Student score analysis management method and system based on artificial intelligence

Abstract

The invention provides a student score analysis management method and system based on artificial intelligence, and relates to the technical field of artificial intelligence data processing. The method comprises the steps of collecting historical test data of a target student group, and performing equivalent difficulty big data clustering analysis based on a knowledge module to form effective group clustering data; performing individual score stability analysis on different student groups according to the group clustering effective data to form individual stability analysis result data; and according to the individual stability analysis result data, performing knowledge module stability clustering to form individual knowledge module learning guide data. According to the method, efficient clustering analysis processing based on the knowledge module is carried out on student scores, and the condition that students master learning contents can be displayed and determined more accurately, reasonably and clearly.

Description

Technical Field

[0001] The present invention relates to the technical field of artificial intelligence data processing, and more particularly, to a method and system for analyzing and managing student grades based on artificial intelligence. Background Art

[0002] Student grades are a verification of students' mastery of previous learning content. Therefore, it is necessary to test students' learning content in a timely manner. As an intuitive manifestation of the degree of mastery of learning content, the value of test scores has decisive reference significance. Currently, most of the statistics and processing of student grades are still based on manual work, which is cumbersome and difficult to show relatively adaptable data processing capabilities when facing a large amount of data. This results in the inability to manage student grade data efficiently and analyze and process it reasonably.

[0003] With the development of artificial intelligence, the technology of automated data processing and analysis has become more and more mature. Artificial intelligence can gradually replace manual work in intelligent analysis and processing, especially in the face of a large amount of data. Therefore, efficient and reasonable analysis and processing of student grades have become a reality.

[0004] Therefore, designing a method and system for analyzing and managing student grades based on artificial intelligence, through efficient clustering analysis and processing of student grades based on knowledge modules, can more accurately and reasonably display and determine the situation of students' mastery of learning content. This is an urgent problem to be solved at present. Summary of the Invention

[0005] The purpose of the present invention is to provide a method for analyzing and managing student grades based on artificial intelligence. By collecting historical test data of the target student group, clustering based on different knowledge modules is performed, and then the stability of the grades of different students is analyzed separately for different knowledge modules. At the same time, the relative situation of the grades of different students in the group is considered as a whole to reasonably perform targeted processing of the mastery of knowledge modules. To a certain extent, it can provide a clear understanding and improvement guidance for students with different grade statuses regarding the mastery of knowledge modules, ensuring that students can achieve greater reasonable learning improvement and learning awareness while reasonably combining their own learning abilities, and promoting the good learning state of individual students.

[0006] The object of the present invention also lies in providing a student achievement analysis and management system based on artificial intelligence. The system can collect historical test data of a target student group for clustering analysis based on knowledge modules, realize the learning stability analysis of knowledge modules for different individual students, provide a clear understanding of the knowledge mastery situation and appropriate improvement guidance, effectively ensure a comprehensive understanding of the knowledge learning and mastery situation of individual students, and is an important material basis for realizing reasonable and efficient student achievement analysis and management.

[0007] In a first aspect, the present invention provides a method for student achievement analysis and management based on artificial intelligence, including: collecting historical test data of a target student group, and performing big data clustering analysis of equivalent difficulty based on knowledge modules to form effective group clustering data; according to the effective group clustering data, performing individual achievement stability analysis on different student groups to form individual stability analysis result data; according to the individual stability analysis result data, performing knowledge module stability clustering to form individual knowledge module learning guidance data.

[0008] In the present invention, the method collects historical test data of a target student group, performs clustering based on different knowledge modules, then separately performs achievement stability analysis on different individual students for different knowledge modules, and at the same time reasonably processes the knowledge module mastery situation by considering the relative situation of the achievements of different individual students in the group as a whole. To a certain extent, it can provide a clear understanding of the knowledge module mastery degree and improvement guidance for individual students with different achievement statuses, ensure that individual students can achieve greater reasonable learning improvement and learning understanding in combination with their own learning abilities, and promote the good learning state of individual students.

[0009] As a possible implementation, collecting historical test data of a target student group, and performing big data clustering analysis of equivalent difficulty based on knowledge modules to form effective group clustering data, includes: performing achievement clustering based on knowledge modules on different individual students in the target student group according to the historical test data to form individual knowledge module achievement clustering data; performing clustering division based on achievement values according to the individual knowledge module achievement clustering data corresponding to different individual students to form effective group clustering data.

[0010] In the present invention, for the historical test data of the target student group, clustering analysis based on knowledge modules is performed. On the one hand, it is necessary to extract different knowledge modules of each student individual in each test from the historical data, and then cluster the test results of different student individuals in each test according to the knowledge modules to form student individual test situation data for knowledge modules. On the other hand, it is necessary to consider that the test results corresponding to the extracted knowledge modules need to be quantified, so as to better evaluate and analyze the learning and mastery of student individuals in the corresponding knowledge modules based on the quantified data information.

[0011] As a possible implementation method, according to the historical test data, grade clustering based on knowledge modules is performed on different student individuals in the target student group to form individual knowledge module grade clustering data, including: extracting the test scores of different knowledge modules of different student individuals in the historical test data and arranging them in the order of time dimension to form an individual knowledge module score order set corresponding to different knowledge modules; determining the single test grade for different knowledge modules, and performing equivalent score fitting analysis on different individual knowledge module score order sets according to different single test grades to form corresponding individual knowledge module grade change data.

