A smart examination system

Abstract

The present invention discloses a smart examination system, which belongs to the field of smart examination technology, and mainly solves the problems of heavy tasks, low efficiency, high difficulty and low flexibility in existing examination organization. The system includes a management end and an examination end; the management end includes a question bank management module, a model management module, an examination management module, an examination situation analysis module, a score management module and a system management module, and the examination end includes an online examination module and a personal center module. Through the above design, the present invention uses an automated process, and the smart examination system can quickly generate test papers, collect answers and review and score test papers, reducing manpower requirements and time consumption, and improving overall examination efficiency.

Description

Technical Field

[0001] The present invention relates to the technical field of smart examination systems, and in particular to a smart examination system. Background Art

[0002] With the expansion and deepening of teaching knowledge points, if you want to analyze the candidates' scores and the degree of knowledge mastery by organizing exams, you need to extract from a large number of questions, and complete the examination review and teaching analysis at the same time. There are problems such as heavy tasks, low efficiency, high difficulty, and low flexibility. To solve these problems, providing an examination platform with a full process from question type matching, intelligent test paper composition, intelligent review, score summary, teaching analysis, etc. will fundamentally reduce manpower requirements and time consumption.

[0003] The existing examination system also has problems such as time-consuming and laborious manual composition and grading, and some factors in the grading process lead to the defect of inconsistent scoring standards. Therefore, the smart examination system improves the efficiency and fairness of the examination through its intelligent composition, intelligent grading, assessment analysis, anti-cheating verification and other functions, while providing strong technical support for the personalization and quality improvement of education. Summary of the invention

[0004] The purpose of the present invention is to provide a smart examination system, which mainly solves the problems of heavy workload, low efficiency, great difficulty and low flexibility in existing examination organization.

[0005] To achieve the above purpose, the technical solution adopted by the present invention is as follows:

[0006] A smart examination system includes a management end and an examination end; the management end includes:

[0007] The question bank management module is used to store, manage and maintain a large number of test question resources, and classify and mark the test questions according to subjects, difficulty and question types;

[0008] Model management module, used to manage and maintain intelligent test-taking models, intelligent scoring models, and competency assessment models, and to generate test questions, intelligent and real-time scoring, and competency assessment;

[0009] The examination management module is used to generate examination activities, provide online marking and examination archiving;

[0010] The examination situation analysis module is used by invigilators to monitor the examination in real time and view the overall situation of the examination;

[0011] The score management module is used to review and submit the test scores, and to rank and count the final scores;

[0012] System management module, used to maintain and manage information on users, menus, roles, organizations, and data dictionaries, and to verify user access permissions;

[0013] The test terminal includes:

[0014] Online exam module, used to verify the identity of candidates and provide cheating detection methods;

[0015] The personal center module is used to display personal basic information, test records, and wrong question collection.

[0016] Furthermore, in the present invention, the configuration method of the question bank management module is as follows:

[0017] Configure the new module of question bank directory to distinguish the subjects of the test questions and maintain the organizational structure of the question bank;

[0018] Configure the question type management module to manage and assign objective questions and subjective questions;

[0019] Configure the question type import module to fill in the question information in the question template according to the template requirements;

[0020] Configure the data deduplication module to perform duplication verification on the imported test question information after the test question verification work is completed. If there are duplicate questions, they will be marked and modified. After the modification is completed, if there are no duplicate questions, the data will be saved to the question bank for maintenance;

[0021] Configure the question bank maintenance module to display the test questions in different levels according to the question bank directory;

[0022] Configure the question bank review module to complete the import of test question data and put the test question data into the database after being reviewed by the administrator after the test question data is imported and is in the pending review status.

