Learning ability evaluation and improvement system based on large model
By using a large-scale model-based learning ability assessment and improvement system, which collects and analyzes users' historical answer information and calculates scores and proficiency, the system solves the problem that traditional exams cannot accurately assess learning ability, and enables more realistic learning ability assessment and the provision of personalized learning resources.
Patent Information
- Application Number
- CN202411150801.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-21
- Publication Date
- 2024-11-19
- Estimated Expiration
- 2044-08-21
AI Technical Summary
Traditional tests cannot accurately assess a user's learning ability, especially since they cannot identify the randomness of a user's understanding of the same knowledge point, resulting in scores that do not truly reflect the user's learning ability.
The learning ability assessment and improvement system based on a large model collects users' historical answer information, calculates scores and proficiency, uses data processing and assessment modules to evaluate learning ability, and provides personalized learning materials through the improvement model to strengthen weak knowledge points.
It enables more accurate assessment of users' learning abilities, identifies weak knowledge points and provides personalized learning resources, forming a closed loop of learning assessment and improvement, and reducing the impact of chance on scores.
Smart Images

Figure CN118982293A_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the technical field of learning ability assessment, and specifically is a learning ability assessment and improvement system based on a large model. Background Art
[0002] Traditional exams use a single score to evaluate a user's learning ability. They cannot tell what specific knowledge a user has or has not mastered, nor can they find out why a user made a mistake in order to remedy the situation. For users with the same score, it is even more impossible to find out the differences in their cognitive states and knowledge structures. The information provided by traditional exams is no longer suitable for individual development needs.
[0003] In order to obtain test results that better reflect the user's learning ability, the existing patent with publication number CN114491050A discloses a learning ability assessment method based on cognitive diagnosis, including: obtaining the user's answer information, pre-processing the user's answer information by labeling, and obtaining labeled answer information and unlabeled answer information; clustering the unlabeled answer information according to the labeled answer information to obtain labels for all answer information; inputting all answer information and its labels into a cognitive diagnosis model, and the cognitive diagnosis model outputs the user's probability of answering correctly, and assessing the user's learning ability based on the user's probability of answering correctly.
[0004] Based on the above, in essence, the existing technology obtains the user's answer information, uses the relevant information in the answer information to analyze the user's answer status during the actual test, and thus comprehensively calculates the user's real score, and does not rely on the traditional paper score. In the actual calculation process, the existing technology can only obtain the user's historical real score based on the historical answer information analysis, and simultaneously complete the learning ability assessment of all users. However, in the user's answering process, the questions corresponding to the same knowledge point are not unique. Therefore, when the user answers multiple questions on the same knowledge point, if there are both wrong and correct situations, there must be a certain degree of randomness in the user's answering of the questions. If the existence of randomness is not taken into account, it may cause the actual real scores of some users to be too high, which cannot truly reflect the user's learning ability.
[0005] To this end, the present invention provides a learning ability assessment and improvement system based on a large model. Summary of the invention
[0006] In order to make up for the deficiencies of the prior art, at least one technical problem raised in the background technology is solved.
[0007] The technical solution adopted by the present invention to solve the technical problem is: a learning ability evaluation and improvement system based on a large model described in the present invention comprises:
[0008] Data collection module, used to collect users' historical answer information;
[0009] The calculation module is used to calculate the user's historical answer scores based on the historical answer information;
[0010] The data processing module obtains the user's historical score rate based on historical answer scores and historical answer information;
[0011] The evaluation module evaluates the user's historical learning ability based on the historical score rate;
[0012] The improvement model outputs the user's historical answering proficiency based on the input user's historical answering information, and improves the user's learning ability based on the user's historical answering proficiency.
[0013] Preferably, the answer information includes: answer accuracy A, answer times C and answer time T; when the user answers the question accurately, A=1, otherwise A=0.
[0014] Preferably, the method by which the calculation module obtains the historical answer score according to the historical answer information is:
[0015] Get the user's answer information;
[0016] According to the formula:
[0017]
[0018] Among them, P represents the basic score, represents the accuracy of the answer to any question i, represents the number of answers to any question i; represents the historical answer scores of any question i of any user j;
[0019] Calculate the historical answer scores of any question i for any user j .
