Assessment construction method based on artificial intelligence

Abstract

The invention discloses an assessment construction method based on artificial intelligence, and belongs to the technical field of workplace training assessment, and the method comprises the steps: S10, generating question data information according to the historical training content of an industry, generating a knowledge base based on the question data information, and generating assessment test paper based on the knowledge base; s20, collecting knowledge point features of the industry historical training content, distinguishing the knowledge point features, and obtaining relevance between examination questions in the examination paper and the knowledge point features; s30, obtaining the number of persons participating in the examination, obtaining the score number of each person participating in the examination, collecting the highest score and the median score in the score number, and checking the difficulty of the examination paper based on the highest score; according to the method, knowledge points and features of industry historical training are obtained, and test paper examination questions generated during historical training examination are collected, so that the knowledge base is deeply integrated and constructed, examination test paper can be intelligently generated in a future time period, the burdensome manual question setting burden of a trainee is relieved, and the question setting efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of workplace training assessment, and in particular to an assessment construction method based on artificial intelligence for creating an assessment question bank and test papers and marking the test papers. Background Art

[0002] In the current training and assessment system, the process of constructing question banks and test papers generally relies on manual operations. This process starts with the trainer's in-depth study of the courseware, and then designs the corresponding question bank based on these teaching materials. Subsequently, these planned question banks will become the basis for generating test papers. The compilation of test papers strictly follows the resource distribution and difficulty gradient in the question bank to ensure that the test content is comprehensive and challenging. In addition, trainers are required to mark and score the papers after the assessment.

[0003] In order to improve the efficiency of the entire training assessment, the patent document application number CN201911387660.9 provides an assessment method, device, equipment and readable storage medium based on artificial intelligence. The technical application believes that face-to-face assessment is often required in life and work. The assessment is often conducted by the examiner interviewing the candidate, asking the examiner questions, and then scoring the examinee's on-site oral answers. The technical application believes that for this type of assessment, the examiner's subjective experience and feelings often affect the examinee's final score, thereby reducing the fairness and accuracy of the assessment results. The technical application uses human-computer dialogue to automatically assess the examinee, and records the entire assessment process by audio and video recording. At the same time, semantic analysis technology and micro-expression emotion analysis technology are used to analyze the examinee's voice answers, and the examinee is comprehensively evaluated in combination with semantics and emotions to improve the fairness and accuracy of the assessment evaluation. At the same time, it is also beneficial to reduce the human cost required for the assessment, reduce the workload of human labor, and improve the efficiency of the assessment.

[0004] Another patent document application number CN202311427859.6 provides a method for building a cognitive comprehensive evaluation model based on artificial intelligence. This technical application has more comprehensive evaluation capabilities and active evaluation methods, strategy optimization capabilities and scalability, which can better meet the needs of people and machines for active evaluation of CCS. In real life, in the process of passing the examination, trainers often have to spend a lot of time studying courseware and designing questions. This process is relatively cumbersome. When faced with a large number of training and assessment arrangements in multiple batches, it is difficult for humans to meet the needs of fast question setting and marking, and it is difficult to ensure that the difficulty and knowledge range of the questions and test papers are consistent, and thus the fairness of the examination papers cannot be guaranteed. Summary of the invention

[0005] In view of the above problems existing in the technical field of existing workplace training and assessment, the present invention is proposed.

[0006] Therefore, one of the objects of the present invention is to provide an assessment construction method based on artificial intelligence, which obtains the knowledge points and characteristics of historical industry training and collects the test papers and questions generated during the historical training assessment to deeply integrate and construct a knowledge base, and then can intelligently generate assessment papers in future time periods, thereby reducing the heavy burden of manual question setting for trainers, and can optimize and adjust the test questions according to the scoring situation to ensure that the difficulty of the assessment paper is moderate, thereby ensuring the fairness of the assessment scoring.

