An intelligent analysis system for student test scores

Abstract

The present invention relates to the technical field of performance data analysis, and specifically to a student test performance intelligent analysis system. The present invention obtains the independent learning performance trend of each student and the overall learning performance trend of all students in each test; analyzes the difference characteristics between the learning performance trends, and obtains the change correlation between the independent learning performance trend and the overall learning performance trend of each student in the current test; then obtains the performance influencing factor of each student in the current test; then adjusts the current original performance data of each student to obtain the current adjusted performance data of each student; clusters all students according to the changes in the current adjusted performance data and the adjacent previous original performance data to obtain student clustering clusters; and analyzes the test performance. The present invention improves the accuracy of the clustering effect and the effectiveness of the performance analysis by adaptively adjusting the test performance of students.

Description

Technical Field

[0001] The present invention relates to the technical field of score data analysis, and in particular to an intelligent analysis system for student test scores. Background Art

[0002] At this stage, in order to understand the learning and development status of a student group, it is necessary to conduct irregular examinations. By comprehensively analyzing the students' test scores each time, we can observe the students' learning changes and ensure that every student does not fall behind in the learning process. This not only improves the effectiveness of education management, but also promotes the development of students' self-cognition and provides a scientific basis for educational decision-making.

[0003] In the existing technology, students' learning trends are analyzed through their previous test scores, and K-means clustering is used to cluster the learning trends of the current students, so as to identify student groups with similar learning trends and carry out targeted management. However, since the fluctuation of students' grades is not only related to the students' recent learning situation, but also has a certain relationship with external factors such as the difficulty of the exam, clustering based solely on the fluctuation of academic performance will ignore the interference of the difficulty of the exam on the students' actual learning outcomes, leading to an incorrect analysis of students' academic performance. Summary of the Invention

[0004] In order to solve the technical problem that ignoring the difficulty of the exam interferes with students' actual learning outcomes, leading to erroneous analysis of students' academic performance, the present invention aims to provide an intelligent analysis system for student exam results. The technical solutions adopted are as follows:

[0005] The present invention proposes a student test score intelligent analysis system, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the following steps are implemented:

[0006] Get the original score data of each student in each test under a subject within the historical range of the current test;

[0007] Based on the changing trends and distribution characteristics of each student's original score data within the neighborhood of each exam, the independent learning performance trend of each student in each exam is obtained, and the overall learning performance trend of all students in each exam is obtained; based on the difference characteristics between the independent learning performance trend and the overall learning performance trend of each student in each exam within the historical range of the current exam, the changing correlation between the independent learning performance trend and the overall learning performance trend of each student in the current exam is obtained;

[0008] Obtain each student's score impact factor for the current exam based on the trend of each student's independent learning performance in the current exam and the corresponding change correlation; adjust each student's current original score data based on the score impact factor for the current exam and the change characteristics of all students' original score data to obtain each student's current adjusted score data; cluster all students based on the change of each student's current adjusted score data and the previous original score data to obtain student clusters;

[0009] The test scores are analyzed according to the student clusters.

[0010] Furthermore, the method for obtaining the independent learning achievement trend includes:

[0011] Construct a line graph of the original score data of all exams within the neighborhood of each exam; obtain the angle between the line segments between adjacent original score data and the horizontal direction;

[0012] After normalizing the angles between all adjacent original score data, the mean of all normalized mapping results is calculated as the overall score change trend;

[0013] Adjust the differences between the same adjacent original score data according to the distribution of the differences between the adjacent original score data to obtain the average score change level;

[0014] Based on the overall score change trend and average score change level of each student, the independent learning effectiveness trend of each student in each exam was obtained. The overall score change trend and average score change level were positively correlated with the independent learning effectiveness trend.

[0015] Furthermore, the method for obtaining the average score change level includes:

[0016] In the neighborhood of each test, the difference between the original score data of the next test and the previous test is calculated to obtain the difference sequence corresponding to the adjacent original score data;

[0017] The number of differences after the corresponding difference of each adjacent original score data in the difference sequence is counted, and negative correlation mapping is performed as the weighted value of the difference corresponding to each adjacent original score data; the difference between adjacent original score data is weighted averaged according to the weighted value of the difference corresponding to each student's different adjacent original score data to obtain the average score change level.