[0020] Preferably, the method for obtaining the historical score rate is:
[0021] Get the historical answer score and the corresponding historical answer information;
[0022] According to the formula:
[0023]
[0024] in, represents the number of questions answered correctly, n represents the number of all questions, Indicates the total theoretical score of all questions; represents the historical score rate of any user j;
[0025] Calculate the historical scoring rate .
[0026] Preferably, based on the historical scoring rate , the method for evaluating user learning ability is:
[0027] Based on the historical scoring rate , set a first threshold R1 and a second threshold R2;
[0028] The user's learning ability is evaluated by setting the first threshold R1 and the second threshold R2:
[0029] when >R1, the corresponding user is marked as excellent;
[0030] when <R2, the corresponding user is marked as poor;
[0031] When R2≤ When ≤R1, the corresponding user is marked as medium;
[0032] The first threshold R1 is greater than the second threshold R2.
[0033] Preferably, the calculation module is also used to calculate the user's historical proficiency for the knowledge point;
[0034] It also includes an analysis module that analyzes the user's mastery of knowledge points based on the user's historical proficiency in the knowledge points.
[0035] Preferably, the method for calculating the user's historical proficiency for a knowledge point is:
[0036] Get the user's historical answer information;
[0037] According to the formula:
[0038]
[0039] Among them, Tstd represents the total standard time for answering questions. Indicates the time to answer any question. is the total time the user spends answering questions. represents the proficiency of any user j; Indicates the weight corresponding to the answer score rate, Represents the weight corresponding to the answering time, where > ;
[0040] Calculate the historical proficiency of any user j for a knowledge point .
[0041] Preferably, the method by which the analysis module analyzes the user's mastery of knowledge points based on historical proficiency is:
[0042] Set a third threshold R3 and a fourth threshold R4;
[0043] when >R3, then all corresponding users are marked as proficient;
[0044] when <R4, then all corresponding users are marked as unfamiliar;
[0045] Wherein the third threshold value R3 is greater than the fourth threshold value R4;
[0046] When R4≤ ≤R3, the K-means algorithm is used to calculate the proficiency between the third threshold and the fourth threshold. The proficiency is greater than the third threshold R3. , proficiency less than or equal to the fourth threshold R4 as the initial cluster.
[0047] Preferably, the data collection module is also used to collect the current answer information of all users marked as unfamiliar; the data processing module is also used to calculate the user's current score rate based on the current answer information; the improvement model is also used to calculate the user's current proficiency based on the current answer information and the current score rate.
[0048] Preferably, the calculation module is also used to calculate the efficiency improvement value , the method is:
[0049] Get the user's historical proficiency With current proficiency , historical scoring rate With the current score rate ;
[0050] According to the formula:
[0051]
[0052] in, represents the influencing factor of proficiency, Indicates the influencing factor of the score rate, Indicates the interval time;
[0053] Calculate the user's efficiency improvement value per unit time .
[0054] The beneficial effects of the present invention are as follows:
[0055] 1. The learning ability evaluation and improvement system based on a large model described in the present invention calculates the user's score rate for the same knowledge point based on the accuracy of answering questions on the same knowledge point through a data processing module, so as to reflect whether there is any randomness in the user's answer to questions on the same knowledge point. Based on the calculated score rate, the user's learning ability is evaluated through an evaluation module. Secondly, in order to identify the user's shortcomings, relevant learning materials or exercises are provided to focus on strengthening those weak knowledge points. The user's historical answering proficiency is output through an improvement model. By obtaining the historical answering proficiency of all users, the user group with a weaker grasp of the knowledge point is screened out, and then this part of the user group is improved to form a closed loop of learning evaluation and improvement;
[0056] 2. The learning ability evaluation and improvement system based on a large model described in the present invention has a score rate The calculation of the system can reduce the problem of some users having high historical answer scores due to chance. Specifically, by obtaining the answer information of all questions corresponding to the same knowledge point of the user, that is, obtaining the number of questions with correct answers. And the number of all questions n, by introducing , reducing the problem of some users having high historical answer scores due to chance, resulting in a high score rate; then marking all users through the first threshold R1 and the second threshold R2, including three levels: excellent, poor and medium, based on which the user's basic learning ability is known. BRIEF DESCRIPTION OF THE DRAWINGS
[0057] The present invention will be further described below in conjunction with the accompanying drawings.