[0023] Furthermore, in the present invention, the configuration method of the model management module is as follows:

[0024] Configure the intelligent test paper generation module, which is used to generate the test paper by filling in the basic configuration information and test paper generation rules and calculating the intelligent test paper generation algorithm; the implementation formula of the intelligent test paper generation algorithm is:

[0025]

[0026] Among them, time() is the total time of intelligently extracting questions, score() is the total score of intelligently extracting questions, the total time of answering questions should not be greater than the prescribed examination time T, and the total score should be equal to the prescribed test score S. represents the set of extracted test questions, where n is the number of extracted test questions, q is the set of knowledge points covered by the questions, Q is the set of all test questions, D is the target test difficulty, ε is the tolerance for difficulty matching, k is the knowledge point involved in the current question, and K is the set of all knowledge points in the question bank; d(q i ) represents the estimated difficulty set when the question is created, k∈p(q i ) indicates that question q in the current test paper p has been generated i Knowledge point k is already included;

[0027] Configure the test paper analysis module to analyze the test paper generated by the intelligent test paper generation algorithm from the perspective of test question difficulty, question type, number of question types, number of questions, assessment direction, and question difficulty range; if the analysis result shows that the generated test paper does not meet the current examination requirements, use the intelligent test paper generation module to regenerate the test paper;

[0028] Configure the test question replacement module, which is used for the administrator to check all test questions and manually select and replace test questions after the test paper analysis is qualified; and for the administrator to replace wrong test questions during the test;

[0029] Configure the intelligent scoring module to obtain the test data in the test paper, and realize intelligent calculation and scoring by setting data calibration and feature extraction in combination with feature weights; data calibration means determining a benchmark score through manual scoring, and the benchmark score will be used as the standard for learning of the intelligent scoring system; feature extraction means setting different feature extraction keywords for different types of test questions for scoring calculation; feature weight means setting different feature weights according to different feature extraction keywords, which are used for score calculation by the intelligent scoring algorithm; the implementation formula of the intelligent scoring algorithm is as follows:

[0030]

[0031] Where n represents the total number of features in the feature set, w i is the feature f i The weight of is the weight normalization value, the sum is 1, match(f i ,A) is the feature f i The degree of match with the candidate's answer A, with a value between 0 and 1, indicating the accuracy of the match, S i Indicates the current feature f i The benchmark score under the test results is set; the score difference quality inspection module is configured to perform quality inspection on the intelligent scoring results; the difference between the intelligent scoring and the manual scoring is compared to ensure the accuracy and consistency of the scoring. If the intelligent scoring results meet the requirements, the scoring model is released; if not, the feature weights can be modified; among them, the manual scoring requires the question setter to provide correct and complete answers to the test questions;

[0032] Configure the visualization report module to generate data from the test difficulty, test time, test question type, and test knowledge point direction after the ability assessment algorithm calculates it, uses visualization reports for display, and provides report export function; the expression of the ability assessment algorithm is:

[0033]

[0034] Among them, α, β, γ are weight coefficients used to adjust the importance of different parameters in the evaluation results, and α+β+γ=1, μ represents the average score of this test, Indicates the range of candidates' scores for the current knowledge point. Indicates the ranking of the candidates' scores on the current knowledge point.

[0035] Furthermore, in the present invention, the configuration method of the test management module is as follows:

[0036] Configure the examination information module to access the test data generated by the model management module, fill in the basic information of the test paper, and publish the test paper after reviewing it, and determine the information of the participating candidates;

[0037] Configure the cheating detection module to turn on or off cheating detection. When the detection is turned on, the system will obtain the test client information and cheating check information in real time;

[0038] Configure the marking module to set marking rules and assign examiners. Exam examiners can manually review or revise the intelligent scoring results. After reading the marking rules, online marking is implemented. If it is configured as intelligent marking, the intelligent scoring results can be manually revised. If it is configured as manual marking, each question needs to be scored.

[0039] Configure the score archiving module to report scores and archive the exams after the exam review is completed; exam archiving includes manual archiving and automatic archiving.

[0040] Furthermore, in the present invention, the configuration method of the examination situation analysis module is as follows:

[0041] Configure the candidate information module to display and verify the candidate information in real time, and display the current test taker information and the real-time screen of the candidate;

[0042] Configure the candidate statistics module to count candidates who have not logged into the system before the exam deadline. The system will automatically generate absence statistics;

[0043] Configure the violation statistics module to count the cheating detection data sent by the online examination module. When there is a violation, choose to send a message manually or automatically to remind the candidate. The system records the number of violations of the candidate. If the number exceeds the set number, the system will force the candidate to log off.