[0012] In the present invention, when clustering the test result data of different knowledge modules of different student individuals in each test extracted from the historical test data, it should be considered that, as an important information for measuring the test results, the test score usually reflects the total score of the test content in a single test. However, due to the quantifiability of the score, after splitting the test content based on knowledge modules, the score of the corresponding knowledge module in this test can be determined by counting the small scores of the corresponding test questions. Of course, different knowledge modules do not appear in each exam. However, for the target student group, since the content of each exam is unified, the test scores extracted for any knowledge module are unified and highly comparable. At the same time, considering that the test difficulty set for different knowledge modules in each test may be different due to other factors considered in the test, and this test difficulty is usually set with a standard reference. Therefore, if the obtained test scores are directly analyzed in the same way, it will lead to deviation due to the difficulty of analysis, which will affect the accuracy. To ensure the rationality of the analysis, the test difficulty of each test, especially the difficulty for knowledge modules, needs to be reasonably considered and reflected in the corresponding test scores, so as to ensure more reasonable and accurate analysis of the learning and mastery situation based on the test scores. In addition, since the learning and mastery of knowledge modules have a time effect, the influence of time factors should also be considered when analyzing the extracted test scores to ensure the rationality and accuracy of the analysis.

[0013] As a possible implementation, determine the single - test level for different knowledge modules, and based on different single - test levels, conduct equivalent score fitting analysis on the score order sets of different individual knowledge modules to form corresponding individual knowledge module performance change data, including: for different knowledge modules, conduct different - level continuous assignment based on the level order according to the test difficulty level to determine the difficulty equivalent values corresponding to different levels; according to the difficulty equivalent values of different levels of the knowledge module and in combination with the single - test level corresponding to each test in the individual knowledge module score order set, determine the difficulty equivalent value of each test; for different knowledge modules, determine the individual knowledge module equivalent score sets corresponding to different student individuals according to the individual knowledge module score order sets of different student individuals and the difficulty equivalent value of each test , where , m is the number of different knowledge modules, n is the number of different student individuals, k is the number of different tests, , , is the test score obtained by the student individual numbered n under the knowledge module numbered m in the test numbered k, is the difficulty equivalent value determined for the knowledge module numbered m in the test numbered k, is the maximum value of the continuous assignment of the level corresponding to the knowledge module numbered m, is the single - test level assignment corresponding to the knowledge module numbered m in the test numbered k; according to the individual knowledge module equivalent score sets corresponding to different student individuals , conduct fitting in the time - dimension order to form an individual knowledge module test change function .

[0014] In the present invention, for the difficulty levels set for different knowledge modules in each test, the difficulty levels are considered uniformly to ensure the rationality of the obtained equivalent scores. Since there are usually reasonable standard references for the difficulty ratings of test questions, the test scores of different tests can be equivalently unified according to such standard references. Of course, there are various equivalent methods. In this application, by assigning scores to the standard of difficulty levels in the order of difficulty levels, the scores corresponding to different difficulty levels are obtained, and then when performing score equivalence, the score of the highest level is used as a reference to determine the relative quantity, and the difficulty equivalent value corresponding to each test is determined, and then the same equivalent score is obtained.

[0015] As a possible implementation, according to the individual knowledge module performance clustering data corresponding to different student individuals, conduct clustering division based on the performance values to form group clustering effective data, including: for the individual knowledge module performance clustering data, according to the individual knowledge module test change function corresponding to different student individuals , the overall test level value of the individual knowledge modules of different student individuals is determined , where ; with the horizontal axis being the test level value and the vertical axis being the number distribution value, based on the overall test level value of the individual knowledge modules , the number statistics based on the level value is carried out to establish the number distribution of the knowledge module test level; a clustering number division value is set. On the number distribution of the knowledge module test level, all student individuals with the number exceeding the clustering number division value are determined as the set of moderately performing student individuals corresponding to the knowledge module, and all student individuals with the number less than the clustering number division value and the overall test level value of the individual knowledge module being less than the minimum overall test level value of the individual knowledge module in the set of moderately performing student individuals are determined as the set of poorly performing student individuals corresponding to the knowledge module, and all the remaining student individuals except the set of poorly performing student individuals and the set of moderately performing student individuals are determined as the set of excellent student individuals corresponding to the knowledge module.

[0016] In the present invention, a practical problem to be considered is that for the learning and mastery of knowledge modules, due to the different learning abilities of different student individuals, the absorption and mastery levels of knowledge are different. Therefore, it is impossible to uniformly quantify the analysis and determination of the learning and mastery of knowledge modules. After dividing the student individuals based on their learning ability, different student individuals with different learning abilities need to be determined, and then the analysis and processing of the learning and mastery of knowledge modules can be carried out more pertinently. Based on this, considering that the learning ability of different student individuals can also be reflected by the test scores, and the way of reflection is through the overall test level of the student individuals. Therefore, after obtaining the equivalent scores of each student individual on the corresponding knowledge module each time, the change in the mastery of the student individual on the knowledge module is determined by fitting, and then the overall test level is determined by integration. Different student groups with different learning abilities are reasonably clustered by using the overall test level. It should be noted here that for the clustering of different student individuals with different learning abilities, this application is mainly divided into three categories: poor, medium, and excellent. On the one hand, considering that the test level is normally distributed in the group, the division into three categories can use the distribution estimation of the normal distribution to make the division more scientific. On the other hand, the division into three categories can avoid the complexity of subsequent analysis caused by too many clusters while fully meeting the requirements of clustering division. The clustering number division value can be set according to the actual situation or determined through big data analysis.