[0007] In order to solve the above technical problems, the present invention provides the following technical solutions: In one aspect, the present invention provides an assessment construction method based on artificial intelligence, comprising the following steps: S10: Generate question data information according to the historical training content of the industry, generate a knowledge base based on the question data information, and generate an assessment test paper based on the knowledge base; S20: Collecting knowledge point features of the industry history training content, distinguishing the knowledge point features, and obtaining the correlation between the test questions in the assessment test paper and the knowledge point features; S30: Obtain the number of people participating in the assessment, and obtain the scores of each person participating in the assessment, collect the highest score and the median score from the scores, and review the difficulty of the assessment paper based on the highest score; S40: based on the preset highest score of the assessment paper, when the highest score among the persons participating in the assessment exceeds the preset highest score, obtaining the questions with high scores of the persons in the corresponding assessment paper, and analyzing the correlation characteristics between the questions and the corresponding training content knowledge points; S50: obtaining the number of persons who exceed the highest score among the persons participating in the assessment, and obtaining the total number of persons participating in the assessment, and calculating the proportion of the number of persons who exceed the highest score based on the total number of persons, and when the number of persons who exceed the highest score among the total number of persons participating in the assessment accounts for 40% to 50% of the total number of persons, it is determined that the difficulty of the corresponding assessment paper is low, and when the number of persons who exceed the highest score is less than 40% of the total number of persons, it is determined that the difficulty of the corresponding assessment paper is high; S60: Based on the fact that the number of people who exceed the highest score is less than 40% of the total number of people, the characteristics of the test questions and / or the types of test questions of the corresponding examination paper are collected. When the knowledge point assessment of the same training content is conducted in the future, the question type is adjusted according to the characteristics of the knowledge point of the training content.

[0008] As a preferred solution of the present invention, in which: in S30, before reviewing the difficulty of the examination paper based on the highest score, it also includes: Count the number of examinees in each scoring segment, where the scoring segments are divided into score segments with a difference of 5 to 10 points each, and collect the high-scoring and low-scoring questions from the examinees in each scoring segment, and analyze and judge the proficiency of each examinee in the knowledge points of the training content based on the collected questions; Collect the same questions among the questions with high scores of each examiner, and remove the same questions from the examination papers of future periods, and also remove other questions related to the knowledge points involved in the same questions; Collect the same questions among the questions with low scores of each examiner, and when generating test questions in the examination papers of future time periods, split the knowledge points involved in the same questions, and generate at least three test questions based on the split knowledge points.

[0009] As a preferred embodiment of the present invention, in S40, the high-scoring questions are questions with scores exceeding two-thirds of the total score specified for the questions. When the highest score among the persons participating in the assessment exceeds the preset highest score, it also includes obtaining the low-scoring questions of the persons in the corresponding assessment papers, and analyzing the correlation characteristics between the questions and the corresponding training content knowledge points, while recording the high-scoring questions and the low-scoring questions, generating a data set, and analyzing the differences in knowledge points involved in the high-scoring questions and the low-scoring questions.

[0010] As a preferred solution of the present invention, in which: in the S10, the knowledge point features are differentiated, including being differentiated into skill knowledge point features, application knowledge point features, definition-type knowledge point features and method-type knowledge point features, and scoring data information is collected based on the historically generated test questions based on the differentiated knowledge point features, and at the same time, the difficulty of each knowledge point feature is sorted from high to low according to the scoring data information. When personnel are trained and assessed in the future, emphasis is placed on generating questions on knowledge points with high difficulty, wherein the emphasis ratio accounts for at least 75% to 80% of the total number of questions in the assessment paper.

[0011] As a preferred solution of the present invention, in S60, the question type adjustment includes increasing and / or decreasing the number of questions, changing the score, changing the question type and adjusting the order of questions, wherein the increase and / or decrease in the number of questions includes increasing and decreasing the number of single-choice questions according to the characteristics of the training knowledge points, and increasing and decreasing the number of multiple-choice questions, the number of fill-in-the-blank questions and the number of answer questions; The score changes include keeping the scores of single-choice questions and fill-in-the-blank questions unchanged, and increasing and / or decreasing the scores of multiple-choice questions and essay questions based on the original scores.

[0012] As a preferred solution of the present invention, in the assessment paper, the number of knowledge points contained in each question is collected. When the number of knowledge points correctly answered by the assessor to the question exceeds two-thirds of the number of knowledge points contained in the corresponding question, a high score is given to the question. Otherwise, a high score is not given to the question.

[0013] As a preferred solution of the present invention, the total number of questions in the assessment paper is obtained, and each question in the assessment paper is individually intercepted. When the number of knowledge points correctly answered by the examinee for each question exceeds two-thirds of the number of knowledge points contained in the corresponding question, the proportion of the number of questions in the total number of questions in the assessment paper is counted. When the proportion exceeds two-thirds of the total number of questions in the assessment paper, the examinee is judged to have passed the assessment, otherwise, it is not judged.