[0044] Configure the dynamic ranking module for candidates to sort them according to their current real-time scores and generate a dynamic ranking list.

[0045] Furthermore, in the present invention, the configuration method of the performance management module is:

[0046] Configure the information access module to access the examinee scores, exam information, and test paper information in the exam management module, generate the final scores and rankings after the scores are reviewed, and provide score data to the personal center module;

[0047] Configure the score list module to display the candidate's name, exam name, exam start and end time, exam duration, assessment results, and review status; and display the ranked list by setting ranking rules.

[0048] Furthermore, in the present invention, the configuration method of the system management module is as follows:

[0049] Configure the organization management module to display organization information in a tree structure hierarchy and support adding and importing organization data;

[0050] Configure the menu management module to display menu name, code, component path, menu type, menu address, and menu status information;

[0051] Configure the role management module to display role name, role code, role status information, and configure role menu access rights;

[0052] Configure the user management module to display user account, user name, user role, and creation time information, and support modification of user information and password.

[0053] Furthermore, in the present invention, the configuration method of the online examination module is as follows:

[0054] Configure the candidate identity authentication module to verify the identity information recorded in the access system management module. Perform identity verification when logging into the examination system. If the identity verification passes, the candidate can enter the examination, otherwise the system will automatically exit the examination.

[0055] Configure the exam instructions loading module to load the exam instructions from the exam management module and show the content, scope, and requirements of the exam to the candidates;

[0056] Configuration detection loading module is used to access the cheating detection configuration in the test management, and conduct real-time detection and verification from the screen switching detection, screen domination detection, split screen detection, screenshot detection, video detection, and pasteboard detection methods. If any illegal operation occurs, the number of violations will be recorded, and the violation information will be submitted to the management end so that the invigilator can remind and record it;

[0057] Configure the test question loading module to display the accessed test question data in order, distinguish between answered questions and unanswered questions, so that candidates can check for omissions and fill in the gaps;

[0058] Configure the wrong test question reporting module to allow candidates to report wrong test questions found during the exam, and the invigilator will record the data.

[0059] Furthermore, in the present invention, the configuration method of the personal center module is as follows:

[0060] Configure the exam record module to display all the exam information that the user has taken;

[0061] Configure the wrong question center module to display and count all wrong questions in the exams you have taken, and provide the function of redoing wrong questions;

[0062] Configure the personal information module to display basic user information, count the total number of exams taken, and the total number of wrong questions.

[0063] Compared with the prior art, the present invention has the following beneficial effects:

[0064] (1) Through the automated process, the intelligent examination system of the present invention can quickly generate examination papers, collect answers, and review and score examination papers, thereby reducing manpower requirements and time consumption and improving the overall examination efficiency. The system uses a variety of technical means, such as intelligent test paper composition, random question options, online monitoring and identity verification, to reduce the possibility of cheating and ensure the fairness of the examination.

[0065] (2) The intelligent examination system of the present invention can collect and analyze a large amount of examination data, providing data support for teaching improvement and examinee learning improvement. It also supports a variety of question types and examination formats, including objective questions, subjective questions, etc., to meet the needs of different subjects and ability assessments.

[0066] (3) The technologies used in the software development of the present invention all adopt free open source technologies, such as Vue, JavaScript, MySql, JAVA, etc., with low development costs; the PC software is developed using Vue, which is a JavaScript framework with good operating performance and cross-platform capabilities, low requirements for the hardware environment of the deployment project, and convenient project transplantation. BRIEF DESCRIPTION OF THE DRAWINGS

[0067] Figure 1 It is a schematic diagram of the block diagram structure of the present invention. DETAILED DESCRIPTION

[0068] The present invention is further described below in conjunction with the accompanying drawings and embodiments. The embodiments of the present invention include but are not limited to the following embodiments.

[0069] The present invention discloses a smart examination system, which realizes automatic and visual report generation of examination papers by configuring a question bank, creating an intelligent test paper composition model, an intelligent scoring model, an ability assessment model and an algorithm, and supports real-time examination situation analysis, displays information such as candidates' real-time scores and rankings, and detects cheating in real time during examinations.