[0017] As a possible implementation, based on the effective data of group clustering, individual performance stability analysis is carried out for different student groups to form individual stability analysis result data, including: for different student individuals concentrated in the poor-module students, stability analysis based on quantitative growth is carried out to form growth stability analysis result data corresponding to the student individuals; for different student individuals concentrated in the medium-module students, stability analysis based on fluctuating upward is carried out to form upward stability analysis result data corresponding to the student individuals; for different student individuals concentrated in the excellent-module students, stability analysis based on quantitative fluctuation is carried out to form quantitative stability analysis result data corresponding to the student individuals.

[0018] In the present invention, the analysis and processing of the learning and mastering situations of different knowledge modules by student individuals with different learning abilities have certain differences due to different learning abilities. Therefore, different considerations need to be made for the stability analysis, so as to more reasonably and accurately reflect the learning and mastering situations of student individuals for knowledge modules and provide appropriate learning guidance for them.

[0019] As a possible implementation, for different student individuals concentrated in the poor-module students, stability analysis based on quantitative growth is carried out to form growth stability analysis result data corresponding to the student individuals, including: for different student individuals concentrated in the poor-module students, according to the corresponding individual knowledge module test change function , determine the corresponding individual knowledge module test change rate function , where i is the number of different student individuals concentrated in the poor-module students under the knowledge module numbered m; set the growth compliance ratio β, and for different student individuals, according to the individual knowledge module test change rate function , conduct the following growth stability analysis: if , then form the growth stability compliance information corresponding to the student individual; if , then form the growth instability information corresponding to the student individual, where represents the ratio of the span length on the horizontal axis of the function to the total span length, and β > 50%.

[0020] In the present invention, for the individual students in the differential clustering, due to their own learning ability, considering that it is necessary to try to improve the learning and mastery of knowledge modules as the learning guidance, the stability analysis of individual students is required to be growth-oriented. The growth compliance ratio can be determined according to the actual situation or through big data analysis. It should be noted that the setting of the growth compliance ratio is mainly to ensure that the learning and mastery of knowledge modules by individual students can reflect an increasing trend in the scores of most test results through the learning during the considered test period, indicating that the individual students in this clustering are working hard within the scope of their learning ability. For the unqualified situation, it provides a reference to guide individual students to reasonably master the learning of knowledge modules.

[0021] As a possible implementation method, for different individual students concentrated in the middle of the module, perform stability analysis based on fluctuating upward to form the upward stability analysis result data corresponding to the individual students, including: for different individual students concentrated in the middle of the module, according to the corresponding individual knowledge module test change function to determine the corresponding individual knowledge module test change rate function , where j is the number of different individual students concentrated in the middle of the module under the knowledge module numbered m; set the upward compliance ratio γ and the maximum fluctuation limit , β > γ > 50%, for different individual students, according to the individual knowledge module test change function and the individual knowledge module test change rate function , perform the following fluctuating upward analysis: If , and , then form the fluctuating upward compliance information corresponding to the individual student, represents the maximum difference in equivalent scores between two different tests; if , and , then form the fluctuating instability information corresponding to the individual student; if , and , then form the upward instability information corresponding to the individual student; if , and , then form the fluctuating upward instability information corresponding to the individual student.

[0022] In the present invention, for the students in the medium clustering group, their learning ability is at the middle level. Therefore, when analyzing the stability, the focus is mainly on the stability of the fluctuating test scores and a certain degree of upward trend. Thus, by setting two parameters, namely the upward-reaching proportion and the maximum fluctuation limit, a reasonable limitation can be achieved. It should be noted that the two parameters can restrict each other to determine the mastery situation of the students in this clustering group. The upward-reaching proportion can ensure that the learning and mastery of knowledge modules during the test period show a certain upward trend in terms of scores. The maximum fluctuation limit ensures that the upward trend will not be masked by the instability caused by large fluctuations, and only a reasonable small-scale fluctuation is a manifestation of stability. Of course, considering that students in the poor clustering group need greater incentives for upward trends, the upward-reaching proportion is less than the growth-reaching proportion, but both exceed 50%.

[0023] As a possible implementation method, for different students concentrated in the excellent student group of modules, a stability analysis based on quantitative fluctuations is carried out to form the corresponding quantitative stability analysis result data for individual students, including: for different students concentrated in the excellent student group of modules, according to the corresponding individual knowledge module test change function , determine the corresponding individual knowledge module test change rate function , where v is the number of different students in the excellent student group of modules under the knowledge module numbered m; set the quantitative reaching proportion ζ and the maximum stable fluctuation limit , ζ = 50%, , for different students, according to the individual knowledge module test change function and the individual knowledge module test change rate function conduct the following fluctuation upward analysis: If , and , then form the corresponding quantitative stable reaching information for the individual student; if , and , then form the corresponding quantitative unstable information for the individual student; if , and , then form the corresponding indefinite quantitative stable information for the individual student; if , and , then form the corresponding indefinite quantitative non-stable information for the individual student.

[0024] In the present invention, for the student individuals in the excellent clustering, their learning and mastery of knowledge modules are already at a relatively high level. Therefore, the stability analysis conducted on them mainly aims to ensure that there will be no significant decline during the test period. Thus, a quantitative compliance ratio of 50% is used as the constraint condition for the change rate of the test score value, and at the same time, the maximum stable fluctuation limit restricts the fluctuation of their stability. Of course, considering that maintaining a relatively high level of learning and mastery of knowledge modules requires a certain learning ability, the maximum stable fluctuation limit is less than the maximum fluctuation limit.