[0058] Figure 1 It is a flow chart of the present invention. DETAILED DESCRIPTION
[0059] In order to make the technical means, creative features, objectives and effects achieved by the present invention easy to understand, the present invention is further explained below in conjunction with specific implementation methods.
[0060] like Figure 1 As shown, a learning ability evaluation and improvement system based on a large model according to an embodiment of the present invention includes:
[0061] Data collection module, used to collect users' historical answer information;
[0062] The calculation module is used to calculate the user's historical answer scores based on the historical answer information;
[0063] The data processing module obtains the user's historical score rate based on historical answer scores and historical answer information;
[0064] The evaluation module evaluates the user's historical learning ability based on the historical score rate;
[0065] The improvement model outputs the user's historical answering proficiency based on the input user's historical answering information, and improves the user's learning ability based on the user's historical answering proficiency.
[0066] Based on the above, in essence, the existing technology obtains the user's answer information, uses the relevant information in the answer information to analyze the user's answer status during the actual test, and thus comprehensively calculates the user's real score, and does not rely on the traditional paper score. In the actual calculation process, the existing technology can only obtain the user's historical real score based on the historical answer information analysis, and simultaneously complete the learning ability assessment of all users. However, in the user's answering process, the questions corresponding to the same knowledge point are not unique. Therefore, when the user answers multiple questions on the same knowledge point, if there are both wrong and correct situations, there must be a certain degree of randomness in the user's answering of the questions. If the existence of randomness is not taken into account, it may cause the actual real scores of some users to be too high, which cannot truly reflect the user's learning ability.
[0067] In one embodiment of the present invention, a test system for user testing and a database capable of storing a large number of test questions are also included. When a user takes a periodic test, the test system randomly combines N test questions to form a set of test papers. Based on the test papers, any user answers the questions. The user's historical answer information is obtained through the data collection module, and the user's historical answer score is calculated by the calculation module. Then, based on the accuracy of the answers to the same knowledge point, the user's score rate for the same knowledge point is calculated through the data processing module, so as to reflect whether there is contingency when the user answers the questions for the same knowledge point. Based on the calculated score rate, the user's learning ability is evaluated through the evaluation module. Specifically, the user's learning ability is divided into three levels based on the score rate, namely excellent, medium and poor. Secondly, in order to identify the user's shortcomings, relevant learning materials or exercises are provided to focus on strengthening those weak knowledge points. The user's historical answer proficiency is output through the improvement model. By obtaining the historical answer proficiency of all users, the user group with weaker knowledge point mastery is screened out, and then the part of the user group is improved and studied, so as to form a closed loop of learning evaluation and improvement.
[0068] It is worth noting that when the user enters the test page of the test system, the test paper consisting of N test questions randomly combined by the test system will be displayed, and each time the user enters the test page for the second time, the test paper will be randomly generated again, and when the user leaves the test page, it is assumed that the user has completed the test, and the user's historical answer scores will be calculated.
[0069] In one embodiment, the answer information includes: answer accuracy A, answer times C and answer time T; when the user answers the question accurately, A=1, otherwise A=0.
[0070] As an embodiment of the present invention, based on obtaining the answer information of any question of the user, including the answer accuracy A, the number of answers C and the answer time T, the real score of the user for any question can be more realistically reflected. Through the expression of the answer accuracy A, if the user finally answers the question incorrectly, it means that the user has not fully mastered the knowledge point. If the answer is correct, the user's mastery of the knowledge point is analyzed based on the number of answers C and the answer time T. When the user submits the same question multiple times, that is, the number of answers C>1, it indicates that the user is uncertain about the question, which further reflects that the user may not have fully mastered the knowledge point. At the same time, if the user's answer time T for any question is too long, it reflects that the user may not have fully mastered the knowledge point.
