Online examination customization method and system based on personalized learning

Abstract

The invention provides an online examination customization method and system based on personalized learning. The method comprises the following steps: acquiring basic information and learning purpose information of a current student; obtaining a knowledge range of personalized learning of the current student and learning cumulative class hours in real time; according to the personalized learning knowledge range of the current student and the learning cumulative class hours, judging whether examination conditions are met or not; if yes, the matched examination questions are extracted from the examination question database according to a preset examination customization algorithm, and corresponding examination papers are generated; and pushing the corresponding test paper to the front-end equipment of the current student so as to carry out online customized examination on the current student. According to the invention, on the basis of the personalized learning condition of each student, the adaptive online customized examination scheme can be made, so that each student can effectively receive knowledge, and the score or learning skill of the student can be improved.

Description

Technical Field

[0001] The present invention relates to the technical field of online teaching management, and in particular to an online examination customization method and system based on personalized learning. Background Art

[0002] With the continuous development and popularization of network technology and the decline in application costs, the Internet is rapidly penetrating into all aspects of social life in a brand-new way. Online knowledge dissemination has gradually gained social recognition for its simplicity, flexibility and large amount of information. As a result, online education has become more and more a research hotspot. Web-based online teaching has become an important method and development direction of online teaching. Online teaching has made our traditional teaching methods no longer restricted by time and space, and students can arrange their studies and exams freely.

[0003] Today's online education is usually unified training. Due to the different qualities of students, as well as their ages, occupations, marital status, and childbearing information, their study times vary. It is difficult to ensure that every student can master the training knowledge in a timely and effective manner. Teachers are also unable to accurately grasp the teaching progress, knowledge difficulty, and amount of knowledge. Faced with the uneven study time of students, training institutions usually have to make a one-size-fits-all decision, so that all students are trained and learn according to the same lesson plan, resulting in the same class, the same teacher, and the same hard-working students, but with very different levels of learning. How to ensure that every student can receive targeted training, effectively absorb knowledge, and improve their own grades or learning skills is an urgent problem to be solved in current learning methods. Summary of the invention

[0004] In order to solve at least one of the above technical problems, the present invention proposes an online examination customization method and system based on personalized learning, which can make an online customized examination plan suitable for each student based on the personalized learning situation of each student, thereby ensuring that each student can effectively receive knowledge and improve their own grades or learning skills.

[0005] The first aspect of the present invention provides an online test customization method based on personalized learning, the method comprising:

[0006] Obtain basic information of current students and information on their learning objectives;

[0007] Obtain the knowledge scope and accumulated hours of current students' personalized learning in real time;

[0008] Determine whether the student meets the examination requirements based on the student's current personalized learning knowledge scope and accumulated learning hours;

[0009] If yes, based on the current student's basic information, learning purpose information, and the knowledge scope of personalized learning, matching test questions are extracted from the test question database according to the preset test customization algorithm, and the corresponding test paper is generated;

[0010] The corresponding test papers are pushed to the front-end device of the current student to conduct an online customized test for the current student.

[0011] Furthermore, based on the knowledge scope of the current student's personalized learning and the accumulated hours of learning, it is determined whether the examination conditions are met, including:

[0012] Construct a knowledge mastery prediction model;

[0013] The knowledge mastery degree prediction model is trained and optimized through sample data to obtain an optimized knowledge mastery degree prediction model;

[0014] The knowledge scope and accumulated hours of the current students' personalized learning are used to predict the current students' knowledge mastery through the knowledge mastery prediction model;

[0015] It is determined whether the predicted knowledge mastery level of the current student is greater than a first preset threshold value. If so, the examination conditions are met.

[0016] Furthermore, after the knowledge scope and accumulated hours of personalized learning of the current student are used to predict the knowledge mastery level of the current student through the knowledge mastery level prediction model, the method further includes:

[0017] Acquire history learning data, each history learning data at least includes basic information and learning purpose information of history students, history learning knowledge scope, accumulated history learning class hours, history test paper difficulty coefficient, and history test scores;

[0018] Based on each history learning data, the basic information and learning purpose information of the history students, the scope of history learning knowledge, and the accumulated history learning hours are calculated to obtain the characteristic value of each history student;

[0019] Perform feature calculation on the basic information and learning purpose information of the current student, the knowledge scope of personalized learning, and the accumulated learning hours to obtain the feature value of the current student;

[0020] Based on each historical learning data, the approximation between the feature value of each historical student and the feature value of the current student is calculated;

[0021] Determine whether the calculated approximation is greater than a second preset threshold, and if so, add the corresponding historical learning data to the approximate feature library;

[0022] Based on each historical learning data in the approximate feature library, the corresponding historical learning knowledge scope and the accumulated historical learning class hours are input into the knowledge mastery degree prediction model to obtain the knowledge mastery degree prediction value of each history student;

[0023] Based on each historical learning data in the approximate feature library, the corresponding historical test paper difficulty coefficient and historical test score are calculated through a preset replacement algorithm to obtain the actual value of each history student's knowledge mastery level;

[0024] Based on each historical learning data in the approximate feature library, the actual value of each history student's knowledge mastery is subtracted from the predicted value of each history student's knowledge mastery to obtain a knowledge mastery vector difference;

[0025] Adding the knowledge mastery degree vector difference values ​​calculated based on each historical learning data in the approximate feature library to obtain the knowledge mastery degree vector difference sum;

[0026] Divide the sum of the knowledge mastery degree vector differences by the total number of historical learning data in the approximate feature library to obtain the average difference of the knowledge mastery degree vector;

[0027] The predicted knowledge mastery level of the current student and the average difference between the knowledge mastery level vectors are vector summed to obtain the corrected knowledge mastery level of the current student.

[0028] Furthermore, the corresponding history test paper difficulty coefficient and history test score are calculated through a preset replacement algorithm to obtain the actual value of each history student's knowledge mastery, including:

[0029] Multiply the corresponding history test paper difficulty coefficient by the history test score, and divide the product by the total score of the history test paper to obtain the actual value of each history student's knowledge mastery.