[0025] Secondly, the present invention provides an artificial intelligence-based student performance analysis and management system, including: a data acquisition unit for collecting historical test data of a target student group; a clustering analysis unit for obtaining the historical test data collected by the data acquisition unit, conducting clustering analysis to form effective group clustering data, and conducting individual performance stability analysis to form individual stability analysis result data; and an output display unit for obtaining the individual stability analysis result data formed by the clustering analysis unit, conducting knowledge module stability clustering, and forming individual knowledge module learning guidance data.

[0026] In the present invention, the system is configured to be able to collect historical test data of a target student group for clustering analysis based on knowledge modules, realize targeted knowledge module learning stability analysis for different student individuals, provide an organic whole for a clear understanding of the knowledge mastery situation and appropriate improvement guidance, and effectively ensure a comprehensive understanding of the knowledge learning and mastery situation of student individuals, which is an important material basis for realizing reasonable and efficient student performance analysis and management.

[0027] The beneficial effects of the artificial intelligence-based student performance analysis and management method and system provided by the present invention are as follows: By collecting historical test data of a target student group, conducting clustering based on different knowledge modules, and then separately conducting individual performance stability analysis for different knowledge modules, and at the same time considering the relative situation of the scores of different student individuals in the group as a whole to reasonably conduct targeted processing of the knowledge module mastery situation, to a certain extent, it can provide a clear understanding of the knowledge module mastery degree and improvement guidance for student individuals with different performance states, ensure that student individuals can achieve greater reasonable learning improvement and learning understanding in combination with their own learning abilities, and promote the good learning state of student individuals.

[0028] The system can collect the historical test data of the target student group and perform clustering analysis based on knowledge modules, realizing the analysis of the learning stability of knowledge modules for different individual students, providing a clear understanding of the knowledge mastery situation and appropriate improvement guidance, effectively ensuring a comprehensive understanding of the knowledge learning and mastery of individual students, and being an important material basis for realizing reasonable and efficient student achievement analysis and management. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments of the present invention will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present invention and should not be regarded as limiting the scope. For those of ordinary skill in the art, without creative efforts, other related drawings can also be obtained based on these drawings.

[0030] Figure 1 It is a step diagram of the method for analyzing and managing student achievements based on artificial intelligence provided by the embodiments of the present invention; Figure 2 It is a structural schematic diagram of the system for analyzing and managing student achievements based on artificial intelligence provided by the embodiments of the present invention; Figure 3 It is a distribution diagram of the number of people with test levels of knowledge modules of the method for analyzing and managing student achievements based on artificial intelligence provided by the embodiments of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0031] The technical solutions in the embodiments of the present invention will be described below with reference to the drawings in the embodiments of the present invention.

[0032] Student achievements are a verification of students' mastery of previous learning content. Therefore, it is necessary to timely test students' learning content. As an intuitive manifestation of measuring the mastery of learning content, the value of test scores has decisive reference. Currently, most of the statistics and processing of student achievements are still based on manual work, with cumbersome workload and difficult to show relatively adaptable data processing capabilities in the face of huge amounts of data, resulting in the situation that the management of student achievement data cannot be efficient and reasonably analyzed and processed.

[0033] With the development of artificial intelligence, the technology of automated data processing and analysis is becoming more and more mature. Artificial intelligence can gradually replace manual work in intelligent analysis and processing, especially in the face of large amounts of data. Therefore, the efficient and reasonable analysis and processing of student achievements have become a reality.

[0034] Reference Figures 1 to 3, the embodiment of the present invention provides a method for analyzing and managing students' grades based on artificial intelligence. This method collects historical test data of the target student group, conducts clustering based on different knowledge modules, and then analyzes the grade stability of different student individuals separately for different knowledge modules. At the same time, it reasonably processes the mastery of knowledge modules by considering the relative situation of the grades of different student individuals in the group as a whole. To a certain extent, it can provide a clear understanding and improvement guidance for the mastery of knowledge modules for student individuals in different grade states, ensuring that student individuals can achieve greater reasonable learning improvement and learning awareness by reasonably combining their own learning abilities, and promoting the good learning state of student individuals.

[0035] The method for analyzing and managing students' grades based on artificial intelligence specifically includes the following steps: S1: Collect historical test data of the target student group, and conduct big data clustering analysis of equivalent difficulty based on knowledge modules to form effective group clustering data.

[0036] Collect historical test data of the target student group, and conduct big data clustering analysis of equivalent difficulty based on knowledge modules to form effective group clustering data, including: According to the historical test data, conduct grade clustering based on knowledge modules for different student individuals in the target student group to form individual knowledge module grade clustering data; According to the individual knowledge module grade clustering data corresponding to different student individuals, conduct clustering division based on grade values to form effective group clustering data.

[0037] Conduct clustering analysis based on knowledge modules on the historical test data of the target student group. On the one hand, it is necessary to extract different knowledge modules of each student individual's each test from the historical data, and then cluster the test results of different student individuals according to the knowledge modules to form data on the test situations of student individuals for knowledge modules. On the other hand, it is necessary to consider that the test results corresponding to the extracted knowledge modules need to be quantified, so as to better evaluate and analyze the learning and mastery situations of student individuals on the corresponding knowledge modules based on the quantified data information.