[0071] The answer information refers to the information of the user's answer to any question, including the answer accuracy A, the number of answers C and the answer time T.
[0072] In one embodiment, the method by which the calculation module obtains the historical answer score according to the historical answer information is:
[0073] Get the user's answer information;
[0074] According to the formula:
[0075]
[0076] Among them, P represents the basic score, represents the accuracy of the answer to any question i, represents the number of answers to any question i; represents the historical answer scores of any question i of any user j;
[0077] Calculate the historical answer scores of any question i for any user j .
[0078] Based on the above, since the system randomly combines the test papers composed of N test questions in the database, they will be regenerated each time the user enters the test page. Therefore, each time the user takes the test, the test questions will not be exactly the same, and according to the number of test questions stored in the database, the more test questions there are, the lower the repetition rate of the test questions regenerated each time. When calculating the historical answer scores of the user each time, the calculation module first obtains the user's historical answer scores based on the historical answer information. As an embodiment of the present invention, the answer types include objective questions such as multiple-choice questions, fill-in-the-blank questions, and judgment questions. The user does not need to enter the answer process when answering, that is, the user gets points if the answer is correct, and loses points if the answer is wrong. Based on the number of answers, when the user answers correctly, when the number of user answers C>1, the user's final historical answer score will show a gradually decreasing trend, that is, after the user submits any question more than once, the score of the question will inevitably be less than the basic score P. When the user answers incorrectly, regardless of the number of times the user answers, the historical answer score of the question is 0.
[0079] In one embodiment, the method for obtaining the historical score rate is:
[0080] Get the historical answer score and the corresponding historical answer information;
[0081] According to the formula:
[0082]
[0083] in, represents the number of questions answered correctly, n represents the number of all questions, Indicates the total theoretical score of all questions; represents the historical score rate of any user j;
[0084] Calculate the historical scoring rate .
[0085] Based on the above, since the questions corresponding to the same knowledge point are not unique, when calculating the user's historical answer scores, relying only on the basic score P and the number of answers C cannot accurately calculate the user's real score. Since the same knowledge point may correspond to multiple questions, if the user fails to master the knowledge point, there will inevitably be some questions that are answered incorrectly and some that are answered correctly, that is, the above-mentioned contingency. In order to eliminate the problem of some users' historical answer scores being higher due to contingency, in one embodiment of the present invention, based on the score rate The calculation of the system can reduce the problem of some users having high historical answer scores due to chance. Specifically, by obtaining the answer information of all questions corresponding to the same knowledge point of the user, that is, obtaining the number of questions with correct answers. And the number of all questions n, by introducing , to reduce the problem of some users having high historical scores resulting in high scoring rates due to chance.
[0086] In one embodiment, based on the historical scoring rate , the method for evaluating user learning ability is:
[0087] Based on the historical scoring rate , set a first threshold R1 and a second threshold R2;
[0088] The user's learning ability is evaluated by setting the first threshold R1 and the second threshold R2:
[0089] when >R1, the corresponding user is marked as excellent;
[0090] when <R2, the corresponding user is marked as poor;
[0091] When R2≤ When ≤R1, the corresponding user is marked as medium;
[0092] The first threshold R1 is greater than the second threshold R2.
[0093] Based on the above, according to the historical score rate of the user obtained , and then evaluate the user's learning ability. In one embodiment of the present invention, a first threshold R1 and a second threshold R2 are set. All users are marked by the first threshold R1 and the second threshold R2, including three levels: excellent, poor and medium. Based on this, the user's basic learning ability is known.
[0094] In one embodiment, the calculation module is further used to calculate the user's historical proficiency for the knowledge point;
[0095] It also includes an analysis module that analyzes the user's mastery of knowledge points based on the user's historical proficiency in the knowledge points.