[0030] Further, based on each historical learning data, feature calculation is performed on the basic information and learning purpose information of the history student, the knowledge scope of history learning, and the accumulated class hours of history learning to obtain the feature value of each history student; feature calculation is performed on the basic information and learning purpose information of the current student, the knowledge scope of personalized learning, and the accumulated class hours of learning to obtain the feature value of the current student; based on each historical learning data, the feature value of each history student is similar to the feature value of the current student, specifically including:

[0031] Based on each piece of history learning data, the age information and educational background information of the history student are extracted from the basic information of the history student, the age information and educational background information of the history student are replaced with the age grade and educational background grade of the history student through a preset age grade table and a preset educational background grade table, respectively, and the learning purpose information of the history student is replaced with the learning purpose grade of the history student through a preset learning purpose grade table, the age grade of the history student is multiplied by the learning purpose grade of the history student to obtain a first product, and the educational background grade of the history student is multiplied by the learning purpose grade of the history student to obtain a second product, the first product and the second product are added to obtain a first value, the age grade of the history student and the educational background grade of the history student are added to obtain a second value, and the first value is divided by the second value to obtain an attribute characteristic value of the history student;

[0032] Based on each historical learning data, the number of corresponding knowledge points is counted according to the historical learning knowledge scope, and the number of corresponding knowledge points is divided by the accumulated historical learning hours to obtain the learning characteristic value of the history student;

[0033] Extracting the age information and educational level information of the current student from the basic information of the current student, replacing the age information and educational level information of the current student into the age level and educational level of the current student through a preset age level table and a preset educational level table, respectively, and replacing the learning purpose information of the current student into the learning purpose level of the current student through a preset learning purpose level table, multiplying the age level of the current student by the learning purpose level of the current student to obtain a third product, and multiplying the educational level of the current student by the learning purpose level of the current student to obtain a fourth product, adding the third product and the fourth product to obtain a third value, adding the age level of the current student and the educational level of the current student to obtain a fourth value, and dividing the third value by the fourth value to obtain an attribute characteristic value of the current student;

[0034] Count the number of corresponding knowledge points from the knowledge scope of the current student's personalized learning, and divide the number of corresponding knowledge points by the current student's accumulated learning hours to obtain the current student's learning characteristic value;

[0035] Based on each historical learning data, the attribute feature value of the historical student and the attribute feature value of the current student are approximated to obtain the attribute feature approximation based on each historical learning data;

[0036] Based on each historical learning data, the learning feature value of the historical students and the learning feature value of the current students are approximated to obtain the learning feature approximation based on each historical learning data;

[0037] The influence weights of preset attribute features and learning features on judging whether the characteristics of historical students and current students are similar are different;

[0038] Based on each historical learning data, the corresponding attribute feature approximation is multiplied by the influence weight of the attribute feature to obtain the attribute feature weight approximation, and the corresponding learning feature approximation is multiplied by the influence weight of the learning feature to obtain the learning feature weight approximation;

[0039] Based on each historical learning data, the corresponding attribute feature weight approximation is added to the corresponding learning feature weight approximation to obtain a fifth value, and the influence weight of the attribute feature is added to the influence weight of the learning feature to obtain a sixth value;

[0040] Based on each historical learning data, the calculated fifth value is divided by the sixth value to obtain the approximation between the characteristic value of the corresponding historical student and the characteristic value of the current student.

[0041] Furthermore, according to the preset test customization algorithm, matching test questions are extracted from the test question database, and corresponding test papers are generated, including:

[0042] According to the knowledge scope of personalized learning, corresponding test questions are extracted from the test question database and added to the preliminary test question bank;

[0043] There are m test questions in the preset preliminary test question bank, and the number of test questions required for an examination paper is n, and m is greater than n;

[0044] Randomly select n test questions from m test questions to form a candidate test paper, thus forming 1. Candidate examination papers and add them to the candidate examination paper database;

[0045] Compare each candidate test paper in the candidate test paper library with the remaining candidate test papers in pairs, and determine which candidate test paper is more suitable for the current student's learning purpose information. If the former is more suitable, add 1 point to the former's learning purpose matching item;

[0046] Compare each candidate test paper in the candidate test paper database with the remaining candidate test papers in pairs, and determine which candidate test paper has a test difficulty that is more suitable for the current student's knowledge level. If the former is more suitable, add 1 point to the former's test difficulty matching item;

[0047] Compare each candidate test paper in the candidate test paper database with the remaining candidate test papers in pairs, and determine which candidate test paper has a more reasonable combination of test questions of different difficulty levels. If the former is more reasonable, add 1 point to the reasonableness of the former's test question combination;

[0048] After all candidate test papers in the candidate test paper database have been compared in pairs, the scores of each candidate test paper in the learning purpose matching item, the test question difficulty matching item, and the test question reasonable matching item are counted;

[0049] The influence weights of the preset learning purpose matching items, test question difficulty matching items, and test question reasonable matching items on the matching test papers are different;

[0050] Multiply the score of each candidate test paper in the learning purpose matching item by the corresponding influence weight to obtain the weight score of the learning purpose matching item; multiply the score of each candidate test paper in the test question difficulty matching item by the corresponding influence weight to obtain the weight score of the test question difficulty matching item; multiply the score of each candidate test paper in the test question reasonable matching item by the corresponding influence weight to obtain the weight score of the test question reasonable matching item;

[0051] The weighted score of the learning purpose matching item, the weighted score of the test question difficulty matching item, and the weighted score of the test question reasonable combination item of each candidate test paper are added together to obtain the matching score of each candidate test paper;

[0052] Each candidate test paper is sorted based on its matching score, and the candidate test paper with the highest matching score is used as the corresponding test paper.

[0053] Furthermore, each candidate test paper in the candidate test paper library is compared with the remaining candidate test papers in pairs, and it is determined which candidate test paper is more suitable for the current student's learning purpose information, specifically including:

[0054] Each exam question in the preset exam database comes with a degree of fit suitable for each learning purpose;

[0055] Based on each test question of the two compared candidate test papers, the degree of fit of the corresponding test question with the current student's learning purpose information is obtained from the accompanying information;

[0056] Count whether the two candidate examination papers have questions with a degree of fit less than a third preset threshold. If the former candidate examination paper does not have such questions, but the latter candidate examination paper does, the former candidate examination paper is judged to be more suitable, and 1 point is added to the former learning purpose matching item; if the former candidate examination paper has such questions, but the latter candidate examination paper does not, the latter candidate examination paper is judged to be more suitable, and no points are added; if both the former candidate examination paper and the latter candidate examination paper have such questions, it is judged that neither of them is suitable, and no points are added; if neither the former candidate examination paper nor the latter candidate examination paper has such questions, proceed to the next step;

[0057] Based on the former candidate test paper, the degree of fit of each test question with the learning purpose information of the current student is multiplied by its corresponding score to obtain the fit score of each test question; the fit scores of all test questions in the former candidate test paper are added to obtain the sum of the fit scores, and the sum of the fit scores is divided by the total score of the former candidate test paper to obtain the learning purpose fit of the former candidate test paper;

[0058] Based on the latter candidate test paper, the degree of fit of each test question with the learning purpose information of the current student is multiplied by its corresponding score to obtain the fit score of each test question; the fit scores of all test questions in the latter candidate test paper are added to obtain the sum of the fit scores, and the sum of the fit scores is divided by the total score of the latter candidate test paper to obtain the learning purpose fit of the latter candidate test paper;

[0059] Compare the degree of fit of the learning purpose of the former candidate test paper with the degree of fit of the learning purpose of the latter candidate test paper. If the former is greater than the latter, the former is judged to be more suitable, and 1 point is added to the learning purpose match of the former. Otherwise, no point is added.