[0070] The smart examination system consists of two parts: the management end and the examination end. The management end includes six modules: question bank management, model management, system management, examination management, examination situation analysis, and score management. The examination end includes two modules: online examination and personal center. Figure 1 shown.

[0071] The smart examination system management terminal is used to manage authorized users, add and configure question banks, manage intelligent test paper models, intelligent scoring models, and ability assessment models, as well as test management and score management, and display and analyze the test situation.

[0072] The smart exam management terminal is mainly composed of six parts, namely question bank management, model management, exam management, exam situation analysis, score management, and system management.

[0073] In this embodiment, the question bank management module is the core component of the system, which is used to store, manage and maintain a large number of test question resources, allowing test questions to be classified and labeled according to subjects, difficulty, question types and other standards to ensure that the organizational structure of the question bank is clear and orderly. Its main configuration modules are as follows:

[0074] A new module for question bank directory is added to distinguish test subjects and maintain the organizational structure of the question bank, including directory name, notes, activation status and other information;

[0075] The question type management module is used to manage and allocate two major question types, namely objective questions and subjective questions, as well as the seven question types under the two major types, such as single-choice questions, multiple-choice questions, true-or-false questions, fill-in-the-blank questions, short-answer questions, essay questions, case analysis questions, etc.

[0076] In the question type import module, fill in the question information in the question template according to the template requirements, such as question type, question content, number of options, etc., and then import it into the system.

[0077] The data deduplication module is used to perform duplication verification on the imported test question information after the test question verification work is completed. If there are duplicate questions, they will be marked and modified. After the modification is completed, if there are no duplicate questions, the data will be saved to the question bank for maintenance. During the test question verification, the system will perform data verification based on the filled-in content information and mark the data that failed the verification to ensure the correctness and completeness of the test question data.

[0078] The question bank maintenance module is used to display the test questions in a hierarchical manner according to the question bank directory, including question type, question content, question path, sponsor, question classification, status and other information.

[0079] The question bank review module is used to put the test questions into the database after the test question data is imported and the test question data is in the pending review status and is reviewed by the administrator.

[0080] In this embodiment, the model management module is used to manage and maintain the intelligent test model, intelligent scoring model, ability assessment model and other contents, and realize the functions of generating test questions, intelligent and real-time scoring, ability assessment and so on. The main configuration modules of the model management module are as follows:

[0081] The intelligent test paper generation module is used to generate test papers by filling in basic configuration information and test paper generation rules and calculating through the intelligent test paper generation algorithm; the implementation formula of the intelligent test paper generation algorithm is:

[0082]

[0083] Among them, time() is the total time of intelligently extracting questions, score() is the total score of intelligently extracting questions, the total time of answering questions should not be greater than the prescribed examination time T, and the total score should be equal to the prescribed test score S. represents the set of extracted test questions, where n is the number of extracted test questions, q is the set of knowledge points covered by the questions, Q is the set of all test questions, D is the target test difficulty, ε is the tolerance for difficulty matching, k is the knowledge point involved in the current question, and K is the set of all knowledge points in the question bank; d(q i ) represents the estimated difficulty set when the question is created, k∈p(q i ) indicates that question q in the current test paper p has been generated i The knowledge point k has been included. Among them, the test configuration rules mainly include configuration items such as test difficulty, test time, test question type, test knowledge points, number of test questions, and test score. By configuring the above parameters, basic parameters are provided for the intelligent test algorithm to extract test questions from the question bank.

[0084] The test paper analysis module is used to analyze the test paper generated by the intelligent test paper generation algorithm in terms of test difficulty, question type, number of question types, number of questions, assessment direction, and question difficulty range; if the analysis result shows that the generated test paper does not meet the current examination requirements, the intelligent test paper generation module is used to regenerate the test paper.

[0085] The test question replacement module is used for the administrator to review all test questions and manually select and replace test questions after the test paper analysis is qualified; and for the administrator to replace wrong test questions when there are wrong test questions during the examination.

