matrix, calculating the path selection probability and pheromone update, performing ant colony optimization modeling, and obtaining an examination room resource allocation optimization decision-making scheme; Step S32: optimizing the candidate scheduling allocation scheme, specifically constructing a deep Q subnet, using the reinforcement learning method in step S23, through state definition and action space definition, to construct a deep Q subnet, and obtain an optimized candidate scheduling allocation scheme; Step S33: optimizing the assessment time management plan, specifically constructing a long short-term memory subnet, by constructing a standard long short-term memory neural network including an input gate, a forget gate, and an output gate as a long short-term memory subnet, and performing time series data prediction through the long short-term memory subnet to obtain an assessment time management optimization plan; Step S34: Train the candidate allocation management model, specifically, train the candidate allocation management model through the ant colony optimization algorithm, the deep Q subnet and the long short-term memory subnet to obtain the candidate allocation management model Model DM ; Step S35: candidate allocation management, specifically using the candidate allocation management model Model DM , perform candidate allocation management based on the candidate allocation management original data to obtain candidate allocation management data.

6. The method for optimizing and managing electronic and electrical skills assessment based on artificial intelligence according to claim 5 is characterized in that: In step S4, the performance management optimization is used to record the electronic and electrical skills assessment results and perform performance clustering and change trend analysis, specifically, based on the performance management optimization original data in the assessment management original data set, the performance management optimization is performed using a long short-term memory neural network combined with mean clustering to obtain performance management optimization data; The said score management optimization specifically includes candidate cluster analysis reference data and candidate score trend prediction data; The long short-term memory neural network combined with mean clustering specifically clusters the examinees' scores by using the K-means clustering method, and uses the standard long short-term memory neural network to analyze the trend of changes in the examinees' scores.

7. The method for optimizing and managing electronic and electrician skill assessment based on artificial intelligence according to claim 6 is characterized in that: In step S5, the skill assessment management is used to comprehensively optimize the electronic and electrical skill assessment management by combining assessment question generation, candidate allocation management and score management optimization. Specifically, before the start of the electronic and electrical assessment, the assessment question generation is applied to obtain the assessment question generation output data, and the questions are adaptively adjusted to set specific assessment questions; In the process of electronic and electrician assessment, by applying candidate allocation management, the candidate allocation management data is obtained, and specific examination room time management and candidate scheduling are carried out; After the electronic and electrician assessment is completed, score management optimization data is obtained by applying score management optimization, and the candidate cluster analysis reference data and candidate score trend prediction data are used as the data source for adjusting the electronic and electrician assessment questions. Manual score management and analysis are then performed to obtain reference information for electronic and electrician assessment teaching and training.

8. An electronic and electrician skill assessment optimization management system based on artificial intelligence, used to implement the electronic and electrician skill assessment optimization management method based on artificial intelligence as described in any one of claims 1 to 7, characterized in that: It includes a host computer control module, an evaluation and assessment module, an intelligent management module and a data transmission module.

9. The electronic and electrician skill assessment optimization management system based on artificial intelligence according to claim 8 is characterized in that: The data transmission module is used for data transmission between modules and collection of original data. Through the collection of original data, an original data set for assessment management is obtained, and the original data set for assessment management is sent to the intelligent management module through data transmission; The intelligent management module is used to apply assessment question generation, candidate allocation management and score management optimization. Through the assessment question generation, candidate allocation management and score management optimization, assessment question generation output data, candidate allocation management data, score management optimization data and electronic and electrician assessment teaching and training reference information are obtained, and the assessment question generation output data is sent to the upper computer control module through the data transmission module, and the assessment result data is received from the evaluation and assessment module and used for intelligent management, and the electronic and electrician assessment teaching and training reference information is used as the comprehensive analysis result of intelligent management; The host computer control module is used to set and implement specific electronic and electrical skills assessment; The evaluation and assessment module is used to construct an embedded assessment device and perform specific assessment tasks.

Images