[0060] Furthermore, each candidate test paper in the candidate test paper database is compared with the remaining candidate test papers in pairs, and it is determined which candidate test paper has a more difficult test that is more suitable for the current student's knowledge level, including:

[0061] Each question in the preset question database is accompanied by a difficulty value;

[0062] Based on each test question of the two compared candidate test papers, the difficulty value of the corresponding test question is obtained from the accompanying information;

[0063] Count whether the two candidate test papers have test questions with a difficulty value less than the fourth preset threshold or greater than the fifth preset threshold. If the former candidate test paper does not have such a test question, but the latter candidate test paper does, then the former candidate test paper is judged to be more suitable, and 1 point is added to the former test question difficulty matching item; if the former candidate test paper has such a test question, but the latter candidate test paper does not, then the latter candidate test paper is judged to be more suitable, and no point is added; if both the former candidate test paper and the latter candidate test paper have such a test question, then it is judged that neither of them is suitable, and no point is added; if neither the former candidate test paper nor the latter candidate test paper has such a test question, then proceed to the next step;

[0064] Based on the former candidate test paper, multiply the difficulty value of each test question by its corresponding score to obtain the difficulty score of each test question; add the difficulty scores of all test questions in the former candidate test paper to obtain the sum of the difficulty scores, and divide the sum of the difficulty scores by the total score of the former candidate test paper to obtain the test question difficulty value of the former candidate test paper;

[0065] Based on the latter candidate test paper, multiply the difficulty value of each test question by its corresponding score to obtain the difficulty score of each test question; add the difficulty scores of all test questions in the latter candidate test paper to obtain the sum of the difficulty scores, and divide the sum of the difficulty scores by the total score of the latter candidate test paper to obtain the test question difficulty value of the latter candidate test paper;

[0066] The difficulty of the pre-set test papers is increased by 20% based on the students' knowledge level to make the test papers most suitable for them.

[0067] Determine which one of the difficulty values ​​of the former candidate test paper and the latter candidate test paper is closer to the current student's knowledge level, and add 20% difficulty. If the former is closer, the former is judged to be more suitable, and 1 point is added to the former's test difficulty match item. Otherwise, no point is added.

[0068] The present invention also proposes an online test customization system based on personalized learning, comprising a memory and a processor, wherein the memory comprises an online test customization method program based on personalized learning, and when the online test customization method program based on personalized learning is executed by the processor, the following steps are implemented:

[0069] Obtain basic information of current students and information on their learning objectives;

[0070] Obtain the knowledge scope and accumulated hours of current students' personalized learning in real time;

[0071] Determine whether the student meets the examination requirements based on the student's current personalized learning knowledge scope and accumulated learning hours;

[0072] If yes, based on the current student's basic information, learning purpose information, and the knowledge scope of personalized learning, matching test questions are extracted from the test question database according to the preset test customization algorithm, and the corresponding test paper is generated;

[0073] The corresponding test papers are pushed to the front-end device of the current student to conduct an online customized test for the current student.

[0074] Furthermore, based on the knowledge scope of the current student's personalized learning and the accumulated hours of learning, it is determined whether the examination conditions are met, including:

[0075] Construct a knowledge mastery prediction model;

[0076] The knowledge mastery degree prediction model is trained and optimized through sample data to obtain an optimized knowledge mastery degree prediction model;

[0077] The knowledge scope and accumulated hours of the current students' personalized learning are used to predict the current students' knowledge mastery through the knowledge mastery prediction model;

[0078] It is determined whether the predicted knowledge mastery level of the current student is greater than a first preset threshold value. If so, the examination conditions are met.

[0079] The present invention proposes an online test customization method and system based on personalized learning, which first obtains the basic information and learning purpose information of the current student, and obtains the knowledge scope of the current student's personalized learning and the accumulated learning hours in real time, and then judges whether the test conditions are met according to the knowledge scope of the current student's personalized learning and the accumulated learning hours; if so, based on the basic information and learning purpose information of the current student, as well as the knowledge scope of personalized learning, and according to the preset test customization algorithm, the matching test questions are extracted from the test question database, and the corresponding test papers are generated; finally, the corresponding test papers are pushed to the front-end device of the online student, so as to conduct an online customized test for the current student. The present invention can make an online customized test plan adapted to each student's personalized learning status, thereby ensuring that each student can effectively receive knowledge and improve their own grades or learning skills.

[0080] Additional aspects and advantages of the present invention will be given in part in the following description and, in part, will be obvious from the following description or learned through practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0081] Figure 1 A flow chart of an online test customization method based on personalized learning according to the present invention is shown;

[0082] Figure 2 A flow chart showing a method for determining whether a current student has met the examination conditions of the present invention;

[0083] Figure 3 A block diagram of an online test customization system based on personalized learning according to the present invention is shown. DETAILED DESCRIPTION

[0084] In order to more clearly understand the above-mentioned purpose, features and advantages of the present invention, the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be noted that the embodiments of the present application and the features in the embodiments can be combined with each other without conflict.

[0085] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Therefore, the protection scope of the present invention is not limited to the specific embodiments disclosed below.

[0086] Figure 1 A flow chart of an online test customization method based on personalized learning according to the present invention is shown.

[0087] like Figure 1 As shown, the first aspect of the present invention proposes an online test customization method based on personalized learning, the method comprising:

[0088] S102, obtaining basic information and learning purpose information of the current student;

[0089] S104, obtaining the knowledge scope and accumulated learning hours of the current student's personalized learning in real time;

[0090] S106, judging whether the examination conditions are met according to the knowledge scope of the current personalized learning of the student and the accumulated learning hours;

[0091] S108, if yes, then based on the basic information of the current student, the learning purpose information, and the knowledge scope of personalized learning, and according to the preset test customization algorithm, matching test questions are extracted from the test question database, and a corresponding test paper is generated;

[0092] S110, pushing the corresponding test paper to the front-end device of the current student, so as to conduct an online customized test for the current student.

[0093] The present invention first obtains the basic information and learning purpose information of the current student, and obtains the knowledge scope and accumulated learning hours of the current student's personalized learning in real time, and then judges whether the examination conditions are met according to the knowledge scope and accumulated learning hours of the current student's personalized learning; if so, based on the basic information and learning purpose information of the current student, as well as the knowledge scope of personalized learning, and according to the preset test customization algorithm, matching test questions are extracted from the test question database, and corresponding test papers are generated; finally, the corresponding test papers are pushed to the front-end device of the online student to conduct an online customized test for the current student. The present invention can make an online customized test plan adapted to each student's personalized learning status, thereby ensuring that each student can effectively receive knowledge and improve their own grades or learning skills.

[0094] According to an embodiment of the present invention, obtaining the basic information and learning purpose information of the current student specifically includes:

[0095] Receive registration requests from current students;

[0096] Generate a corresponding registration information filling form according to the current student's registration request and feed it back to the current student's front-end device, and the registration information filling form at least includes the current student's basic information (such as age, education, work status, etc.) and learning purpose information (such as the purpose of learning English is the middle school entrance examination, college entrance examination, postgraduate entrance examination, IELTS, overseas exchange, etc.);

[0097] Receive the registration information form completed by the current student and create a learning file for the current student.