[0038] According to the historical test data, conduct grade clustering based on knowledge modules for different student individuals in the target student group to form individual knowledge module grade clustering data, including: Extract the test scores of different knowledge modules of different student individuals in the historical test data, and arrange them in chronological order to form an ordered set of individual knowledge module scores corresponding to different knowledge modules; Determine the single - test grade for different knowledge modules, and conduct equivalent score fitting analysis on different ordered sets of individual knowledge module scores according to different single - test grades to form corresponding individual knowledge module grade change data.

[0039] Extract the test result data of different knowledge modules for each test of different student individuals from historical test data for clustering. It should be noted that as an important piece of information for measuring test results, although the test score usually reflects the total score of the test content in a single test, due to the quantifiability of the score, after splitting the test content based on knowledge modules, the scores of corresponding test questions can be used for statistics on knowledge modules, and then the scores of corresponding knowledge modules in this test can be determined. Of course, different knowledge modules do not appear in every exam. However, for the target student group, since the content of each exam is unified, the test scores extracted for any knowledge module are unified and highly comparable. At the same time, considering that the test difficulty set for different knowledge modules in each test may vary due to other factors considered in the test, and this test difficulty is usually set with a standard reference. Therefore, if the obtained test scores are directly analyzed uniformly, it will lead to deviations caused by difficulties in analysis. To ensure the rationality of the analysis, the test difficulty for each test, especially the difficulty for knowledge modules, needs to be reasonably considered and reflected in the corresponding test scores, so as to ensure more reasonable and accurate analysis when analyzing the learning and mastery situation based on test scores. In addition, since the learning and mastery situation of knowledge modules has a time effect, the time factor should also be considered when analyzing the extracted test scores to ensure the rationality and accuracy of the analysis.

[0040] Determine the single - test level for different knowledge modules, and based on different single - test levels, conduct equivalent - score fitting analysis on the score - order sets of different individual knowledge modules to form corresponding individual knowledge - module performance - change data, including: for different knowledge modules, assign different consecutive values based on the level order according to the test - difficulty level to determine the difficulty equivalent values corresponding to different levels; according to the difficulty equivalent values of different levels of knowledge modules and combined with the single - test level corresponding to each test in the individual knowledge - module score - order set, determine the difficulty equivalent value of each test; for different knowledge modules, based on the individual knowledge - module score - order sets of different student individuals and the difficulty equivalent values of each test, determine the individual knowledge - module equivalent - score sets corresponding to different student individuals , where, , m is the number of different knowledge modules, n is the number of different student individuals, k is the number of different tests, , , is the test score obtained by the student individual numbered n under the knowledge module numbered m in the test numbered k, is the difficulty equivalent value determined for the knowledge module numbered m in the test numbered k, The maximum value assigned continuously to the level corresponding to the knowledge module numbered m Assign a single test level value corresponding to the knowledge module numbered m in the k-th test; according to the equivalent score set of individual knowledge modules corresponding to different student individuals Fit in the order of the time dimension to form an individual knowledge module test change function .

[0041] For the difficulty levels set for different knowledge modules in each test, the difficulty levels are considered uniformly to ensure the rationality of the obtained equivalent scores. Since there are usually reasonable standard references for the difficulty ratings of test questions, the test scores of different times can be equivalently unified according to this standard reference. Of course, there are various equivalent methods. In this application, the scores corresponding to different difficulty levels are obtained by assigning sequential difficulty levels to the difficulty level standards. Then, when performing score equivalence, the score of the highest level is used as a reference to determine the relative quantity, and the difficulty equivalent value corresponding to each test is determined, and then the same equivalent score is obtained.

[0042] According to the clustering data of individual knowledge module scores corresponding to different student individuals, perform clustering division based on the score values to form effective group clustering data, including: for the clustering data of individual knowledge module scores of different students, according to the individual knowledge module test change function corresponding to different student individuals Determine the overall test level value of the individual knowledge module of different student individuals , where ; with the test level value as the horizontal axis and the number distribution value as the vertical axis, according to the overall test level value of the individual knowledge module , perform a count of the number of people based on the level value to establish a distribution of the number of people at the knowledge module test level; set a clustering number division value. On the distribution of the number of people at the knowledge module test level, all student individuals with a number of people exceeding the clustering number division value are determined as the set of moderately performing student individuals corresponding to the knowledge module, and all student individuals with a number of people less than the clustering number division value and an overall test level value of the individual knowledge module less than the minimum overall test level value of the individual knowledge module in the set of moderately performing student individuals are determined as the set of poorly performing student individuals corresponding to the knowledge module, and all the remaining student individuals except the set of poorly performing student individuals and the set of moderately performing student individuals are determined as the set of excellent student individuals corresponding to the knowledge module.

[0043] A practical problem to be considered is that due to the different learning abilities of different student individuals, the absorption and mastery levels of knowledge modules will vary. Therefore, it is impossible to uniformly quantify the analysis and determination of the learning and mastery of knowledge modules. After classifying student individuals based on their learning abilities, it is necessary to determine student individuals with different learning abilities and then conduct more targeted analysis and processing of the learning and mastery of knowledge modules. Based on this, considering that the learning abilities of student individuals with different learning abilities can also be reflected by test scores, and the way of reflection is through the overall test level of student individuals. Therefore, after obtaining the equivalent scores of each student individual on the corresponding knowledge module each time, the change in the mastery of the knowledge module by the student individual is determined through fitting, and then the overall test level is determined through integration. The overall test level is used to divide student groups with different learning abilities for reasonable clustering. It should be noted here that for the clustering of student individuals with different learning abilities, this application is mainly divided into three categories: poor, medium, and excellent. On the one hand, considering that the test level is normally distributed in the group, the division into three categories can use the distribution estimation of the normal distribution to make the division more scientific. On the other hand, the division into three categories can avoid the complexity of subsequent analysis caused by too many clusters while fully meeting the requirements of clustering. The division value of the clustering number can be set according to the actual situation or determined through big data analysis.