[0014] As a preferred solution of the present invention, when the proportion of the counted questions exceeds two-thirds of the total number of questions in the assessment paper, the type of the question is obtained, and when conducting training assessments in future time periods, the type of questions that are not given high scores is converted to the same type as the question whose proportion of the counted questions exceeds two-thirds of the total number of questions in the assessment paper.

[0015] On the other hand, the present invention provides a terminal comprising a processor, an input interface, an output interface and a memory, wherein the memory is used to store a computer program, the computer program comprises program instructions, and the processor is configured to call the program instructions to execute the steps of any one of the above artificial intelligence-based assessment construction methods.

[0016] The present invention also provides a computer-readable storage medium, which stores a computer program. The computer program includes program instructions, and when the program instructions are executed by a processor, the processor executes the steps of any one of the above artificial intelligence-based assessment construction methods.

[0017] Beneficial effects: The present invention obtains the knowledge points and characteristics of the industry's historical training, and collects the test questions generated during the historical training assessment, thereby deeply integrating and constructing a knowledge base. Based on the knowledge base constructed by deep integration, the assessment papers can be intelligently generated in the future time period, which reduces the trainer's heavy burden of manual question setting, so that the test questions on the assessment paper can be automatically generated according to the characteristics of the trained knowledge points. At the same time, the difficulty of the test paper can be judged according to the test scoring situation, and the test questions generated in the future time period can be optimized and adjusted according to the scoring situation, so that the generated test questions are both in line with the training content and challenging, so as to ensure that the difficulty of the assessment paper is moderate. In addition, the present invention makes a final judgment on the assessor based on the overall correct answer number of the knowledge point rather than the total score of the test paper, so as to analyze the assessor's proficiency in the knowledge points, thereby ensuring the fairness of the assessment pass or fail. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative labor. Among them: Figure 1 A schematic diagram of a method flow chart of an embodiment of the present invention; Figure 2 It is a schematic diagram of the process structure of an embodiment of the present invention. DETAILED DESCRIPTION

[0019] In order to make the purpose, technical solution and advantages of the embodiment of the present invention clearer, the technical solution of the embodiment of the present invention will be clearly and completely described below in conjunction with the drawings of the embodiment of the present invention. Obviously, the described embodiment is a part of the embodiment of the present invention, not all of the embodiments. Based on the described embodiment of the present invention, all other embodiments obtained by ordinary technicians in this field belong to the scope of protection of the present invention.

[0020] During the entire training and assessment process, trainers have to spend a lot of time studying courseware and designing questions, which is a rather cumbersome process. When faced with a large number of training and assessment arrangements in multiple batches, it is difficult for humans to meet the needs of quickly setting questions and marking papers, and it is difficult to ensure that the difficulty level and knowledge point scope of the questions and papers are consistent, and thus it is impossible to ensure the fairness of the assessment papers.

[0021] Based on this, the present invention proposes an assessment construction method based on artificial intelligence, which obtains the knowledge points and characteristics of historical industry training and collects the test papers and questions generated during the historical training assessment to deeply integrate and construct a knowledge base, and then can intelligently generate assessment papers in future time periods, reducing the heavy burden of manual question setting for trainers, and can optimize and adjust the questions according to the scoring situation to ensure that the difficulty of the assessment papers is moderate, thereby ensuring the fairness of the assessment scores.

[0022] The present invention is further described in detail below through embodiments and in conjunction with the accompanying drawings.