[0098] like Figure 2 As shown, according to an embodiment of the present invention, judging whether the examination conditions are met according to the knowledge scope of the current student's personalized learning and the accumulated learning hours specifically includes:

[0099] S202, constructing a knowledge mastery prediction model;

[0100] S204, training and optimizing the knowledge mastery degree prediction model through sample data to obtain an optimized knowledge mastery degree prediction model;

[0101] S206, using the knowledge scope and accumulated hours of the current student's personalized learning to predict the current student's knowledge mastery through a knowledge mastery prediction model;

[0102] S208, determining whether the predicted knowledge mastery level of the current student is greater than a first preset threshold, and if so, the examination conditions are met.

[0103] It is understandable that if the current students learn a large range of knowledge in a very short time, then the mastery of the knowledge range is very low. For example, some students only quickly browse online and cannot actually master the knowledge. At this time, the predicted knowledge mastery is very low. If it is lower than the first preset threshold, it is considered that the test conditions cannot be met. Therefore, the premise of generating a customized test plan is that the students need to reach the minimum mastery level, that is, the first preset threshold. Otherwise, it will cause unnecessary tests, waste students' time, and fail to achieve the effect of the test.

[0104] According to an embodiment of the present invention, after the knowledge scope and accumulated learning hours of the current student's personalized learning are used to predict the current student's knowledge mastery through the knowledge mastery prediction model, the method further includes:

[0105] Acquire history learning data, each history learning data at least includes basic information and learning purpose information of history students, history learning knowledge scope, accumulated history learning class hours, history test paper difficulty coefficient, and history test scores;

[0106] Based on each history learning data, the basic information and learning purpose information of the history students, the knowledge scope of history learning, and the accumulated class hours of history learning are calculated to obtain the characteristic value of each history student;

[0107] Perform feature calculation on the basic information and learning purpose information of the current student, the knowledge scope of personalized learning, and the accumulated learning hours to obtain the feature value of the current student;

[0108] Based on each historical learning data, the approximation between the feature value of each historical student and the feature value of the current student is calculated;

[0109] Determine whether the calculated approximation is greater than a second preset threshold, and if so, add the corresponding historical learning data to the approximate feature library;

[0110] Based on each historical learning data in the approximate feature library, the corresponding historical learning knowledge scope and the accumulated historical learning class hours are input into the knowledge mastery degree prediction model to obtain the knowledge mastery degree prediction value of each history student;

[0111] Based on each historical learning data in the approximate feature library, the corresponding historical test paper difficulty coefficient and historical test score are calculated through a preset replacement algorithm to obtain the actual value of each history student's knowledge mastery level;

[0112] Based on each historical learning data in the approximate feature library, the actual value of each history student's knowledge mastery is subtracted from the predicted value of each history student's knowledge mastery to obtain a knowledge mastery vector difference;

[0113] Adding the knowledge mastery degree vector difference values ​​calculated based on each historical learning data in the approximate feature library to obtain the knowledge mastery degree vector difference sum;

[0114] Divide the sum of the knowledge mastery degree vector differences by the total number of historical learning data in the approximate feature library to obtain the average difference of the knowledge mastery degree vector;

[0115] The predicted knowledge mastery level of the current student and the average difference between the knowledge mastery level vectors are vector summed to obtain the corrected knowledge mastery level of the current student.

[0116] It can be understood that the present invention obtains the predicted value of the knowledge mastery level of each history student based on the analysis and calculation of multiple historical learning data, and then performs difference calculation and analysis between the actual value of the knowledge mastery level of each history student and the predicted value of the knowledge mastery level of each history student, so as to obtain the correction value of the prediction model, and corrects the knowledge mastery level of the current student predicted by the model through the correction value, so as to predict the current student's knowledge mastery level more accurately, which is conducive to the subsequent precise customization of test questions for the current student.

[0117] According to an embodiment of the present invention, the corresponding history test paper difficulty coefficient and history test score are calculated by a preset replacement algorithm to obtain the actual value of the knowledge mastery level of each history student, specifically including:

[0118] Multiply the corresponding history test paper difficulty coefficient by the history test score, and divide the product by the total score of the history test paper to obtain the actual value of each history student's knowledge mastery.

[0119] It can be understood that the average difficulty coefficient of the test paper is 1. If the difficulty coefficient is higher than 1, it means that the test paper is difficult and the test scores are generally low. At this time, the test scores of history students should be multiplied by the difficulty coefficient higher than 1, which is the actual value of the knowledge mastery of history students. If it is lower than 1, it means that the test paper is easy and the test scores are generally high. At this time, the test scores of history students should be multiplied by the difficulty coefficient lower than 1, which is the actual value of the knowledge mastery of history students.

[0120] According to an embodiment of the present invention, based on each historical learning data, the basic information and learning purpose information of the history student, the knowledge scope of the history learning, and the accumulated class hours of the history learning are feature calculated to obtain the feature value of each history student; the basic information and learning purpose information of the current student, the knowledge scope of personalized learning, and the accumulated class hours of learning are feature calculated to obtain the feature value of the current student; based on each historical learning data, the feature value of each history student is similar to the feature value of the current student, specifically including:

[0121] Based on each piece of history learning data, the age information and educational background information of the history student are extracted from the basic information of the history student, the age information and educational background information of the history student are replaced with the age grade and educational background grade of the history student through a preset age grade table and a preset educational background grade table, respectively, and the learning purpose information of the history student is replaced with the learning purpose grade of the history student through a preset learning purpose grade table, the age grade of the history student is multiplied by the learning purpose grade of the history student to obtain a first product, and the educational background grade of the history student is multiplied by the learning purpose grade of the history student to obtain a second product, the first product and the second product are added to obtain a first value, the age grade of the history student and the educational background grade of the history student are added to obtain a second value, and the first value is divided by the second value to obtain an attribute characteristic value of the history student;

[0122] Based on each historical learning data, the number of corresponding knowledge points is counted according to the historical learning knowledge scope, and the number of corresponding knowledge points is divided by the accumulated historical learning hours to obtain the learning characteristic value of the history student;

[0123] Extracting the age information and educational level information of the current student from the basic information of the current student, replacing the age information and educational level information of the current student into the age level and educational level of the current student through a preset age level table and a preset educational level table, respectively, and replacing the learning purpose information of the current student into the learning purpose level of the current student through a preset learning purpose level table, multiplying the age level of the current student by the learning purpose level of the current student to obtain a third product, and multiplying the educational level of the current student by the learning purpose level of the current student to obtain a fourth product, adding the third product and the fourth product to obtain a third value, adding the age level of the current student and the educational level of the current student to obtain a fourth value, and dividing the third value by the fourth value to obtain an attribute characteristic value of the current student;

[0124] Count the number of corresponding knowledge points from the knowledge scope of the current student's personalized learning, and divide the number of corresponding knowledge points by the current student's accumulated learning hours to obtain the current student's learning characteristic value;

[0125] Based on each historical learning data, the attribute feature value of the historical student and the attribute feature value of the current student are approximated to obtain the attribute feature approximation based on each historical learning data;

[0126] Based on each historical learning data, the learning feature value of the historical students and the learning feature value of the current students are approximated to obtain the learning feature approximation based on each historical learning data;

[0127] The influence weights of preset attribute features and learning features on judging whether the characteristics of historical students and current students are similar are different;

[0128] Based on each historical learning data, the corresponding attribute feature approximation is multiplied by the influence weight of the attribute feature to obtain the attribute feature weight approximation, and the corresponding learning feature approximation is multiplied by the influence weight of the learning feature to obtain the learning feature weight approximation;

[0129] Based on each historical learning data, the corresponding attribute feature weight approximation is added to the corresponding learning feature weight approximation to obtain a fifth value, and the influence weight of the attribute feature is added to the influence weight of the learning feature to obtain a sixth value;

[0130] Based on each historical learning data, the calculated fifth value is divided by the sixth value to obtain the approximation between the characteristic value of the corresponding historical student and the characteristic value of the current student.