[0044] S2: According to the effective data of group clustering, conduct an analysis of the individual performance stability of different student groups to form individual stability analysis result data.

[0045] According to the effective data of group clustering, conduct an analysis of the individual performance stability of different student groups to form individual stability analysis result data, including: conduct a stability analysis based on quantitative growth for different student individuals concentrated in poor-module students to form growth stability analysis result data corresponding to the student individuals; conduct a stability analysis based on fluctuating upward for different student individuals concentrated in medium-module students to form upward stability analysis result data corresponding to the student individuals; conduct a stability analysis based on quantitative fluctuation for different student individuals concentrated in excellent-module students to form quantitative stability analysis result data corresponding to the student individuals.

[0046] The analysis and processing of the learning and mastery of different knowledge modules by student individuals with different learning abilities have certain differences due to different learning abilities. Therefore, the analysis of stability needs to be considered separately, so as to more reasonably and accurately reflect the learning and mastery of knowledge modules by student individuals and provide appropriate learning guidance for them.

[0047] For different individual students concentrated in the module poor students, perform stability analysis based on quantitative growth to form the growth stability analysis result data corresponding to the individual students, including: for different individual students concentrated in the module poor students, according to the corresponding individual knowledge module test change function to determine the corresponding individual knowledge module test change rate function , where i is the number of different individual students in the module poor students concentrated under the knowledge module numbered m; set the growth compliance ratio β, for different individual students, according to the individual knowledge module test change rate function , perform the following growth stability analysis: if , then form the growth stability compliance information corresponding to the individual student; if , then form the growth instability information corresponding to the individual student, where represents the proportion of the span length of on the horizontal axis of the function in the total span length, and β > 50%.

[0048] For the individual students in the poor clustering, due to their own learning ability, it is considered necessary to take efforts to improve the learning and mastery of knowledge modules as the learning guidance. Therefore, the stability analysis of individual students is based on growth requirements. The growth compliance ratio can be determined according to the actual situation or through big data analysis. It should be noted that the setting of the growth compliance ratio is mainly to ensure that the learning and mastery of knowledge modules by individual students can reflect an increasing trend in most test results in terms of scores through the learning during the considered test time period. This indicates that the individual students in this clustering are working hard within the scope of their learning ability, and for the unqualified situation, it provides a reference to guide individual students to reasonably master the learning of knowledge modules.

[0049] For different individual students concentrated in the module medium students, perform stability analysis based on fluctuation upward to form the upward stability analysis result data corresponding to the individual students, including: for different individual students concentrated in the module medium students, according to the corresponding individual knowledge module test change function to determine the corresponding individual knowledge module test change rate function , where j is the number of different individual students in the module medium students concentrated under the knowledge module numbered m; set the upward compliance ratio γ and the maximum fluctuation limit , β > γ > 50%, for different individual students, according to the individual knowledge module test change function and the individual knowledge module test change rate function , perform the following fluctuation upward analysis: if , and , the fluctuating upward reaching standard information corresponding to the individual student is formed. represents the maximum difference in equivalent scores between two different tests; if , and , the fluctuating unstable information corresponding to the individual student is formed; if , and , the upward unstable information corresponding to the individual student is formed; if , and , the fluctuating upward unstable information corresponding to the individual student is formed.

[0050] For the individual students in the medium clustering, their learning ability is at the middle level. Therefore, in the analysis of stability, the focus is mainly on the stability of the fluctuating test scores and a certain degree of upward trend. So, two parameters, namely the upward reaching standard ratio and the maximum fluctuation limit, are set for reasonable limitation. It should be noted that the two parameters can restrict each other to determine the mastery situation of the individual students in this clustering. The upward reaching standard ratio can ensure that the learning and mastery of knowledge modules during the test period show a certain upward trend in terms of scores. The maximum fluctuation limit ensures that the upward trend will not be masked by the instability caused by large fluctuations. A reasonable small - amplitude fluctuation is the manifestation of stability. Of course, considering the upward trend of individual students in the poor clustering requires greater incentives. Therefore, the upward reaching standard ratio is less than the growth reaching standard ratio, but both exceed 50%.

[0051] For different individual students concentrated in the module excellent students, a stability analysis based on quantitative fluctuation is carried out to form the quantitative stability analysis result data corresponding to the individual students, including: for different individual students concentrated in the module excellent students, according to the corresponding individual knowledge module test change function , the corresponding individual knowledge module test change rate function is determined, where v is the number of different individual students in the module excellent students under the knowledge module numbered m; the quantitative reaching standard ratio ζ and the maximum stable fluctuation limit are set, ζ = 50%, , for different individual students, according to the individual knowledge module test change function and the individual knowledge module test change rate function , the following fluctuating upward analysis is carried out: if , and , the quantitative stable reaching standard information corresponding to the individual student is formed; if , and , the quantitative unstable information corresponding to the individual student is formed; if , and , the non - quantitative stable information corresponding to the individual student is formed; if , and Then, the indefinite and unstable information corresponding to the individual student is formed.