[0018] Furthermore, the method for obtaining the overall learning achievement trend includes:

[0019] The independent learning performance trends of all students in each exam are averaged to obtain the overall learning performance trend of all students in each exam.

[0020] Furthermore, the method for obtaining the change correlation includes:

[0021] Based on the change difference between each student's independent learning achievement trend and the overall learning achievement trend in the current exam and the matching distance between the independent learning achievement trend and the overall learning achievement trend in all exams within the historical range of the current exam, the change correlation between each student's independent learning achievement trend and the overall learning achievement trend in the current exam is obtained. The change difference and matching distance are both negatively correlated with the change correlation.

[0022] Furthermore, the method for obtaining the performance influencing factor includes:

[0023] Based on the trend of each student's independent learning performance in the current exam and the corresponding change correlation, the performance impact factor of each student in the current exam is obtained. The trend and change correlation of independent learning performance are both positively correlated with the performance impact factor.

[0024] Furthermore, the method for obtaining the current adjusted score data includes:

[0025] Determine whether each student's current raw score data needs to be adjusted based on the change characteristics of all students' raw score data. If adjustment is necessary, obtain the adjustment factor based on each student's score impact factor under the current exam and the current raw score data;

[0026] After calculating the sum of each student's current original score data and the adjustment factor, normalization mapping is performed to obtain each student's current adjusted score data.

[0027] Furthermore, judging whether it is necessary to adjust the current original score data of each student according to the change characteristics of the original score data of all students includes:

[0028] According to the central tendency of the original score data of all students in each test, the overall score level of each test is obtained;

[0029] If the overall score level of the current exam is not equal to the overall score level of the exam before the current exam, it is determined that the current original score data needs to be adjusted; otherwise, no adjustment is required.

[0030] Furthermore, the method for obtaining the adjustment factor includes:

[0031] Calculate the product of each student's score impact factor under the current exam and the current original score data as the first product;

[0032] If the overall performance level of the current test is greater than the overall performance level of the test before the current test, the first product is negated and the obtained value is used as the adjustment factor;

[0033] If the overall performance level of the current test is lower than the overall performance level of the test before the current test, the first product is used as an adjustment factor.

[0034] Furthermore, the method for obtaining the student clusters includes:

[0035] The difference between each student's current adjusted score data and the previous original score data is calculated, and K-means clustering is performed on all students to obtain student clusters.

[0036] The present invention has the following beneficial effects:

[0037] The present invention obtains the independent learning achievement trend of each student in each exam based on the change trend and distribution characteristics of the original score data of each student in the neighborhood range of each exam, and obtains the overall learning achievement trend of all students in each exam. The independent learning achievement trend reflects the learning situation and progress speed of individual students, and the overall learning achievement trend reflects the learning level and development trend of all students. In order to prevent students' scores from being affected by the difficulty of the exam and unable to reflect their true learning level, the present invention obtains the change correlation between the independent learning achievement trend and the overall learning achievement trend of each student in each exam in the historical range of the current exam, reveals the correlation between individual students and overall learning achievement, and evaluates the influence of the difficulty of the exam on individual students. The invention can be used to analyze the influence of factors; obtain the performance influence factor of each student in the current exam according to the trend of each student's independent learning performance in the current exam and the corresponding change correlation, quantify the degree of influence of the student's learning performance on the performance in the current exam, and help students understand the specific impact of the current learning status on the performance; adjust the current original performance data of each student according to the performance influence factor of each student in the current exam and the change characteristics of the original performance data of all students, obtain the current adjusted performance data of each student, and reflect the student's real learning performance and progress; cluster all students according to the change of the current adjusted performance data and the adjacent previous original performance data, obtain student clustering clusters, identify student groups with similar learning performance trends and performance change characteristics, and conduct targeted analysis of students with different learning change situations; and analyze the test scores. The present invention improves the accuracy of clustering effect and the effectiveness of performance analysis by adaptively adjusting the test scores of students. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] In order to more clearly illustrate the technical solutions and advantages of the embodiments of the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the prior art descriptions. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0039] Figure 1 A flowchart of an implementation method of a student test score intelligent analysis system provided by one embodiment of the present invention;

[0040] Figure 2 A schematic diagram of a student test score intelligent analysis system provided by one embodiment of the present invention. DETAILED DESCRIPTION

[0041] To further illustrate the technical means and effectiveness of the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, describes in detail the specific implementation, structure, features, and effectiveness of a student test score intelligent analysis system proposed in accordance with the present invention. In the following description, different references to "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics of one or more embodiments may be combined in any suitable manner.