[0131] It can be understood that the present invention calculates characteristic values ​​and calculates the approximation between the characteristic values ​​of historical students and the characteristic values ​​of current students, thereby screening out historical learning data with a higher degree of similarity to the characteristics of the current student, and then calculates correction values ​​based on the historical learning data with a higher degree of characteristic similarity, thereby obtaining correction values ​​that are more suitable for the current student's situation, further facilitating obtaining a more accurate level of knowledge mastery of the current student.

[0132] According to a specific embodiment of the present invention, the approximation calculation of the attribute feature values ​​of the historical students and the attribute feature values ​​of the current student is performed, specifically including:

[0133] Subtract the attribute feature value of the current student from the attribute feature value of the previous student, and take the absolute value of the difference;

[0134] The absolute value of the difference is divided by the attribute characteristic value of the current student to obtain the seventh value;

[0135] Subtract the seventh value from 1 to obtain the approximation between the attribute characteristic value of the historical student and the attribute characteristic value of the current student.

[0136] It can be understood that the calculation method of the approximation between the learning characteristic values ​​of the historical students and the learning characteristic values ​​of the current students is the same as the calculation method of the approximation of the attribute characteristic values ​​mentioned above, which will not be repeated here.

[0137] According to an embodiment of the present invention, extracting matching test questions from a test question database according to a preset test customization algorithm and generating corresponding test papers specifically includes:

[0138] According to the knowledge scope of personalized learning, corresponding test questions are extracted from the test question database and added to the preliminary test question bank;

[0139] There are m test questions in the preset preliminary test question bank, and the number of test questions required for an examination paper is n, and m is greater than n;

[0140] Randomly select n test questions from m test questions to form a candidate test paper, thus forming 1. Candidate examination papers and add them to the candidate examination paper database;

[0141] Compare each candidate test paper in the candidate test paper library with the remaining candidate test papers in pairs, and determine which candidate test paper is more suitable for the current student's learning purpose information. If the former is more suitable, add 1 point to the former's learning purpose matching item;

[0142] Compare each candidate test paper in the candidate test paper database with the remaining candidate test papers in pairs, and determine which candidate test paper has a test difficulty that is more suitable for the current student's knowledge level. If the former is more suitable, add 1 point to the former's test difficulty matching item;

[0143] Compare each candidate test paper in the candidate test paper database with the remaining candidate test papers in pairs, and determine which candidate test paper has a more reasonable combination of test questions of different difficulty levels. If the former is more reasonable, add 1 point to the reasonableness of the former's test question combination;

[0144] After all candidate test papers in the candidate test paper database have been compared in pairs, the scores of each candidate test paper in the learning purpose matching item, the test question difficulty matching item, and the test question reasonable matching item are counted;

[0145] The influence weights of the preset learning purpose matching items, test question difficulty matching items, and test question reasonable matching items on the matching test papers are different;

[0146] Multiply the score of each candidate test paper in the learning purpose matching item by the corresponding influence weight to obtain the weight score of the learning purpose matching item; multiply the score of each candidate test paper in the test question difficulty matching item by the corresponding influence weight to obtain the weight score of the test question difficulty matching item; multiply the score of each candidate test paper in the test question reasonable matching item by the corresponding influence weight to obtain the weight score of the test question reasonable matching item;

[0147] The weighted score of the learning purpose matching item, the weighted score of the test question difficulty matching item, and the weighted score of the test question reasonable combination item of each candidate test paper are added together to obtain the matching score of each candidate test paper;

[0148] Each candidate test paper is sorted based on its matching score, and the candidate test paper with the highest matching score is used as the corresponding test paper.

[0149] According to an embodiment of the present invention, each candidate test paper in the candidate test paper library is compared with the remaining other candidate test papers in pairs, and it is determined which candidate test paper is more suitable for the learning purpose information of the current student, specifically including:

[0150] Each exam question in the preset exam database comes with a degree of fit suitable for each learning purpose;

[0151] Based on each test question of the two compared candidate test papers, the degree of fit of the corresponding test question with the current student's learning purpose information is obtained from the accompanying information;

[0152] Count whether the two candidate examination papers have questions with a degree of fit less than a third preset threshold. If the former candidate examination paper does not have such questions, but the latter candidate examination paper does, the former candidate examination paper is judged to be more suitable, and 1 point is added to the former learning purpose matching item; if the former candidate examination paper has such questions, but the latter candidate examination paper does not, the latter candidate examination paper is judged to be more suitable, and no points are added; if both the former candidate examination paper and the latter candidate examination paper have such questions, it is judged that neither of them is suitable, and no points are added; if neither the former candidate examination paper nor the latter candidate examination paper has such questions, proceed to the next step;

[0153] Based on the former candidate test paper, the degree of fit of each test question with the learning purpose information of the current student is multiplied by its corresponding score to obtain the fit score of each test question; the fit scores of all test questions in the former candidate test paper are added to obtain the sum of the fit scores, and the sum of the fit scores is divided by the total score of the former candidate test paper to obtain the learning purpose fit of the former candidate test paper;

[0154] Based on the latter candidate test paper, the degree of fit of each test question with the learning purpose information of the current student is multiplied by its corresponding score to obtain the fit score of each test question; the fit scores of all test questions in the latter candidate test paper are added to obtain the sum of the fit scores, and the sum of the fit scores is divided by the total score of the latter candidate test paper to obtain the learning purpose fit of the latter candidate test paper;

[0155] Compare the degree of fit of the learning purpose of the former candidate test paper with the degree of fit of the learning purpose of the latter candidate test paper. If the former is greater than the latter, the former is judged to be more suitable, and 1 point is added to the learning purpose match of the former. Otherwise, no point is added.