[0023] Reference Figure 1 to Figure 2 , is an embodiment of the present invention, which provides an assessment construction method based on artificial intelligence, comprising the following steps: S10: Generate question data information according to the historical training content of the industry, generate a knowledge base based on the question data information, and generate an assessment test paper based on the knowledge base; Specifically, in this embodiment, the knowledge point features are differentiated, including being differentiated into skill knowledge point features, application knowledge point features, definitional knowledge point features and methodical knowledge point features, and scoring data information is collected based on the historically generated test questions based on the differentiated knowledge point features, and at the same time, the difficulty of each knowledge point feature is sorted from high to low according to the scoring data information. When training and assessing personnel in the future period, the focus is on generating questions of knowledge points with high difficulty, wherein the focus ratio accounts for at least 75% to 80% of the total number of questions in the assessment test paper; S20: Collect the knowledge point features of the industry's historical training content, differentiate the knowledge point features, and obtain the correlation between the test questions in the examination paper and the knowledge point features; S30: Obtain the number of people participating in the assessment, and obtain the scores of each person participating in the assessment, collect the highest score and the median score from the scores, and review the difficulty of the assessment paper based on the highest score; It should be noted that, before reviewing the difficulty of the examination paper based on the highest score, the present embodiment also includes: Count the number of examinees in each scoring range. The scoring range includes a score range of 5 to 10 points. Collect the high-scoring and low-scoring questions from the examinees in each scoring range, and analyze and judge the proficiency of each examinee in the training content knowledge points based on the collected questions; Collect the same questions among the questions with high scores from each examiner, and remove the same questions from the examination papers in the future period, including removing other questions related to the knowledge points involved in the same questions; Collect the same questions among the questions with low scores from each examiner, and when generating test questions in the examination papers of the future period, split the knowledge points involved in the same questions, and generate at least three test questions based on the split knowledge points; This embodiment collects the same questions among the questions with low scores of each examiner, and splits the knowledge point according to the knowledge point involved in the question, so as to generate at least three test questions from the original question, thereby reducing the difficulty of the test questions involving the knowledge point, which is more conducive to assessing the proficiency level of each examiner in the knowledge point; S40: based on the preset highest score of the assessment paper, when the highest score among the persons participating in the assessment exceeds the preset highest score, the questions with high scores of the persons in the corresponding assessment paper are obtained, and the correlation characteristics between the questions and the corresponding training content knowledge points are analyzed; It should be emphasized in this embodiment that the high-scoring questions are questions with scores exceeding two-thirds of the total score specified for the questions. When the highest score among the persons participating in the assessment exceeds the preset highest score, it also includes obtaining the low-scoring questions of the persons in the corresponding assessment papers, analyzing the correlation characteristics between the questions and the corresponding training content knowledge points, and recording the high-scoring questions and the low-scoring questions at the same time, generating a data set, and analyzing the differences in knowledge points involved in the high-scoring questions and the low-scoring questions; In this embodiment, the number of knowledge points included in each question is collected in the examination paper. When the number of knowledge points correctly answered by the examinee to a question exceeds two-thirds of the number of knowledge points included in the corresponding question, the question is given a high score. Otherwise, the question is not given a high score. On the basis of the above, this embodiment further obtains the total number of questions in the examination paper, and separately intercepts each question in the examination paper. When the number of knowledge points correctly answered by the examinee for each question exceeds two-thirds of the number of knowledge points contained in the corresponding question, the proportion of the number of questions to the total number of questions in the examination paper is counted. When the proportion exceeds two-thirds of the total number of questions in the examination paper, it is determined that the examinee has passed the examination, otherwise, it is not determined; According to the experience of various assessment standards in the past, the traditional evaluation method is often to determine whether the assessee has passed the assessment based on the final score. However, in the assessment paper, various question types are different, the knowledge points they contain are also different, and the difficulty and number of the knowledge points contained are also different. If in a certain training assessment, the number of high-difficulty knowledge point questions is lower than the number of low-difficulty knowledge point questions, then in this assessment paper, most of the scores are based on low-difficulty knowledge point questions, and the scores of high-difficulty knowledge point questions only account for a small number. Therefore, in this case, it is unfair to judge the familiarity of the assessee's knowledge point mastery. Therefore, this embodiment separately intercepts and calculates each question in the assessment paper, and limits the number of knowledge points that the assessee correctly answers to each question to more than two-thirds of the number of knowledge points contained in the corresponding question, and on this basis, the proportion of the number of questions to the total number of questions in the assessment paper is calculated. When the proportion exceeds two-thirds of the total number of questions in the assessment paper, it is determined that the assessee has a proficient grasp of the training knowledge points and has passed the assessment, thereby ensuring the fairness of the assessment.