[0042] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0043] The specific scheme of the intelligent analysis system for student test scores provided by the present invention is described in detail below with reference to the accompanying drawings.

[0044] See also Figure 1 , which shows a flow chart of an implementation method of a student test score intelligent analysis system provided by an embodiment of the present invention. The embodiment of the present invention provides a student test score intelligent analysis system, see Figure 2 , including a memory 201, a processor 202, and a computer program 203 stored in the memory 201 and executable on the processor 202. When the processor 202 executes the computer program 203, the following method steps may be implemented. The flowchart corresponding to the method steps is as follows: Figure 1 As shown, in the specific implementation, the method specifically includes:

[0045] Step S1: Obtain the original score data of each student in each exam under a subject within the historical range of the current exam.

[0046] In an embodiment of the present invention, taking into account the fluctuations in students' test scores caused by the difficulty of the exam, in order to avoid misjudgment of students' learning outcomes, timely discover students' academic performance, and provide timely feedback and guidance for teaching, it is necessary to analyze the test scores of all students; first, the student learning status after each exam is evaluated, and the test scores of all students are obtained from the student test score recording system. The scores of one subject are summarized together and analyzed separately to obtain the original score data of each student in each exam under one subject within the historical range of the current exam.

[0047] It should be noted that, in one embodiment of the present invention, the historical range of the current exam is a time range consisting of the current exam and the corresponding historical exams recorded in the student exam score recording system.

[0048] It should be noted that in order to facilitate the subsequent processing of the score data, all original score data are normalized, and the obtained values are used as new original score data for subsequent calculations and cleaning, which helps to remove invalid or erroneous records, handle missing values, and ensure the accuracy and completeness of the data.

[0049] Step S2: Obtain the independent learning performance trend of each student in each exam based on the change trend and distribution characteristics of the original score data of each student in the neighborhood range of each exam, and obtain the overall learning performance trend of all students in each exam; obtain the change correlation between the independent learning performance trend and the overall learning performance trend of each student in each exam within the historical range of the current exam.

[0050] By analyzing a student's historical performance data, we can see whether their grades are trending upward or downward. This trend can intuitively reflect a student's learning motivation and progress. The closer the test scores are to each exam, the more reliable the analysis of their learning status is, and the more accurately they reflect each student's true learning performance trends. Therefore, based on the changing trends and distribution characteristics of each student's raw performance data within the neighborhood of each exam, we can derive each student's independent learning performance trends for each exam.

[0051] Preferably, in one embodiment of the present invention, the method for obtaining the independent learning achievement trend includes:

[0052] Within the neighborhood of each exam, a line graph of the raw score data for all exams is constructed; the angle between the line segments of adjacent raw score data and the horizontal direction is obtained; after normalizing the angles between all adjacent raw score data, the mean of all normalized mapping results is calculated as the overall score change trend; the differences between the same adjacent raw score data are adjusted based on the time series distribution of the differences between adjacent raw score data to obtain the average score change level;

[0053] Based on each student's overall score trend and average score change, we determined their independent learning performance trends for each exam. Both overall and average score trends were positively correlated with independent learning performance trends. A positive correlation indicates that the dependent variable increases as the independent variable increases, and decreases as the independent variable decreases. The specific relationship can be multiplication, addition, or the idempotence of an exponential function, determined by actual application.

[0054] Preferably, in one embodiment of the present invention, the method for obtaining the average score change level includes:

[0055] In the neighborhood of each test, the difference between the original score data of the next test and the previous test is calculated to obtain the difference sequence corresponding to the adjacent original score data;

[0056] The number of differences after the corresponding difference of each adjacent original score data in the difference sequence is counted, and negative correlation mapping is performed as the weighted value of the difference corresponding to each adjacent original score data; the difference between adjacent original score data is weighted averaged according to the weighted value of the difference corresponding to each student's different adjacent original score data to obtain the average score change level.