[0156] According to an embodiment of the present invention, each candidate test paper in the candidate test paper library is compared with the remaining other candidate test papers in pairs, and it is determined which candidate test paper has a test question difficulty that is more suitable for the current student's knowledge mastery, specifically including:

[0157] Each question in the preset question database is accompanied by a difficulty value;

[0158] Based on each test question of the two compared candidate test papers, the difficulty value of the corresponding test question is obtained from the accompanying information;

[0159] Count whether the two candidate test papers have test questions with a difficulty value less than the fourth preset threshold or greater than the fifth preset threshold. If the former candidate test paper does not have such a test question, but the latter candidate test paper does, then the former candidate test paper is judged to be more suitable, and 1 point is added to the former test question difficulty matching item; if the former candidate test paper has such a test question, but the latter candidate test paper does not, then the latter candidate test paper is judged to be more suitable, and no point is added; if both the former candidate test paper and the latter candidate test paper have such a test question, then it is judged that neither of them is suitable, and no point is added; if neither the former candidate test paper nor the latter candidate test paper has such a test question, then proceed to the next step;

[0160] Based on the former candidate test paper, multiply the difficulty value of each test question by its corresponding score to obtain the difficulty score of each test question; add the difficulty scores of all test questions in the former candidate test paper to obtain the sum of the difficulty scores, and divide the sum of the difficulty scores by the total score of the former candidate test paper to obtain the test question difficulty value of the former candidate test paper;

[0161] Based on the latter candidate test paper, multiply the difficulty value of each test question by its corresponding score to obtain the difficulty score of each test question; add the difficulty scores of all test questions in the latter candidate test paper to obtain the sum of the difficulty scores, and divide the sum of the difficulty scores by the total score of the latter candidate test paper to obtain the test question difficulty value of the latter candidate test paper;

[0162] The difficulty of the pre-set test papers is increased by 20% based on the students' knowledge level to make the test papers most suitable for them.

[0163] Determine which one of the difficulty values ​​of the former candidate test paper and the latter candidate test paper is closer to the current student's knowledge level, and add 20% difficulty. If the former is closer, the former is judged to be more suitable, and 1 point is added to the former's test difficulty match item. Otherwise, no point is added.

[0164] According to a specific embodiment of the present invention, each candidate test paper in the candidate test paper library is compared with the remaining other candidate test papers in pairs, and it is determined which candidate test paper has a more reasonable ratio of question difficulty, specifically including:

[0165] Statistically calculate the score proportions of difficult questions, general questions and simple questions in the former test paper, and the score proportions of difficult questions, general questions and simple questions in the latter test paper;

[0166] If the score proportion of the difficult questions in the former test paper is not greater than the sixth preset threshold, and the score proportion of the simple questions is not greater than the seventh preset threshold; at the same time, the score proportion of the difficult questions in the latter test paper is greater than the sixth preset threshold, or the score proportion of the simple questions is greater than the seventh preset threshold; then it is judged that the difficulty level matching proportion of the candidate test paper of the former is more reasonable, and 1 point is added to the reasonable matching item of the former test paper; if the score proportion of the difficult questions in both the former and the latter test papers is not greater than the sixth preset threshold, and the score proportion of the simple questions is not greater than the seventh preset threshold, then proceed to the next step;

[0167] The optimal score ratio of difficult questions, general questions and simple questions in the preset examination paper is K1:K2:K3;

[0168] Statistically calculate the score proportions of difficult questions, general questions and simple questions in the former test paper, and the score proportions of difficult questions, general questions and simple questions in the latter test paper;

[0169] Determine whether the score ratio of the difficult questions, general questions and simple questions in the former test paper is closer to the optimal score ratio K1:K2:K3 than the score ratio of the difficult questions, general questions and simple questions in the latter test paper;

[0170] If the former is closer, 1 point will be added to the reasonable collocation of the former question; otherwise, no point will be added.

[0171] Figure 3 A block diagram of an online test customization system based on personalized learning according to the present invention is shown.

[0172] like Figure 3 As shown, the second aspect of the present invention further proposes an online test customization system 3 based on personalized learning, including a memory 31 and a processor 32, wherein the memory includes an online test customization method program based on personalized learning, and when the online test customization method program based on personalized learning is executed by the processor, the following steps are implemented:

[0173] Obtain basic information of current students and information on their learning objectives;

[0174] Obtain the knowledge scope and accumulated hours of current students' personalized learning in real time;

[0175] Determine whether the student meets the examination requirements based on the student's current personalized learning knowledge scope and accumulated learning hours;

[0176] If yes, based on the current student's basic information, learning purpose information, and the knowledge scope of personalized learning, matching test questions are extracted from the test question database according to the preset test customization algorithm, and the corresponding test paper is generated;

[0177] The corresponding test papers are pushed to the front-end device of the current student to conduct an online customized test for the current student.

[0178] According to an embodiment of the present invention, judging whether the examination conditions are met based on the knowledge scope of the current student's personalized learning and the accumulated learning hours specifically includes:

[0179] Construct a knowledge mastery prediction model;

[0180] The knowledge mastery degree prediction model is trained and optimized through sample data to obtain an optimized knowledge mastery degree prediction model;

[0181] The knowledge scope and accumulated hours of the current students' personalized learning are used to predict the current students' knowledge mastery through the knowledge mastery prediction model;

[0182] It is determined whether the predicted knowledge mastery level of the current student is greater than a first preset threshold value. If so, the examination conditions are met.

[0183] The online test customization method and system based on personalized learning proposed by the present invention first obtains the basic information and learning purpose information of the current student, and obtains the knowledge scope of the current student's personalized learning and the accumulated learning hours in real time, and then judges whether the test conditions are met according to the knowledge scope of the current student's personalized learning and the accumulated learning hours; if so, based on the basic information and learning purpose information of the current student, as well as the knowledge scope of personalized learning, and according to the preset test customization algorithm, matching test questions are extracted from the test question database, and corresponding test papers are generated; finally, the corresponding test papers are pushed to the front-end device of the online student to conduct an online customized test for the current student. The present invention can make an online customized test plan that is suitable for each student based on his or her personalized learning status, thereby ensuring that each student can effectively receive knowledge and improve his or her own grades or learning skills.

[0184] In the several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as: multiple units or components can be combined, or can be integrated into another system, or some features can be ignored or not executed. In addition, the coupling, direct coupling, or communication connection between the components shown or discussed can be through some interfaces, and the indirect coupling or communication connection of the devices or units can be electrical, mechanical or other forms.

[0185] The units described above as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units; they may be located in one place or distributed on multiple network units; some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0186] In addition, all functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may be separately used as a unit, or two or more units may be integrated into one unit; the above-mentioned integrated units may be implemented in the form of hardware or in the form of hardware plus software functional units.

[0187] Those skilled in the art can understand that: all or part of the steps of implementing the above method embodiments can be completed by hardware related to program instructions, and the aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it executes the steps of the above method embodiments; and the aforementioned storage medium includes: mobile storage devices, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), disks or optical disks, and other media that can store program codes.

[0188] Alternatively, if the above-mentioned integrated unit of the present invention is implemented in the form of a software function module and sold or used as an independent product, it can also be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the embodiment of the present invention can be essentially or partly reflected in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the methods described in each embodiment of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as mobile storage devices, ROM, RAM, magnetic disks or optical disks.

[0189] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. An online test customization method based on personalized learning, characterized in that: The method comprises: Obtain basic information of current students and information on their learning objectives; Obtain the knowledge scope and accumulated hours of current students' personalized learning in real time; Determine whether the student meets the examination requirements based on the student's current personalized learning knowledge scope and accumulated learning hours; If yes, based on the current student's basic information, learning purpose information, and the knowledge scope of personalized learning, matching test questions are extracted from the test question database according to the preset test customization algorithm, and the corresponding test paper is generated; The corresponding test papers are pushed to the front-end device of the current student to conduct an online customized test for the current student.