[0024] At the same time, this embodiment also includes: When the proportion of the number of questions counted exceeds two-thirds of the total number of questions on the assessment paper, the type of the question is obtained, and when conducting training assessments in future periods, the type of questions that are not given high scores is converted to the same type as the type of questions whose proportion of the number of questions counted exceeds two-thirds of the total number of questions on the assessment paper; S50: obtaining the number of people who exceed the highest score among the people participating in the assessment, and obtaining the total number of people participating in the assessment, and calculating the proportion of people who exceed the highest score based on the total number of people. When the number of people who exceed the highest score accounts for 40% to 50% of the total number of people participating in the assessment, it is determined that the difficulty of the corresponding assessment paper is low; when the number of people who exceed the highest score is less than 40% of the total number of people, it is determined that the difficulty of the corresponding assessment paper is high; S60: Based on the number of people who exceed the highest score being less than 40% of the total number of people, the characteristics and / or types of questions of the corresponding examination papers are collected. When the knowledge point examination of the same training content is conducted in the future, the question types are adjusted according to the characteristics of the knowledge points of the training content; It should be noted that the adjustment of question types in this embodiment includes increasing and / or decreasing the number of questions, changing the score, changing the question type and adjusting the order of questions, wherein increasing and / or decreasing the number of questions includes increasing and decreasing the number of single-choice questions according to the characteristics of the training knowledge points, and increasing and decreasing the number of multiple-choice questions, fill-in-the-blank questions and answer questions; The change in score value includes keeping the score values ​​of single-choice questions and fill-in-the-blank questions unchanged, and increasing and / or decreasing the score values ​​of multiple-choice questions and essay questions based on the original score values; This embodiment can prevent the participants from guessing or betting on questions by adjusting the order of questions, so as to ensure the fairness of the examination and the rationality of the difficulty distribution. A terminal comprises a processor, an input interface, an output interface and a memory, wherein the memory is used to store a computer program, the computer program comprises program instructions, and the processor is configured to call the program instructions to execute the steps of any one of the above artificial intelligence-based assessment construction methods.

[0025] A computer-readable storage medium stores a computer program, wherein the computer program includes program instructions, and when the program instructions are executed by a processor, the processor executes the steps of any one of the above-mentioned artificial intelligence-based assessment construction methods.

[0026] In summary, the present invention obtains the knowledge points and characteristics of the industry's historical training, and collects the test questions generated during the historical training assessment, thereby deeply integrating and constructing a knowledge base. Based on the knowledge base constructed by deep integration, the assessment papers can be intelligently generated in the future time period, which reduces the trainer's heavy burden of manual question setting, so that the test questions on the assessment paper can be automatically generated according to the characteristics of the trained knowledge points. At the same time, the difficulty of the test paper can be judged according to the test scoring situation, and the test questions generated in the future time period can be optimized and adjusted according to the scoring situation, so that the generated test questions are both in line with the training content and challenging, so as to ensure that the difficulty of the assessment paper is moderate. In addition, the present invention makes a final judgment on the assessor based on the overall correct answer number of the knowledge point rather than the total score of the test paper, so as to analyze the assessor's proficiency in the knowledge points, thereby ensuring the fairness of the assessment.

[0027] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. The assessment construction method based on artificial intelligence is characterized by: The following steps are involved: S10: Generate question data information according to the historical training content of the industry, generate a knowledge base based on the question data information, and generate an assessment test paper based on the knowledge base; S20: Collecting knowledge point features of the industry history training content, distinguishing the knowledge point features, and obtaining the correlation between the test questions in the assessment test paper and the knowledge point features; S30: Obtain the number of people participating in the assessment, and obtain the scores of each person participating in the assessment, collect the highest score and the median score from the scores, and review the difficulty of the assessment paper based on the highest score; S40: based on the preset highest score of the assessment paper, when the highest score among the persons participating in the assessment exceeds the preset highest score, obtaining the questions with high scores of the persons in the corresponding assessment paper, and analyzing the correlation characteristics between the questions and the corresponding training content knowledge points; S50: obtaining the number of persons who exceed the highest score among the persons participating in the assessment, and obtaining the total number of persons participating in the assessment, and calculating the proportion of the number of persons who exceed the highest score based on the total number of persons, and when the number of persons who exceed the highest score among the total number of persons participating in the assessment accounts for 40% to 50% of the total number of persons, it is determined that the difficulty of the corresponding assessment paper is low, and when the number of persons who exceed the highest score is less than 40% of the total number of persons, it is determined that the difficulty of the corresponding assessment paper is high; S60: Based on the fact that the number of people who exceed the highest score is less than 40% of the total number of people, the characteristics of the test questions and / or the types of test questions of the corresponding examination paper are collected. When the knowledge point assessment of the same training content is conducted in the future, the question type is adjusted according to the characteristics of the knowledge point of the training content.