[0096] In order to obtain the user's mastery of knowledge points, it is also necessary to calculate the proficiency. The calculation module is used to calculate the user's proficiency in knowledge points, and then the user's mastery of knowledge points is analyzed through historical proficiency. Unlike learning ability, the above-mentioned score rate is used to characterize the user's basic learning ability and preliminarily reflect the user's mastery of knowledge points, but it still cannot accurately characterize the proficiency of all users for knowledge points. For example, when the user's score rate is high, if one user has high answer accuracy but many submissions, and the other user has low answer accuracy but few submissions; for the above two users, their mastery of knowledge points cannot be distinguished by the score rate alone.
[0097] In one embodiment, the method for calculating the user's historical proficiency for a knowledge point is:
[0098] Get the user's historical answer information;
[0099] According to the formula:
[0100]
[0101] Among them, Tstd represents the total standard time for answering questions. Indicates the time to answer any question. is the total time the user spends answering questions. represents the proficiency of any user j; Indicates the weight corresponding to the answer score rate, Represents the weight corresponding to the answering time, where > ;
[0102] Calculate the historical proficiency of any user j for a knowledge point .
[0103] Based on the above, in order to characterize the user's mastery of knowledge points through data, in one embodiment of the present invention, the historical proficiency is obtained through the formula , using data to characterize the degree of mastery of any user on a knowledge point, based on the above formula, by setting the user's total answering time The ratio of the total answering standard time Tstd. The larger the ratio, the higher the user's historical proficiency. The smaller the relative value, the larger the value. At the same time, in order to eliminate the relative influence of the scoring rate and the answering time, that is, when there is a user with a high scoring rate and a long time or a slightly lower scoring rate but a short time, the historical proficiency calculated in the two cases There may be some overlap. To avoid this, the score rate and answer time are given different weights. In one embodiment of the present invention, > , indicating that the impact of the scoring rate is greater than the impact of the answering time.
[0104] In one embodiment, the method by which the analysis module analyzes the user's mastery of knowledge points based on historical proficiency is:
[0105] Set a third threshold R3 and a fourth threshold R4;
[0106] when >R3, then all corresponding users are marked as proficient;
[0107] when <R4, then all corresponding users are marked as unfamiliar;
[0108] Wherein the third threshold value R3 is greater than the fourth threshold value R4;
[0109] When R4≤ ≤R3, the K-means algorithm is used to calculate the proficiency between the third threshold and the fourth threshold. The proficiency is greater than the third threshold R3. , proficiency less than or equal to the fourth threshold R4 as the initial cluster.
[0110] As an embodiment of the present invention, the third threshold R3 and the fourth threshold R4 are set to calculate the historical proficiency corresponding to all users. Classify to find out which users need to be improved and the historical proficiency of some users Exists in the interval [R4, R3], so it is also necessary to use the K-means algorithm to analyze the proficiency between the third threshold and the fourth threshold. The users are divided into groups so that all can be marked as proficient or unfamiliar. The unfamiliar users can then be trained to improve, thus completing the closed loop from evaluation to improvement.
[0111] In one embodiment, the data collection module is further used to collect current answer information of all users marked as unfamiliar; the data processing module is further used to calculate the user's current score rate based on the current answer information; the improvement model is further used to calculate the user's current proficiency based on the current answer information and the current score rate.
[0112] As an embodiment of the present invention, the user's historical answer scores, historical score rates and historical proficiency calculated above are all taken as historical records and can be stored in a data storage unit. In order to verify whether the user has mastered the knowledge points after learning and improvement, further testing is required for verification; wherein the data collection module also includes a data storage unit.
[0113] In one embodiment, the calculation module is also used to calculate the efficiency improvement value , the method is:
[0114] Get the user's historical proficiency With current proficiency , historical scoring rate With the current score rate ;
[0115] According to the formula:
[0116]
[0117] in, represents the influencing factor of proficiency, Indicates the influencing factor of the score rate, Indicates the interval time;
[0118] Calculate the user's efficiency improvement value per unit time .