[0057] In one embodiment of the present invention, the formula for the independent learning achievement trend is expressed as:

[0058]

[0059] Among them, T i represents the trend of the independent learning performance of the i-th student in each exam; N represents the number of differences between adjacent original score data within the neighborhood of each exam; θ i，n represents the angle between the line segment of the nth adjacent original score data of the i-th student and the horizontal direction; g(θ i，n ) represents the angle θ i，n Normalized value; D i，n It represents the number of differences after the nth adjacent original score data corresponding to the difference in the difference sequence of the i-th student in the neighborhood range of each test; B i，n Represents the difference between the nth adjacent original score data of the i-th student.

[0060] In the formula of independent learning performance trend, since the angle can be positive or negative, g(θ i，n ) means normalizing the maximum and minimum values of the angles, scaling all angles to the range of 0-1; Indicates calculating the mean of the normalized angles between all adjacent raw score data as the overall score change trend. The larger the angle, the larger the mean of the normalized angle, and the greater the overall score change trend. It represents the level of change in average score. The greater the difference between adjacent original score data, the greater the trend of change in student scores. For the i-th student, the greater the number of differences after the corresponding difference of the n-th adjacent original score data in the difference sequence within the neighborhood range of each exam, the farther the analyzed change trend is from the current original score data, and the lower the credibility of the analysis of the student's current learning status. On the contrary, the smaller the number of differences, the closer it is to the current original score data, the higher the credibility of the analysis of the student's current learning status, and the more necessary it is to increase the change trend to more accurately measure the trend of each student's independent learning effectiveness. Therefore, the larger the angle, the greater the overall score change trend, the greater the level of average score change, and the corresponding greater independent learning effectiveness trend.

[0061] It should be noted that, in one embodiment of the present invention, the neighborhood range of each exam is the range formed by each exam and the corresponding historical exams in the student exam score recording system. In other embodiments of the present invention, the size of the neighborhood range can be specifically set according to specific circumstances and is not limited or elaborated herein.

[0062] It should be noted that, in one embodiment of the present invention, the method of calculating the number of differences after the corresponding difference of each adjacent original score data in the difference sequence and performing negative correlation mapping is to calculate the inverse of the number of differences obtained by statistics and perform negative correlation mapping, and construct a negative correlation relationship in which the larger the number of differences, the farther from the current original score data, and the lower the credibility of the analysis of the current original score data; in other embodiments of the present invention, the number of differences obtained by statistics can be used to construct a correlation relationship in which the larger the number of differences, the farther from the current original score data, and the lower the credibility of the analysis of the current original score data; the order of the corresponding difference of each adjacent original score data in the difference sequence can also be directly normalized as a weighted value; the smaller the order, the farther from the current original score data, and the lower the credibility of the analysis of the current original score data. The specific means are technical means well known to those skilled in the art and will not be elaborated here.

[0063] It should be noted that when realizing a negative correlation, if the reciprocal is used, it is necessary to add an artificial threshold to the denominator of the fraction, such as a value of 0.01, to avoid the situation where the denominator of the fraction is 0.

[0064] By analyzing the overall learning performance trend, you can fully understand students' learning needs and performance. Get the overall learning performance trend of all students in each exam.

[0065] Preferably, in one embodiment of the present invention, the method for obtaining the overall learning achievement trend includes:

[0066] The independent learning performance trends of all students in each exam are averaged to obtain the overall learning performance trend of all students in each exam.

[0067] It should be noted that in other embodiments of the present invention, other methods such as the median or mode of the independent learning performance trends of all students can also be selected to represent the overall learning performance trend of all students. The specific means are technical means well known to those skilled in the art and will not be elaborated here.

[0068] To eliminate the impact of fluctuations in exam difficulty on student learning outcomes and accurately assess student performance, we analyze the differences between each student's independent learning performance trend and their overall learning performance trend. The smaller the difference between a student's learning performance trend and the overall learning performance trend of all students, the higher the consistency, reflecting the general degree to which external factors such as exam difficulty influence changes in that student's performance relative to the overall student population. This provides a better understanding of students' learning status and development trends. Therefore, based on the differences between each student's independent learning performance trend and their overall learning performance trend for each exam within the current exam's historical range, we obtain a correlation between the changes in each student's independent learning performance trend and their overall learning performance trend for the current exam.