2. The artificial intelligence-based assessment construction method according to claim 1, characterized in that: In S30, before reviewing the difficulty of the examination paper based on the highest score, the method further includes: Count the number of examinees in each scoring segment, where the scoring segments are divided into score segments with a difference of 5 to 10 points each, and collect the high-scoring and low-scoring questions from the examinees in each scoring segment, and analyze and judge the proficiency of each examinee in the knowledge points of the training content based on the collected questions; Collect the same questions among the questions with high scores of each examiner, and remove the same questions from the examination papers of future periods, and also remove other questions related to the knowledge points involved in the same questions; Collect the same questions among the questions with low scores of each examiner, and when generating test questions in the examination papers of future time periods, split the knowledge points involved in the same questions, and generate at least three test questions based on the split knowledge points.

3. The artificial intelligence-based assessment construction method according to claim 1, characterized in that: In S40, the high-scoring questions are questions with scores exceeding two-thirds of the total score specified for the questions. When the highest score among the persons participating in the assessment exceeds the preset highest score, it also includes obtaining the low-scoring questions of the persons in the corresponding assessment papers, and analyzing the correlation characteristics between the questions and the corresponding training content knowledge points, while recording the high-scoring questions and the low-scoring questions, generating a data set, and analyzing the differences in knowledge points involved in the high-scoring questions and the low-scoring questions.

4. The artificial intelligence-based assessment construction method according to claim 1, characterized in that: In S10, the knowledge point features are differentiated, including being differentiated into skill knowledge point features, application knowledge point features, definition-type knowledge point features and method-type knowledge point features, and scoring data information is collected based on historically generated test questions based on the differentiated knowledge point features. At the same time, the difficulty of each knowledge point feature is sorted from high to low according to the scoring data information. When training and assessing personnel in the future, emphasis is placed on generating questions on knowledge points with high difficulty, wherein the emphasis ratio accounts for at least 75% to 80% of the total number of questions in the assessment paper.

5. The artificial intelligence-based assessment construction method according to claim 1, characterized in that: In S60, the question type adjustment includes increasing and / or decreasing the number of questions, changing the score, changing the question type and adjusting the order of questions, wherein the increasing and / or decreasing the number of questions includes increasing and decreasing the number of single-choice questions according to the characteristics of the training knowledge points, and increasing and decreasing the number of multiple-choice questions, fill-in-the-blank questions and answer questions; The score changes include keeping the scores of single-choice questions and fill-in-the-blank questions unchanged, and increasing and / or decreasing the scores of multiple-choice questions and essay questions based on the original scores.

6. The artificial intelligence-based assessment construction method according to claim 1, characterized in that: In the examination paper, the number of knowledge points contained in each question is collected. When the number of knowledge points correctly answered by the examinee to the question exceeds two-thirds of the number of knowledge points contained in the corresponding question, a high score is given to the question. Otherwise, a high score is not given to the question.

7. The artificial intelligence-based assessment construction method according to claim 6, characterized in that: The total number of questions in the examination paper is obtained, and each question in the examination paper is individually intercepted. When the number of knowledge points correctly answered by the examinee for each question exceeds two-thirds of the number of knowledge points contained in the corresponding question, the proportion of the number of questions in the total number of questions in the examination paper is calculated. When the proportion exceeds two-thirds of the total number of questions in the examination paper, the examinee is determined to have passed the examination, otherwise, it is not determined.

8. The artificial intelligence-based assessment construction method according to claim 7, characterized in that: When the counted number of questions accounts for more than two-thirds of the total number of questions on the assessment paper, the type of the question is obtained, and when conducting training assessments in future periods, the type of questions that are not given high scores will be converted to the same type as the question whose counted number of questions accounts for more than two-thirds of the total number of questions on the assessment paper.

9. A terminal, characterized in that: The method comprises a processor, an input interface, an output interface and a memory, wherein the memory is used to store a computer program, the computer program comprises program instructions, and the processor is configured to call the program instructions to execute the method according to any one of claims 1 to 8.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, wherein the computer program includes program instructions, and when the program instructions are executed by a processor, the processor is enabled to perform the method according to any one of claims 1 to 8.

Images