[0119] By using the user's historical proficiency With current proficiency , historical scoring rate With the current score rate Calculate the user's efficiency improvement value , based on the efficiency improvement value It reflects the changes in the user's mastery of knowledge points after the learning and improvement process, thereby helping the user to master the knowledge points faster and help adjust the specific operations of the learning and improvement process.
[0120] Working principle: The user's historical answer information is obtained through the data collection module, and the user's historical answer score is calculated by the calculation module. Then, the data processing module is used to calculate the user's score rate for the same knowledge point based on the accuracy of the answer to the same knowledge point. This reflects whether there is any randomness when the user answers the questions on the same knowledge point. Based on the calculated score rate, the evaluation module is used to evaluate the user's learning ability. Specifically, the user's learning ability is divided into three levels based on the score rate, namely excellent, medium and poor. Secondly, in order to identify the user's shortcomings, relevant learning materials or exercises are provided to focus on strengthening those weak knowledge points. The model is also improved to output the user's historical answer proficiency. By obtaining the historical answer proficiency of all users, the users with weak knowledge points can be screened out. user groups, and then improve learning for this part of user groups, so as to form a closed loop of learning evaluation and improvement; when calculating the historical answer scores of users each time, first the calculation module preferentially obtains the historical answer scores of users according to the historical answer information. As an embodiment of the present invention, the answer types include objective questions such as multiple-choice questions, fill-in-the-blank questions, and judgment questions. Users do not need to enter the answer process when answering, that is, if the user answers correctly, he / she will get points, and if the answer is wrong, he / she will lose points. Based on the number of answers, if the user answers correctly, when the number of user answers C>1, the user's final historical answer score will show a gradually decreasing trend, that is, after the user submits any question more than once, the score of the question will inevitably be less than the basic score P. If the user answers incorrectly, regardless of the number of user answers, the historical answer score of the question is 0;
[0121] Based on the score rate The calculation of the system can reduce the problem of some users having high historical answer scores due to chance. Specifically, by obtaining the answer information of all questions corresponding to the same knowledge point of the user, that is, obtaining the number of questions with correct answers. And the number of all questions n, by introducing , reduce the problem of high scoring rate caused by some users' high historical answer scores due to chance; use the calculation module to calculate the user's proficiency in knowledge points, and then analyze the user's mastery of knowledge points through historical proficiency. Unlike learning ability, the above scoring rate is used to characterize the user's basic learning ability and preliminarily reflect the user's mastery of knowledge points, but it still cannot accurately characterize the proficiency of all users for knowledge points. For example, when the user's scoring rate is high, if one user has high answer accuracy but many submissions, and the other user has low answer accuracy but few submissions; for the above two users, their mastery of knowledge points cannot be distinguished only by the scoring rate.
[0122] Obtaining historical proficiency through formula , using data to characterize the degree of mastery of any user's knowledge points, based on the above formula, by setting the user's total answering time The ratio of the total answering standard time Tstd. The larger the ratio, the higher the user's historical proficiency. The smaller the relative value, the larger the value. At the same time, in order to eliminate the relative influence of the scoring rate and the answering time, that is, when there is a user with a high scoring rate and a long time or a slightly lower scoring rate but a short time, the historical proficiency calculated in the two cases There may be some overlap. To avoid this, the score rate And give different weights to the answering time.
[0123] The above shows and describes the basic principles, main features and advantages of the present invention. It should be understood by those skilled in the art that the present invention is not limited to the above embodiments. The above embodiments and descriptions are only for explaining the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, which fall within the scope of the present invention. The scope of protection of the present invention is defined by the attached claims and their equivalents.
Claims
1. A learning ability evaluation and improvement system based on a large model, characterized in that: include: Data collection module, used to collect users' historical answer information; The calculation module is used to calculate the user's historical answer scores based on the historical answer information; The data processing module obtains the user's historical score rate based on historical answer scores and historical answer information; The evaluation module evaluates the user's historical learning ability based on the historical score rate; The improvement model outputs the user's historical answering proficiency based on the input user's historical answering information, and improves the user's learning ability based on the user's historical answering proficiency.