2. The online test customization method based on personalized learning according to claim 1, characterized in that: Based on the knowledge scope of the current personalized learning and the accumulated hours of learning, determine whether the examination conditions are met, including: Construct a knowledge mastery prediction model; The knowledge mastery degree prediction model is trained and optimized through sample data to obtain an optimized knowledge mastery degree prediction model; The knowledge scope and accumulated hours of the current students' personalized learning are used to predict the current students' knowledge mastery through the knowledge mastery prediction model; It is determined whether the predicted knowledge mastery level of the current student is greater than a first preset threshold value. If so, the examination conditions are met.

3. The online test customization method based on personalized learning according to claim 2 is characterized in that: After the knowledge scope and accumulated hours of learning of the current student are used to predict the knowledge mastery of the current student through the knowledge mastery prediction model, the method further includes: Acquire history learning data, each of which includes at least basic information and learning purpose information of history students, history learning knowledge scope, accumulated history learning hours, history test paper difficulty coefficient, and history test scores; Based on each history learning data, the basic information and learning purpose information of the history students, the knowledge scope of history learning, and the accumulated class hours of history learning are calculated to obtain the characteristic value of each history student; Perform feature calculation on the basic information and learning purpose information of the current student, the knowledge scope of personalized learning, and the accumulated learning hours to obtain the feature value of the current student; Based on each historical learning data, the approximation between the feature value of each historical student and the feature value of the current student is calculated; Determine whether the calculated approximation is greater than a second preset threshold, and if so, add the corresponding historical learning data to the approximate feature library; Based on each historical learning data in the approximate feature library, the corresponding historical learning knowledge scope and the accumulated historical learning class hours are input into the knowledge mastery degree prediction model to obtain the knowledge mastery degree prediction value of each history student; Based on each historical learning data in the approximate feature library, the corresponding historical test paper difficulty coefficient and historical test score are calculated through a preset replacement algorithm to obtain the actual value of each history student's knowledge mastery level; Based on each historical learning data in the approximate feature library, the actual value of each history student's knowledge mastery is subtracted from the predicted value of each history student's knowledge mastery to obtain a knowledge mastery vector difference; Adding the knowledge mastery degree vector difference values ​​calculated based on each historical learning data in the approximate feature library to obtain the knowledge mastery degree vector difference sum; Divide the sum of the knowledge mastery degree vector differences by the total number of historical learning data in the approximate feature library to obtain the average difference of the knowledge mastery degree vector; The predicted knowledge mastery level of the current student and the average difference between the knowledge mastery level vectors are vector summed to obtain the corrected knowledge mastery level of the current student.

4. The online test customization method based on personalized learning according to claim 3 is characterized in that: The corresponding history test paper difficulty coefficient and history test score are calculated through a preset replacement algorithm to obtain the actual value of each history student's knowledge mastery, including: Multiply the corresponding history test paper difficulty coefficient by the history test score, and divide the product by the total score of the history test paper to obtain the actual value of each history student's knowledge mastery.

5. The online test customization method based on personalized learning according to claim 3 is characterized in that: Based on each history learning data, the basic information and learning purpose information of the history students, the knowledge scope of history learning, and the accumulated class hours of history learning are calculated to obtain the characteristic value of each history student; Perform feature calculation on the basic information and learning purpose information of the current student, the knowledge scope of personalized learning, and the accumulated learning hours to obtain the feature value of the current student; Based on each historical learning data, the characteristic value of each historical student is approximated with the characteristic value of the current student, specifically including: Based on each piece of history learning data, the age information and educational background information of the history student are extracted from the basic information of the history student, the age information and educational background information of the history student are replaced with the age grade and educational background grade of the history student through a preset age grade table and a preset educational background grade table, respectively, and the learning purpose information of the history student is replaced with the learning purpose grade of the history student through a preset learning purpose grade table, the age grade of the history student is multiplied by the learning purpose grade of the history student to obtain a first product, and the educational background grade of the history student is multiplied by the learning purpose grade of the history student to obtain a second product, the first product and the second product are added to obtain a first value, the age grade of the history student and the educational background grade of the history student are added to obtain a second value, and the first value is divided by the second value to obtain an attribute characteristic value of the history student; Based on each historical learning data, the number of corresponding knowledge points is counted according to the historical learning knowledge scope, and the number of corresponding knowledge points is divided by the accumulated historical learning hours to obtain the learning characteristic value of the history student; Extracting the age information and educational level information of the current student from the basic information of the current student, replacing the age information and educational level information of the current student into the age level and educational level of the current student through a preset age level table and a preset educational level table, respectively, and replacing the learning purpose information of the current student into the learning purpose level of the current student through a preset learning purpose level table, multiplying the age level of the current student by the learning purpose level of the current student to obtain a third product, and multiplying the educational level of the current student by the learning purpose level of the current student to obtain a fourth product, adding the third product and the fourth product to obtain a third value, adding the age level of the current student and the educational level of the current student to obtain a fourth value, and dividing the third value by the fourth value to obtain an attribute characteristic value of the current student; Count the number of corresponding knowledge points from the knowledge scope of the current student's personalized learning, and divide the number of corresponding knowledge points by the current student's accumulated learning hours to obtain the current student's learning characteristic value; Based on each historical learning data, the attribute feature value of the historical student and the attribute feature value of the current student are approximated to obtain the attribute feature approximation based on each historical learning data; Based on each historical learning data, the learning feature value of the historical students and the learning feature value of the current students are approximated to obtain the learning feature approximation based on each historical learning data; The influence weights of preset attribute features and learning features on judging whether the characteristics of historical students and current students are similar are different; Based on each historical learning data, the corresponding attribute feature approximation is multiplied by the influence weight of the attribute feature to obtain the attribute feature weight approximation, and the corresponding learning feature approximation is multiplied by the influence weight of the learning feature to obtain the learning feature weight approximation; Based on each historical learning data, the corresponding attribute feature weight approximation is added to the corresponding learning feature weight approximation to obtain a fifth value, and the influence weight of the attribute feature is added to the influence weight of the learning feature to obtain a sixth value; Based on each historical learning data, the calculated fifth value is divided by the sixth value to obtain the approximation between the characteristic value of the corresponding historical student and the characteristic value of the current student.