[0052] For the students in the excellent clustering, their learning and mastery of knowledge modules are already at a relatively high level. Therefore, the stability analysis carried out on them is mainly to ensure that there will be no serious decline during the test period. So, a quantitative compliance ratio of 50% is used as the constraint condition for the change rate of the test score, and at the same time, the maximum stable fluctuation limit restricts the fluctuation of their stability. Of course, considering that maintaining a relatively high level of learning and mastery of knowledge modules requires a certain learning ability, the maximum stable fluctuation limit is less than the maximum fluctuation limit.

[0053] S3: According to the individual stability analysis result data, perform knowledge module stability clustering to form individual knowledge module learning guidance data.

[0054] For an individual student, they will learn multiple knowledge modules. Therefore, the analysis of the learning and mastery of different knowledge modules is clustered based on the individual student, so as to provide feedback on the learning and mastery of all knowledge modules for the individual student, enabling the individual student to have a clear and distinct understanding of their overall learning situation and learning guidance for different knowledge modules.

[0055] The present invention also provides a student achievement analysis and management system based on artificial intelligence. The system includes: a data acquisition unit for acquiring historical test data of a target student group; a clustering analysis unit for obtaining the historical test data acquired by the data acquisition unit, performing clustering analysis to form group clustering effective data, and performing individual achievement stability analysis to form individual stability analysis result data; an output display unit for obtaining the individual stability analysis result data formed by the clustering analysis unit, performing knowledge module stability clustering to form individual knowledge module learning guidance data.

[0056] The system is configured to be able to acquire historical test data of a target student group and perform clustering analysis based on knowledge modules, realizing targeted knowledge module learning stability analysis for different individual students, providing a clear understanding of the knowledge mastery situation and appropriate improvement guidance as an organic whole, effectively ensuring a comprehensive understanding of the individual student's knowledge learning and mastery situation, and being an important material basis for realizing reasonable and efficient student achievement analysis and management.

[0057] In summary, the beneficial effects of the student achievement analysis and management method and system based on artificial intelligence provided by the embodiments of the present invention are as follows: This method collects the historical test data of the target student group, conducts clustering based on different knowledge modules, and then separately analyzes the performance stability of different student individuals for different knowledge modules. At the same time, considering the relative situation of the performance of different student individuals in the group as a whole, it reasonably conducts targeted processing of the mastery of knowledge modules. To a certain extent, it can provide a clear understanding and improvement guidance for the mastery of knowledge modules for student individuals in different performance states, ensuring that student individuals can achieve greater reasonable learning improvement and learning awareness while reasonably combining their own learning abilities, and promoting the good learning state of student individuals.

[0058] This system is configured to collect the historical test data of the target student group for clustering analysis based on knowledge modules, realize the targeted analysis of the learning stability of knowledge modules for different student individuals, and provide an organic whole of a clear understanding of the knowledge mastery situation and appropriate improvement guidance, effectively ensuring a comprehensive understanding of the knowledge learning and mastery situation of student individuals, and is an important material basis for realizing reasonable and efficient student performance analysis and management.

[0059] The above is only the specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present application can easily think of changes or substitutions, which should all be covered by the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.

Claims

1. A student performance analysis and management method based on artificial intelligence, characterized in that: include: Collect historical test data of the target student group and conduct cluster analysis of equivalent difficulty big data based on knowledge modules to form effective group clustering data; According to the effective data of group clustering, individual performance stability analysis is performed on different student groups to form individual stability analysis result data; Based on the individual stability analysis result data, knowledge module stability clustering is performed to form individual knowledge module learning guidance data.

2. The method for analyzing and managing student performance based on artificial intelligence according to claim 1, characterized in that: The historical test data of the target student group is collected, and the equivalent difficulty big data cluster analysis based on the knowledge module is performed to form group clustering effective data, including: According to the historical test data, clustering the scores of different individual students in the target student group based on knowledge modules to form individual knowledge module score clustering data; According to the individual knowledge module score clustering data corresponding to different individual students, clustering division based on score values ​​is performed to form group clustering effective data.

3. The method for analyzing and managing student performance based on artificial intelligence according to claim 2 is characterized in that: The step of clustering the scores of different individual students in the target student group based on the historical test data to form individual knowledge module score clustering data includes: Extracting the test scores of different knowledge modules of different individual students in the historical test data, and arranging them in order of time dimension to form a sequence set of individual knowledge module scores corresponding to different knowledge modules; The single test levels are determined for different knowledge modules, and according to the different single test levels, the equivalent score fitting analysis is performed on the different individual knowledge module score sequence sets to form corresponding individual knowledge module score change data.

4. The method for analyzing and managing student performance based on artificial intelligence according to claim 3 is characterized in that: The determination of the single test level for different knowledge modules and the equivalent score fitting analysis of the score sequence sets of different individual knowledge modules according to the different single test levels to form the corresponding individual knowledge module score change data include: For different knowledge modules, different levels are continuously assigned based on the level order according to the test difficulty level, and the difficulty equivalent values ​​corresponding to different levels are determined; Determine the difficulty equivalent value of each test according to the difficulty equivalent values ​​of different levels of the knowledge modules and in combination with the single test level corresponding to each test in the individual knowledge module score sequence set; For different knowledge modules, according to the individual knowledge module score sequence set of different individual students and the difficulty equivalent value of each test, the individual knowledge module equivalent score set corresponding to different individual students is determined. ,in, , m is the number of different knowledge modules, n is the number of different individual students, k is the number of different tests, , , is the test score obtained by the student with number n in the test with number k under the knowledge module with number m, is the difficulty equivalent value determined by the knowledge module numbered m in the test on the side numbered k, is the maximum value of the continuous assignment of the level corresponding to the knowledge module numbered m, Assign a value to the single test level corresponding to the knowledge module numbered m in the test numbered k; According to the equivalent score set of the individual knowledge modules corresponding to different individual students , fitting is performed on the time dimension sequence to form the test change function of the individual knowledge module .