2. A learning ability assessment and improvement system based on a large model according to claim 1, characterized in that: The answer information includes: answer accuracy A, answer times C and answer time T; when the user answers the question accurately, A=1, otherwise A=0.
3. A learning ability assessment and improvement system based on a large model according to claim 2, characterized in that: The method for the calculation module to obtain the historical answer score according to the historical answer information is: Get the user's answer information; According to the formula: , Among them, P represents the basic score, represents the accuracy of the answer to any question i, represents the number of answers to any question i; represents the historical answer scores of any question i of any user j; Calculate the historical answer scores of any question i for any user j .
4. A learning ability assessment and improvement system based on a large model according to claim 3, characterized in that: The method for obtaining the historical score rate is: Get the historical answer score and the corresponding historical answer information; According to the formula: ,, in, represents the number of questions answered correctly, n represents the number of all questions, Indicates the total theoretical score of all questions; represents the historical score rate of any user j; Calculate the historical scoring rate .
5. A learning ability assessment and improvement system based on a large model according to claim 4, characterized in that: Based on the historical scoring rate , the method for evaluating user learning ability is: Based on the historical scoring rate , set a first threshold R1 and a second threshold R2; The user's learning ability is evaluated by setting the first threshold R1 and the second threshold R2: when >R1, the corresponding user is marked as excellent; when <R2, the corresponding user is marked as poor; When R2≤ When ≤R1, the corresponding user is marked as medium; The first threshold R1 is greater than the second threshold R2.
6. A learning ability assessment and improvement system based on a large model according to claim 5, characterized in that: The calculation module is also used to calculate the user's historical proficiency for the knowledge point; It also includes an analysis module that analyzes the user's mastery of knowledge points based on the user's historical proficiency in the knowledge points.
7. A learning ability assessment and improvement system based on a large model according to claim 6, characterized in that: The method for calculating the user's historical proficiency for a knowledge point is: Get the user's historical answer information; According to the formula: , Among them, Tstd represents the total standard time for answering questions. Indicates the time to answer any question. is the total time the user spends answering questions. represents the proficiency of any user j; Indicates the weight corresponding to the answer score rate, Represents the weight corresponding to the answering time, where > ; Calculate the historical proficiency of any user j for a knowledge point .
8. A learning ability assessment and improvement system based on a large model according to claim 7, characterized in that: The method by which the analysis module analyzes the user's mastery of knowledge points based on historical proficiency is: Set a third threshold R3 and a fourth threshold R4; when >R3, then all corresponding users are marked as proficient; when <R4, then all corresponding users are marked as unfamiliar; Wherein the third threshold value R3 is greater than the fourth threshold value R4; When R4≤ ≤R3, the K-means algorithm is used to calculate the proficiency between the third threshold and the fourth threshold. The proficiency is greater than the third threshold R3. , proficiency less than or equal to the fourth threshold R4 as the initial cluster.
9. A learning ability assessment and improvement system based on a large model according to claim 8, characterized in that: The data collection module is further used to collect the current answer information of all users marked as unfamiliar; the data processing module is further used to calculate the user's current score rate based on the current answer information; the improvement model is further used to calculate the user's current proficiency based on the current answer information and the current score rate.
10. A learning ability assessment and improvement system based on a large model according to claim 9, characterized in that: The calculation module is also used to calculate the efficiency improvement value , the method is: Get the user's historical proficiency With current proficiency , Historical scoring rate With the current score rate ; According to the formula: , in, represents the influencing factor of proficiency, Indicates the influencing factor of the score rate, Indicates the interval time; Calculate the user's efficiency improvement value per unit time .
Citation Information
Patent Citations
Personalized assessment method for subject grasping ability of student
CN108053117A
Learning evaluation method and device
CN111582694A
Method for measuring mastery degree of knowledge points
CN112116506A
Learning ability evaluation method and system based on cognitive diagnosis
CN114491050A
Multi-knowledge-point fusion knowledge tracking method based on auxiliary task enhancement
CN118333156A