[0086] The intelligent scoring module is used to obtain the test data in the test paper, and realize intelligent calculation and scoring through the set data calibration and feature extraction, combined with feature weights; among them, data calibration means to determine a benchmark score through manual scoring, and this benchmark score will be used as the standard for the intelligent scoring system to learn; feature extraction means to set different feature extraction keywords for different types of test questions for scoring calculation; feature weight means to set different feature weights according to the set different feature extraction keywords, which are used for score calculation by the intelligent scoring algorithm. Among them, the implementation formula of the intelligent scoring algorithm is:

[0087]

[0088] Where n represents the total number of features in the feature set, w i is the feature f i The weight of is the weight normalization value, the sum is 1, match(f i ,A) is the feature f i The degree of match with the candidate's answer A, whose value is usually between 0 and 1, indicates the accuracy of the match. i Indicates the current feature f i The following benchmark scores.

[0089] The score difference quality inspection module is used to perform quality checks on the intelligent scoring results; compare the differences between intelligent scoring and manual scoring to ensure the accuracy and consistency of scoring. If the intelligent scoring results are consistent, the scoring model will be published; if they are inconsistent with the results, the feature weights can be modified; among them, manual scoring requires the question setter to provide correct and complete answers to the test questions.

[0090] The visualization report module is used to generate data from the test difficulty, test time, test question types, and assessment knowledge points after calculation by the ability assessment algorithm, and display it using a visualization report, and provide a report export function; among them, the expression of the ability assessment algorithm is.

[0091]

[0092] Among them, α, β, γ are weight coefficients used to adjust the importance of different parameters in the evaluation results, and α+β+γ=1, μ represents the average score of this test, Indicates the range of candidates' scores for the current knowledge point. Indicates the ranking of the candidates' scores on the current knowledge point.

[0093] In this embodiment, the test management module mainly includes test paper composition, online marking, test archiving and other modules, which are used to generate test activities, provide online marking and test archiving capabilities, and provide data for the score management module and the personal center module. The main configuration modules of the test management module are as follows:

[0094] The exam information module is used to access the test data generated by the model management module and fill in the basic information of the test paper, such as the activity name, sponsor, test paper cover, test time, archiving settings, score disclosure settings, test instructions, preface, etc. The test paper will be released after review and the information of participating candidates will be determined;

[0095] The cheating detection module is used to turn on or off cheating detection. When the detection is turned on, the system will obtain the test client information and cheating inspection information in real time.

[0096] The marking module is used to set marking rules and assign examiners. Exam examiners can conduct manual review or revise the intelligent scoring results. After reading the marking rules, online marking is implemented. If it is configured for intelligent marking, the intelligent scoring results can be revised manually. If it is configured for manual marking, each question needs to be scored.

[0097] The score archiving module is used to report scores and archive exams after the exam review is completed. After all the exam papers are reviewed, the exam can be archived, and the name of the exam activity, the organizer, the release time, the number of candidates, the number of candidates who have taken the exam, the number of candidates who have not taken the exam, the reference rate and other information will be displayed. It also supports the export of archive reports. Exam archiving includes manual archiving and automatic archiving. Manual archiving means manually archiving at the exam archive after the exam is completed. If the exam is configured for automatic archiving and the last archiving time is set, the system will make a judgment based on the time and realize automatic archiving.

[0098] In this embodiment, the test situation analysis module is mainly used for invigilators to monitor the test in real time and view the overall test situation. To implement the test situation analysis, the module accesses the candidate data and test paper information from the online test module and the test management module to perform cheating detection and real-time score display. The specific configuration modules of the test situation analysis module are as follows:

[0099] The candidate information module is used to display and verify candidate information in real time, display current test taker information and real-time images of candidates; by verifying the identity of candidates, it prevents cheating such as taking exams on behalf of others and ensures the fairness of the exam.

[0100] The candidate statistics module is used to count the candidates who have not logged into the system before the exam deadline. The system will automatically generate absence statistics.

[0101] The violation statistics module is used to count the cheating detection data sent by the online examination module when the system is connected. When there is an illegal operation, you can choose to send a message manually or automatically to remind the candidate. The system records the number of violations of the candidate. If the number exceeds the set number, the system will force the candidate to log off.

[0102] The dynamic ranking module for candidates is used to sort candidates according to their current real-time scores and generate a dynamic ranking list.