[0069] Preferably, in one embodiment of the present invention, the method for obtaining change correlation includes:

[0070] Based on the change difference between each student's independent learning performance trend and overall learning performance trend for the current exam, as well as the matching distance between the independent learning performance trend and overall learning performance trend for all exams within the current exam's history, the change correlation between each student's independent learning performance trend and overall learning performance trend for the current exam is obtained. Both the change difference and the matching distance are negatively correlated with the change correlation. A negative correlation indicates that the dependent variable decreases as the independent variable increases, and the dependent variable increases as the independent variable decreases. This can be a subtractive relationship, a divisive relationship, or other relationships, determined by actual application.

[0071] In one embodiment of the present invention, the formula for change correlation is expressed as:

[0072]

[0073] Among them, x represents the serial number of the current test; Rix represents the correlation between the independent learning performance trend and the overall learning performance trend of the i-th student in the current exam; T ix represents the trend of independent learning performance of the i-th student under the current exam; T x Indicates the overall learning performance trend of all students under the current test; represents the trend of independent learning performance of the i-th student in all exams within the historical range of the current exam; Indicates the overall learning performance trend of all students in all exams within the historical range of the current exam; It represents the DTW distance between the independent learning performance trend of each student and the overall learning performance trend of all students in all exams within the historical range of the current exam, that is, the matching distance; exp() represents the exponential function with a natural constant as the base; sigmoid{} represents the logistic function.

[0074] In the formula of the change correlation, the exponential function with the natural constant as the base is used to convert Perform negative correlation mapping, |T ix -T x | represents the difference between the trend of the i-th student's independent learning performance under the current exam and the overall learning performance trend. The larger the difference, the greater the difference between the corresponding student and the overall learning performance trend of all students, the smaller the correlation of the changes, and the less likely it is to be affected by the difficulty of the exam. The larger it is, the greater the matching distance between the independent learning achievement trend and the overall learning achievement trend of students within the historical range, the smaller the correlation, and the smaller the change correlation.

[0075] It should be noted that, in one embodiment of the present invention, the matching distance is obtained by calculating the DTW distance, and a negative correlation mapping is performed on the DTW distance to establish a correlation relationship in which the larger the DTW distance, the larger the matching distance, and the smaller the change correlation; in other embodiments of the present invention, the negative correlation mapping of the DTW distance can also be performed directly by calculating the Pearson correlation coefficient between the independent learning performance trend of each student and the overall learning performance trend of all students within the historical range to establish a correlation relationship in which the larger the correlation coefficient, the greater the change correlation. The specific DTW distance and correlation coefficient are technical means well known to those skilled in the art and will not be elaborated here.

[0076] It should be noted that, in other embodiments of the present invention, Alternatively, normalization and addition may be performed first and then negative correlation mapping may be performed to construct a negative correlation relationship in which the greater the change difference and matching distance, the smaller the change correlation. The specific means are well known to those skilled in the art and will not be elaborated here.

[0077] Step S3: Obtain each student's score influence factor in the current exam based on the trend of each student's independent learning performance in the current exam and the corresponding change correlation; adjust each student's current original score data based on the score influence factor in the current exam and the change characteristics of all students' original score data to obtain each student's current adjusted score data; cluster all students based on the changes in each student's current adjusted score data and the adjacent previous original score data to obtain student clusters.

[0078] Each student or group has different criteria for evaluating exam difficulty, making it impossible to determine test difficulty based on subjective factors. Therefore, using the overall learning performance trend of all students as a benchmark provides a relatively objective evaluation standard that reflects the general level of performance of most students after being affected. Analyzing the correlation between each student's learning performance trend and the overall learning performance trend reveals the inherent connection between exam difficulty and student scores: the greater the correlation, the greater the impact, and the smaller the correlation, the smaller the impact. Combined with each student's individual learning performance trend, this intuitively reflects their learning status and outcomes, allowing for a comprehensive analysis of the impact of exam difficulty on their performance. Therefore, based on each student's individual learning performance trend and the corresponding correlation, we determine the performance impact factor for each student on the current exam.