5. The method for analyzing and managing student performance based on artificial intelligence according to claim 4 is characterized in that: The step of performing clustering based on the performance values ​​of the individual knowledge module performance clustering data corresponding to different individual students to form group clustering effective data includes: For the clustering data of the scores of different individual knowledge modules, the change function is tested according to the individual knowledge modules corresponding to different individual students. , determine the overall test level value of the individual knowledge module of different students ,in, ; The horizontal axis is the test level value, and the vertical axis is the number of people distribution value. According to the overall test level value of the individual knowledge module , conduct population statistics based on level values ​​and establish the distribution of people at the knowledge module test level; A cluster population division value is set. In the distribution of the number of people in the knowledge module test level, all the student individuals whose number exceeds the cluster population division value are determined as the module average student individual set corresponding to the knowledge module; all the student individuals whose number is less than the cluster population division value and whose individual knowledge module overall test level value is less than the minimum individual knowledge module overall test level value corresponding to the module average student individual set are determined as the module poor student individual set corresponding to the knowledge module; all the remaining student individuals except the module poor student individual set and the module average student individual set are determined as the module excellent student individual set corresponding to the knowledge module.

6. The method for analyzing and managing student performance based on artificial intelligence according to claim 5 is characterized in that: According to the effective data of group clustering, individual performance stability analysis is performed on different student groups to form individual stability analysis result data, including: Conducting stability analysis based on quantitative growth on different individual students in the module difference individual student set to form growth stability analysis result data corresponding to the individual students; For different individual students in the individual student set in the module, a stability analysis based on fluctuation floating is performed to form floating stability analysis result data corresponding to the individual students; A stability analysis based on quantitative fluctuation is performed on different individual students in the set of excellent individual students in the module to form quantitative stability analysis result data corresponding to the individual students.

7. The method for analyzing and managing student performance based on artificial intelligence according to claim 6 is characterized in that: The stability analysis based on quantitative growth is performed on different individual students in the module difference individual student set to form growth stability analysis result data corresponding to the individual students, including: For the different individual students in the module difference student individual set, the change function is tested according to the corresponding individual knowledge module , determine the corresponding individual knowledge module test change rate function , where i is the number of different individual students in the module difference student individual set under the knowledge module numbered m; Set the growth target proportion β, and test the change rate function for different individual students according to the individual knowledge module , the following growth stability analysis is performed: like , then the growth stability reaching standard information corresponding to the individual student is formed; like , then the growth instability information corresponding to the individual student is formed, where express The span length on the horizontal axis of the function accounts for the proportion of the total span length, β>50%.

8. The method for analyzing and managing student performance based on artificial intelligence according to claim 7 is characterized in that: The stability analysis based on fluctuation floating is performed on different individual students in the individual student set of the module to form floating stability analysis result data corresponding to the individual students, including: For different individual students in the individual student set in the module, the change function is tested according to the corresponding individual knowledge module. , determine the corresponding individual knowledge module test change rate function , where j is the number of different individual students in the individual student set in the module under the knowledge module numbered m; Set the floating target percentage γ and the maximum fluctuation limit , β>γ>50%, for different individual students, test the change function according to the individual knowledge module and the individual knowledge module test change rate function , perform the following fluctuation upward analysis: like ,and , then the fluctuation upward standard information corresponding to the individual student is formed, Indicates the maximum difference in equivalent scores between two different tests; like ,and , then the fluctuation and instability information corresponding to the individual student is formed; like ,and , then the upward unstable information corresponding to the individual student is formed; like ,and , then the fluctuation and floating unstable information corresponding to the individual student is formed.

9. The method for analyzing and managing student performance based on artificial intelligence according to claim 8 is characterized in that: The stability analysis based on quantitative fluctuation is performed on different individual students in the set of excellent individual students in the module to form quantitative stability analysis result data corresponding to the individual students, including: For different students in the module's excellent student individual set, test the change function according to the corresponding individual knowledge module , determine the corresponding individual knowledge module test change rate function , where v is the number of different individual students in the excellent individual student set of the module under the knowledge module numbered m; Set the quantitative compliance ratio ζ and the maximum stable fluctuation limit ,ζ=50%, , for different individual students, test the change function according to the individual knowledge module and the individual knowledge module test change rate function Perform the following fluctuation analysis: like ,and , then the quantitative stable standard-reaching information corresponding to the individual student is formed; like ,and , then the quantitative unstable information corresponding to the individual student is formed; like ,and , then the indefinite stable information corresponding to the individual student is formed; like ,and , then an indefinite amount of unstable information corresponding to the individual student is formed.

10. The student performance analysis and management system based on artificial intelligence is characterized by: include: A data collection unit for collecting historical test data of the target student group; A cluster analysis unit, which obtains the historical test data collected by the data collection unit, performs cluster analysis to form group clustering effective data, and performs individual performance stability analysis to form individual stability analysis result data; The output display unit is used to obtain the individual stability analysis result data formed by the cluster analysis unit, perform knowledge module stability clustering, and form individual knowledge module learning guidance data.

Images