[0103] In this embodiment, the score management module is used to review and submit the test scores, and to rank and count the final scores, and mainly includes modules such as score publishing and ranking and counting. The specific configuration modules of the score management module are as follows:

[0104] The information access module is used to access the candidate scores, test information, and test paper information in the test management module, generate the final scores and score rankings after score review, and provide score data to the personal center module.

[0105] The score list module is used to display the candidate's name, test name, test start and end time, test time, assessment results, and review status; the ranking list is displayed by setting the ranking rules. This module can realize batch review in the review status, check the scores of multiple candidates and submit them for review to improve the review efficiency. It can also realize single review, check and review the candidate's scores one by one to ensure the correctness of the scores. The ranking rules calculate the process scores by combining the test time with the system, and calculate the average score based on the test scores, and sort the candidates according to the average score. The ranking list displays information such as serial number, name, institution, gender, test score, process score, total score, average score, etc.

[0106] In this embodiment, the system management module is used to maintain and manage information such as users, menus, roles, and organizations, and to perform permission verification on user access to ensure system security. The specific module configuration of the system management module is as follows:

[0107] The organization management module is used to display organization information in a tree-like structure and supports adding and importing organization data;

[0108] Menu management module, used to display menu name, code, component path, menu type, menu address, and menu status information;

[0109] The role management module is used to display the role name, role code, role status information, and configure the role menu access rights;

[0110] The user management module is used to display user account, user name, user role, and creation time information, and supports modification of user information and password.

[0111] The test end of the smart test system consists of two modules, namely the online test module and the personal center module. The online test module is used to verify the identity of the examinee and provide a series of cheating detection methods to ensure the fairness of the test.

[0112] In this embodiment, the configuration method of the online examination module is as follows:

[0113] Candidate identity verification module, used to verify the identity information recorded by the access system management module, and perform identity verification when logging into the examination system. If the identity verification is passed, the candidate can enter the examination, otherwise the system will automatically exit the examination;

[0114] The exam instructions loading module is used to load the exam instructions from the exam management module and show the content, scope and requirements of the exam to the candidates.

[0115] The detection loading module is used to access the cheating detection configuration in the test management, and conduct real-time detection and verification from the screen switching detection, screen domination detection, split screen detection, screenshot detection, video detection, and pasteboard detection methods. If any illegal operation occurs, the number of violations will be recorded, and the violation information will be submitted to the management end so that the invigilator can remind and record it;

[0116] The test question loading module is used to display the received test question data in order, distinguishing between answered questions and unanswered questions, so as to facilitate the candidates to check for omissions and fill in the gaps;

[0117] The wrong question reporting module is used for candidates to report wrong questions found during the examination, and the invigilator records the data.

[0118] In this embodiment, the personal center module is used to display personal basic information, test records, wrong question center, etc. The specific configuration functions of the personal center module are as follows:

[0119] The exam record module is used to display all the information about the exams that the user has taken, such as the name of the exam activity, exam time, exam duration, process scores, assessment scores, average scores, etc. It also provides a function for redoing wrong questions.

[0120] Configure the wrong question center module to display and count all the wrong questions in the exams you have taken, and provide a wrong question redo function; used to strengthen the grasp of knowledge points.

[0121] The personal information module is used to display basic user information, count the total number of exams taken, and the total number of wrong questions.

[0122] The technologies used in the software development of the present invention all adopt free open source technologies, such as Vue, JavaScript, MySql, JAVA, etc., with low development cost; the PC software is developed using Vue, which is a JavaScript framework with good operating performance and cross-platform performance, low requirements for the hardware environment of the deployment project, and convenient project transplantation.

[0123] Through the above design, the present invention uses an automated process, and the smart examination system can quickly generate examination papers, collect answers, and review and score examination papers, thereby reducing manpower requirements and time consumption and improving overall examination efficiency.

[0124] The above embodiment is only one of the preferred implementation modes of the present invention and should not be used to limit the protection scope of the present invention. Any changes or modifications that are made to the main design concept and spirit of the present invention and have no substantive significance, and the technical problems they solve are still consistent with the present invention, should be included in the protection scope of the present invention.