[0079] Preferably, in one embodiment of the present invention, the method for obtaining the performance influencing factor includes:

[0080] Based on each student's independent learning performance trend and corresponding change correlation in the current exam, we obtain each student's score impact factor for the current exam. Both the independent learning performance trend and the change correlation are positively correlated with the score impact factor. A positive correlation indicates that the dependent variable increases as the independent variable increases, and decreases as the independent variable decreases. The specific relationship can be multiplication, addition, or the idempotence of an exponential function, determined by actual application.

[0081] In one embodiment of the present invention, the formula for the performance impact factor is expressed as:

[0082] Y ix =R ix ×T ix ;

[0083] Among them, x represents the serial number of the current exam; Y ix R represents the influencing factor of the i-th student's performance in the current exam; ix T represents the correlation between the independent learning performance trend and the overall learning performance trend of the i-th student at the current moment; ixRepresents the trend of independent learning performance of the i-th student under the current exam.

[0084] In the formula of the grade influencing factor, in the current exam, when the correlation between the independent learning achievement trend of the i-th student and the change in the overall learning achievement trend is high, the student is more affected by the difficulty of the exam, the greater the impact on the grade, and the greater the degree of adjustment that should be made; when the correlation is smaller, it means that the student is less affected by the fluctuation of the difficulty of the exam, the smaller the impact on the grade, and the smaller the degree of adjustment that should be made; the greater the independent learning achievement trend of the i-th student, the more the overall trend of the student's grade change is upward, the greater the degree of adjustment for this grade, reflecting the actual learning level; the smaller the independent learning achievement trend of the i-th student, the more the trend of the student's grade change is downward, and the smaller the degree of adjustment for the current exam grade.

[0085] It should be noted that in other embodiments of the present invention, other basic mathematical operations such as addition can also be used to construct a positive correlation in which the greater the trend of independent learning effectiveness, the greater the corresponding change correlation, and the greater the performance influencing factor. The specific means are technical means well known to those skilled in the art and will not be elaborated here.

[0086] By considering the changes in the original score data of all students in each exam, the dynamic changes in students' learning effects over a period of time can be intuitively reflected; by analyzing the score influencing factors of each student, a comprehensive and quantitative learning effectiveness evaluation indicator is provided, which can more accurately understand the students' learning status and development trends, and then adjust the original score data to avoid score deviations caused by factors such as the difficulty of the exam, and obtain adjusted score data that is more in line with the students' actual learning level; in order to more accurately evaluate students' recent learning performance and reduce the impact of fluctuations in exam difficulty, a comprehensive analysis of the change level of the original score data and the score influencing factors is performed; therefore, according to the changes in the original score data of all students in each exam and the score influencing factors of each student, the current original score data of each student is adjusted to obtain the current adjusted score data of each student.

[0087] Preferably, in one embodiment of the present invention, the method for obtaining the current adjusted performance data includes:

[0088] Determine whether each student's current raw score data needs to be adjusted based on the change characteristics of all students' raw score data. If adjustment is necessary, obtain the adjustment factor based on each student's score impact factor under the current exam and the current raw score data;

[0089] After calculating the sum of each student's current original score data and the adjustment factor, normalization mapping is performed to obtain each student's current adjusted score data.

[0090] Preferably, in one embodiment of the present invention, judging whether it is necessary to adjust the current original score data of each student according to the change characteristics of the original score data of all students includes:

[0091] According to the central tendency of the original score data of all students in each test, the overall score level of each test is obtained;

[0092] If the overall score level of the current exam is not equal to the overall score level of the exam before the current exam, it is determined that the current original score data needs to be adjusted; otherwise, no adjustment is required.

[0093] It should be noted that comparing the overall score levels between exams can indicate the changing characteristics of the original score data of all students. The differences between the overall score levels indicate that the original score data of all students between different exams have changed as a whole, which will be affected by the difficulty of the exam and therefore needs to be adjusted.

[0094] It should be noted that, in one embodiment of the present invention, the central tendency of the original score data of all students can be expressed by calculating the mean of the original score data of all students to obtain the overall score level of all students. The larger the mean, the greater the central tendency of the original score data of all students, that is, the greater the overall score level. In other embodiments of the present invention, the mode or median of the original score data of all students can also be selected to reflect the central tendency and level of the scores of all students. The specific means are technical means well known to those skilled in the art and will not be elaborated here.