6. The online test customization method based on personalized learning according to claim 2, characterized in that: According to the preset test customization algorithm, matching test questions are extracted from the test question database and the corresponding test papers are generated, including: According to the knowledge scope of personalized learning, corresponding test questions are extracted from the test question database and added to the preliminary test question bank; There are m test questions in the preset preliminary test question bank, and the number of test questions required for an examination paper is n, and m is greater than n; Randomly select n test questions from m test questions to form a candidate test paper, thus forming 1. Candidate examination papers and add them to the candidate examination paper database; Compare each candidate test paper in the candidate test paper library with the remaining candidate test papers in pairs, and determine which candidate test paper is more suitable for the current student's learning purpose information. If the former is more suitable, add 1 point to the former's learning purpose matching item; Compare each candidate test paper in the candidate test paper database with the remaining candidate test papers in pairs, and determine which candidate test paper has a test difficulty that is more suitable for the current student's knowledge level. If the former is more suitable, add 1 point to the former's test difficulty matching item; Compare each candidate test paper in the candidate test paper database with the remaining candidate test papers in pairs, and determine which candidate test paper has a more reasonable combination of test questions of different difficulty levels. If the former is more reasonable, add 1 point to the reasonableness of the former's test question combination; After all candidate test papers in the candidate test paper database have been compared in pairs, the scores of each candidate test paper in the learning purpose matching item, the test question difficulty matching item, and the test question reasonable matching item are counted; The influence weights of the preset learning purpose matching items, test question difficulty matching items, and test question reasonable matching items on the matching test papers are different; Multiply the score of each candidate test paper in the learning purpose matching item by the corresponding influence weight to obtain the weight score of the learning purpose matching item; multiply the score of each candidate test paper in the test question difficulty matching item by the corresponding influence weight to obtain the weight score of the test question difficulty matching item; multiply the score of each candidate test paper in the test question reasonable matching item by the corresponding influence weight to obtain the weight score of the test question reasonable matching item; The weighted score of the learning purpose matching item, the weighted score of the test question difficulty matching item, and the weighted score of the test question reasonable combination item of each candidate test paper are added together to obtain the matching score of each candidate test paper; Each candidate test paper is sorted based on its matching score, and the candidate test paper with the highest matching score is used as the corresponding test paper.

7. The online test customization method based on personalized learning according to claim 6, characterized in that: Each candidate test paper in the candidate test paper library is compared with the remaining candidate test papers in pairs, and it is determined which candidate test paper is more suitable for the current student's learning purpose information, including: Each exam question in the preset exam database comes with a degree of fit suitable for each learning purpose; Based on each test question of the two compared candidate test papers, the degree of fit of the corresponding test question with the current student's learning purpose information is obtained from the accompanying information; Count whether the two candidate examination papers have questions with a degree of fit less than a third preset threshold. If the former candidate examination paper does not have such questions, but the latter candidate examination paper does, the former candidate examination paper is judged to be more suitable, and 1 point is added to the former learning purpose matching item; if the former candidate examination paper has such questions, but the latter candidate examination paper does not, the latter candidate examination paper is judged to be more suitable, and no points are added; if both the former candidate examination paper and the latter candidate examination paper have such questions, it is judged that neither of them is suitable, and no points are added; if neither the former candidate examination paper nor the latter candidate examination paper has such questions, proceed to the next step; Based on the former candidate test paper, the degree of fit of each test question with the learning purpose information of the current student is multiplied by its corresponding score to obtain the fit score of each test question; the fit scores of all test questions in the former candidate test paper are added to obtain the sum of the fit scores, and the sum of the fit scores is divided by the total score of the former candidate test paper to obtain the learning purpose fit of the former candidate test paper; Based on the latter candidate test paper, the degree of fit of each test question with the learning purpose information of the current student is multiplied by its corresponding score to obtain the fit score of each test question; the fit scores of all test questions in the latter candidate test paper are added to obtain the sum of the fit scores, and the sum of the fit scores is divided by the total score of the latter candidate test paper to obtain the learning purpose fit of the latter candidate test paper; Compare the degree of fit of the learning purpose of the former candidate test paper with the degree of fit of the learning purpose of the latter candidate test paper. If the former is greater than the latter, the former is judged to be more suitable, and 1 point is added to the learning purpose match of the former. Otherwise, no point is added.

8. The online test customization method based on personalized learning according to claim 6, characterized in that: Each candidate test paper in the candidate test paper database is compared with the remaining candidate test papers, and the difficulty of the test questions of which candidate test paper is more suitable for the current students' knowledge mastery level is determined, including: Each question in the preset question database is accompanied by a difficulty value; Based on each test question of the two compared candidate test papers, the difficulty value of the corresponding test question is obtained from the accompanying information; Count whether the two candidate test papers have test questions with a difficulty value less than the fourth preset threshold or greater than the fifth preset threshold. If the former candidate test paper does not have such a test question, but the latter candidate test paper does, then the former candidate test paper is judged to be more suitable, and 1 point is added to the former test question difficulty matching item; if the former candidate test paper has such a test question, but the latter candidate test paper does not, then the latter candidate test paper is judged to be more suitable, and no point is added; if both the former candidate test paper and the latter candidate test paper have such a test question, then it is judged that neither of them is suitable, and no point is added; if neither the former candidate test paper nor the latter candidate test paper has such a test question, then proceed to the next step; Based on the former candidate test paper, multiply the difficulty value of each test question by its corresponding score to obtain the difficulty score of each test question; add the difficulty scores of all test questions in the former candidate test paper to obtain the sum of the difficulty scores, and divide the sum of the difficulty scores by the total score of the former candidate test paper to obtain the test question difficulty value of the former candidate test paper; Based on the latter candidate test paper, multiply the difficulty value of each test question by its corresponding score to obtain the difficulty score of each test question; add the difficulty scores of all test questions in the latter candidate test paper to obtain the sum of the difficulty scores, and divide the sum of the difficulty scores by the total score of the latter candidate test paper to obtain the test question difficulty value of the latter candidate test paper; The difficulty of the questions in the preset examination papers is increased by 20% based on the students' knowledge level to make the most appropriate examination papers; Determine which one of the difficulty values ​​of the former candidate test paper and the latter candidate test paper is closer to the current student's knowledge level, and add 20% difficulty. If the former is closer, the former is judged to be more suitable, and 1 point is added to the former's test difficulty match item. Otherwise, no point is added.

9. An online test customization system based on personalized learning, characterized in that: The system comprises a memory and a processor, wherein the memory comprises an online test customization method program based on personalized learning, and when the online test customization method program based on personalized learning is executed by the processor, the following steps are implemented: Obtain basic information of current students and information on their learning objectives; Obtain the knowledge scope and accumulated hours of current students' personalized learning in real time; Determine whether the student meets the examination requirements based on the student's current personalized learning knowledge scope and accumulated learning hours; If yes, based on the current student's basic information, learning purpose information, and the knowledge scope of personalized learning, matching test questions are extracted from the test question database according to the preset test customization algorithm, and the corresponding test paper is generated; The corresponding test papers are pushed to the front-end device of the current student to conduct an online customized test for the current student.

10. The online test customization system based on personalized learning according to claim 9, characterized in that: Based on the knowledge scope of the current personalized learning and the accumulated hours of learning, determine whether the examination conditions are met, including: Construct a knowledge mastery prediction model; The knowledge mastery degree prediction model is trained and optimized through sample data to obtain an optimized knowledge mastery degree prediction model; The knowledge scope and accumulated hours of the current students' personalized learning are used to predict the current students' knowledge mastery through the knowledge mastery prediction model; It is determined whether the predicted knowledge mastery level of the current student is greater than a first preset threshold value. If so, the examination conditions are met.

Images