Claims

1. A smart examination system, characterized in that: It includes a management end and an examination end; the management end includes: The question bank management module is used to store, manage and maintain a large number of test question resources, and classify and mark the test questions according to subjects, difficulty and question types; The model management module is used to manage and maintain the intelligent test-taking model, the intelligent scoring model, and the ability assessment model, and to generate test questions, intelligent and real-time scoring, and ability assessment. The configuration method of the model management module is as follows: Configure the intelligent test paper generation module, which is used to generate test papers by filling in basic configuration information and test paper generation rules and calculating through the intelligent test paper generation algorithm; the implementation formula of the intelligent test paper generation algorithm is: Among them, time() is the total time of intelligently extracting questions, score() is the total score of intelligently extracting questions, the total time of answering questions should not be greater than the prescribed examination time T, and the total score should be equal to the prescribed test score S. represents the set of extracted test questions, where n is the number of extracted test questions, q is the set of knowledge points covered by the questions, Q is the set of all test questions, and D is the target test difficulty. Indicates the tolerance for difficulty matching, k indicates the knowledge points involved in the current question, and K indicates the set of all knowledge points in the question bank; Indicates the estimated difficulty set when the question is created. Indicates that the questions in the current test paper p have been generated Knowledge point k is already included; Configure the test paper analysis module to analyze the test paper generated by the intelligent test paper generation algorithm from the perspective of test question difficulty, question type, number of question types, number of questions, assessment direction, and question difficulty range; if the analysis result shows that the generated test paper does not meet the current examination requirements, use the intelligent test paper generation module to regenerate the test paper; Configure the test question replacement module, which is used for the administrator to check all test questions and manually select and replace test questions after the test paper analysis is qualified; and for the administrator to replace wrong test questions during the test; Configure the intelligent scoring module to obtain the test data in the test paper, and realize intelligent calculation and scoring by setting data calibration and feature extraction in combination with feature weights; data calibration means determining a benchmark score through manual scoring, and the benchmark score will be used as the standard for learning of the intelligent scoring system; feature extraction means setting different feature extraction keywords for different types of test questions for scoring calculation; feature weight means setting different feature weights according to different feature extraction keywords, which are used for score calculation by the intelligent scoring algorithm; the implementation formula of the intelligent scoring algorithm is as follows: Where N represents the total number of features in the feature set, It is a feature The weight of is the weight normalization value, the sum is 1, It is a feature The degree of match with the candidate's answer A, with a value between 0 and 1, indicates the accuracy of the match. Indicates the current feature Benchmark scores below; Configure the score difference quality inspection module to perform quality inspection on the intelligent scoring results; compare the differences between intelligent scoring and manual scoring to ensure the accuracy and consistency of scoring. If the intelligent scoring results meet the requirements, the scoring model will be released; if not, the feature weights can be modified; for manual scoring, the question setter provides correct and complete answers to the test questions; Configure the visualization report module to generate data from the test difficulty, test time, test question type, and test knowledge point direction after the ability assessment algorithm calculates it, uses visualization reports for display, and provides report export function; the expression of the ability assessment algorithm is: in α , β , γ is the weight coefficient, which is used to adjust the importance of different parameters in the evaluation results, and α+β+γ=1, Indicates the average score of this test. Indicates the range of candidates' scores for the current knowledge point. Indicates the ranking of the examinee's scores for the current knowledge point; The examination management module is used to generate examination activities, provide online marking and examination archiving; The examination situation analysis module is used by invigilators to monitor the examination in real time and view the overall situation of the examination; The score management module is used to review and submit the test scores, and to rank and count the final scores; System management module, used to maintain and manage users, menus, roles, organizations, information, and perform permission verification on user access; The test terminal includes: Online exam module, used to verify the identity of candidates and provide cheating detection methods; The personal center module is used to display personal basic information, test records, and wrong question collection.