[0095] Preferably, in one embodiment of the present invention, the method for obtaining the adjustment factor includes:

[0096] Calculate the product of each student's score impact factor under the current exam and the current original score data as the first product;

[0097] If the overall score level of the current exam is greater than the overall score level of the exam before the current exam, the first product is negated and the obtained value is used as the adjustment factor; if the overall score level of the current exam is less than the overall score level of the exam before the current exam, the first product is used as the adjustment factor.

[0098] In one embodiment of the present invention, the formula for adjusting the current performance data is expressed as:

[0099]

[0100] Among them, x represents the serial number of the current exam; Z ix Represents the current adjusted grade data of the i-th student; It represents the mean of the original score data of all students in the current exam, that is, the overall score level in the current exam; It represents the mean of the original score data of all students in the previous exam, that is, the overall score level in the previous exam; Q ix represents the original score data of the i-th student; Y ix represents the influencing factor of the i-th student's performance in the current exam; sigmoid{} represents the logistic function.

[0101] In the current formula for adjusting grade data, Y ix ×Q ix Represents the first product. The larger the score impact factor, the greater the deviation, and the larger the adjustment of the original score data is required; the smaller the score impact factor, the smaller the deviation, and the smaller the adjustment of the original score data is required; if the overall score level of the current exam is lower than the overall score level of the previous exam, it means that the overall score trend of all students is declining, and the current original score data needs to be adjusted upward to reflect the actual learning level, with Y ix ×Q ix As an adjustment factor, the larger the score impact factor, the larger the adjustment factor, and the larger the current adjusted score data; the smaller the score impact factor, the smaller the adjustment factor, and the smaller the current adjusted score data; if the overall score level of the current exam is greater than the overall score level of the previous exam, it means that the overall score trend of all students is rising, and the current original score data needs to be adjusted downward by -(Y ix ×Q ix ) as an adjustment factor, the larger the score impact factor is, the larger the first product is, and the larger the current original product data is adjusted downward; conversely, the smaller the score impact factor is, the smaller the first product is, the smaller the current original product data is adjusted downward, and the original score data is closer to the student's true level.

[0102] By comparing students' current adjusted grade data with their previous original grade data, we can identify patterns in student grade changes, which may reflect students' learning progress, stability, and decline. Clustering can classify students with similar learning status, reduce the workload of teachers in dealing with individual differences among students, and provide more targeted guidance and support for students in each cluster. Therefore, according to the difference between each student's current adjusted grade data and the adjacent previous original grade data, all students are clustered to obtain student clusters.

[0103] Preferably, in one embodiment of the present invention, the method for obtaining student clusters includes:

[0104] The difference between each student's current adjusted score data and the previous original score data is calculated, and K-means clustering is performed on all students to obtain student clusters.

[0105] The K-means clustering algorithm clusters multiple cluster objects into K specified clusters based on their similarities. Each cluster object belongs to, and only belongs to, one cluster with the smallest distance to the center of the student cluster. It should be noted that in one embodiment of the present invention, the number of clusters is pre-set to 3 by the implementer based on actual circumstances. In other embodiments of the present invention, the number of clusters can be specifically set based on specific circumstances, and this is not limited or elaborated upon here. The specific K-means clustering algorithm is a technical means well known to those skilled in the art and will not be elaborated upon here.

[0106] Step s4: Analyze the test scores according to the student clusters.

[0107] Different student clusters represent different learning needs and characteristics. Cluster analysis can more clearly identify these differences. The results of cluster analysis can help students understand their own learning characteristics within the overall population, allowing them to develop personalized learning paths that better suit their learning styles and needs. Therefore, we analyze test scores based on student clusters.

[0108] It should be noted that after obtaining student clusters, student performance can be analyzed. This includes calculating the mean difference between the current adjusted performance data and the previous original performance data for all students within each cluster, which serves as the intra-cluster performance mean. The clusters are then arranged in descending order to represent students with progressing, stable, and declining learning states. Based on each student's learning outcomes, student problems can be promptly identified and personalized learning strategies can be developed. For example, if a student is making progress, the status quo can be maintained; if a student is stable, the status quo can be maintained while considering strategies for improvement; and if a student is declining, the individual or external factors can be promptly analyzed and changes can be made. Subsequently, individual subject performance can be analyzed to promptly identify changes in student performance across each subject.