2. A smart examination system according to claim 1, characterized in that: The configuration method of the question bank management module is as follows: Configure the new module of question bank directory to distinguish the subjects of the test questions and maintain the organizational structure of the question bank; Configure the question type management module to manage and assign objective questions and subjective questions; Configure the question type import module to fill in the question information in the question template according to the template requirements; Configure the data deduplication module to perform duplication verification on the imported test question information after the test question verification work is completed. If there are duplicate questions, they will be marked and modified. After the modification is completed, if there are no duplicate questions, the data will be saved to the question bank for maintenance; Configure the question bank maintenance module to display the test questions in different levels according to the question bank directory; Configure the question bank review module to complete the import of test question data and put the test question data into the database after being reviewed by the administrator after the test question data is imported and is in the pending review status.

3. A smart examination system according to claim 2, characterized in that: The configuration method of the examination management module is as follows: Configure the examination information module to access the test data generated by the model management module, fill in the basic information of the test paper, and publish the test paper after reviewing it, and determine the information of the participating candidates; Configure the cheating detection module to turn on or off cheating detection. When the detection is turned on, the system will obtain the test client information and cheating check information in real time; Configure the marking module to set marking rules and assign examiners. Exam examiners can manually review or revise the intelligent scoring results. After reading the marking rules, online marking is implemented. If it is configured as intelligent marking, the intelligent scoring results can be manually revised. If it is configured as manual marking, each question needs to be scored. Configure the score archiving module to report scores and archive the exams after the exam review is completed; exam archiving includes manual archiving and automatic archiving.

4. A smart examination system according to claim 3, characterized in that: The configuration method of the examination situation analysis module is as follows: Configure the candidate information module to display and verify the candidate information in real time, and display the current test taker information and the real-time screen of the candidate; Configure the candidate statistics module to count candidates who have not logged into the system before the exam deadline. The system will automatically generate absence statistics; Configure the violation statistics module to count the cheating detection data sent by the online examination module. When there is a violation, choose to send a message manually or automatically to remind the candidate. The system records the number of violations of the candidate. If the number exceeds the set number, the system will force the candidate to log off. Configure the dynamic ranking module for candidates to sort them according to their current real-time scores and generate a dynamic ranking list.

5. The intelligent examination system according to claim 4, characterized in that: The configuration method of the score management module is: Configure the information access module to access the examinee scores, exam information, and test paper information in the exam management module, generate the final scores and rankings after the scores are reviewed, and provide score data to the personal center module; Configure the score list module to display the candidate's name, exam name, exam start and end time, exam duration, assessment results, and review status; and display the ranked list by setting ranking rules.

6. The intelligent examination system according to claim 5, characterized in that: The configuration method of the system management module is as follows: Configure the organization management module to display organization information in a tree structure hierarchy and support adding and importing organization data; Configure the menu management module to display menu name, code, component path, menu type, menu address, and menu status information; Configure the role management module to display role name, role code, role status information, and configure role menu access rights; Configure the user management module to display user account, user name, user role, and creation time information, and support modification of user information and password.

7. The intelligent examination system according to claim 6, characterized in that: The configuration method of the online examination module is as follows: Configure the candidate identity authentication module to verify the identity information recorded in the access system management module. Perform identity verification when logging into the examination system. If the identity verification passes, the candidate can enter the examination, otherwise the system will automatically exit the examination. Configure the exam instructions loading module to load the exam instructions from the exam management module and show the content, scope, and requirements of the exam to the candidates; Configuration detection loading module is used to access the cheating detection configuration in the test management, and conduct real-time detection and verification from the screen switching detection, screen domination detection, split screen detection, screenshot detection, video detection, and pasteboard detection methods. If any illegal operation occurs, the number of violations will be recorded, and the violation information will be submitted to the management end so that the invigilator can remind and record it; Configure the test question loading module to display the accessed test question data in order, distinguish between answered questions and unanswered questions, so that candidates can check for omissions and fill in the gaps; Configure the wrong test question reporting module to allow candidates to report wrong test questions found during the exam, and the invigilator will record the data.

8. The intelligent examination system according to claim 7, characterized in that: The configuration method of the personal center module is as follows: Configure the exam record module to display all the exam information that the user has taken; Configure the wrong question center module to display and count all wrong questions in the exams you have taken, and provide the function of redoing wrong questions; Configure the personal information module to display basic user information, count the total number of exams taken, and the total number of wrong questions.

Images