[0109] In summary, the present invention obtains the independent learning achievement trend of each student in each exam based on the change trend and distribution characteristics of the original score data of each student in the neighborhood range of each exam, and obtains the overall learning achievement trend of all students in each exam; obtains the change correlation between the independent learning achievement trend and the overall learning achievement trend of each student in the current exam based on the difference characteristics between the independent learning achievement trend and the overall learning achievement trend of each student in each exam within the historical range of the current exam; and then obtains the performance impact factor of each student in the current exam; adjusts the current original score data of each student in combination with the change characteristics of the original score data of all students to obtain the current adjusted score data of each student; clusters all students according to the change of each student's current adjusted score data and the adjacent previous original score data to obtain student clustering clusters; and analyzes the exam scores. The present invention improves the accuracy of clustering effect and the effectiveness of score analysis by adaptively adjusting students' scores.

[0110] It should be noted that the order in which the embodiments of the present invention are described above is for illustrative purposes only and does not necessarily represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the specific order or sequential order shown to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0111] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on the differences from other embodiments.

Claims

1. A student test score intelligent analysis system, comprising a computer program stored in a memory and executable on a processor, characterized in that: When the processor executes the computer program, the following steps are implemented: Get the original score data of each student in each test under a subject within the historical range of the current test; Based on the change trend and distribution characteristics of each student's original score data within the neighborhood of each exam, the independent learning achievement trend of each student in each exam is obtained, and the overall learning achievement trend of all students in each exam is obtained; the change correlation between the independent learning achievement trend and the overall learning achievement trend of each student in the current exam is obtained. The formula for the change correlation is expressed as: ;in, Indicates the serial number of the current exam; Indicates the current exam The correlation between the trend of individual students' independent learning performance and the overall learning performance trend; Indicates the The trend of students' independent learning performance under the current exam; Indicates the overall learning performance trend of all students under the current test; Indicates the The trend of a student's independent learning performance in all exams within the current exam history; Indicates the overall learning performance trend of all students in all exams within the historical range of the current exam; The DTW distance between the independent learning performance trend of each student and the overall learning performance trend of all students in all exams within the historical range of the current exam, that is, the matching distance; represents an exponential function with a natural constant as its base; represents the logistic function; The performance impact factor of each student in the current exam is obtained based on the trend of each student's independent learning performance in the current exam and the corresponding change correlation. The formula of the performance impact factor is expressed as follows: ,in, Indicates the influencing factor of the i-th student's performance in the current exam; According to the score influence factor of each student in the current exam and the change characteristics of the original score data of all students, the current original score data of each student is adjusted to obtain the current adjusted score data of each student. The formula of the current adjusted score data is expressed as follows: ;in, Indicates the Current adjusted grade data for each student; Represents the mean of the original score data of all students in the current exam; It represents the mean of the original score data of all students in the previous exam of the current exam; Indicates the The original performance data of each student; According to the changes in each student's current adjusted score data and the previous original score data, all students are clustered to obtain student clusters for analysis of test scores.

2. A student test score intelligent analysis system according to claim 1, characterized in that: The method for obtaining the independent learning achievement trend includes: Construct a line graph of the original score data of all exams within the neighborhood of each exam; obtain the angle between the line segments between adjacent original score data and the horizontal direction; After normalizing the angles between all adjacent original score data, the mean of all normalized mapping results is calculated as the overall score change trend; Adjust the differences between the same adjacent original score data according to the distribution of the differences between the adjacent original score data to obtain the average score change level; Based on the overall score change trend and average score change level of each student, the independent learning effectiveness trend of each student in each exam was obtained. The overall score change trend and average score change level were positively correlated with the independent learning effectiveness trend.

3. The intelligent analysis system for student test scores according to claim 2, characterized in that: The method for obtaining the average score change level includes: In the neighborhood of each test, the difference between the original score data of the next test and the previous test is calculated to obtain the difference sequence corresponding to the adjacent original score data; The number of differences after the corresponding difference of each adjacent original score data in the difference sequence is counted, and negative correlation mapping is performed as the weighted value of the difference corresponding to each adjacent original score data; the difference between adjacent original score data is weighted averaged according to the weighted value of the difference corresponding to each student's different adjacent original score data to obtain the average score change level.

4. The intelligent analysis system for student test scores according to claim 1, characterized in that: The method for obtaining the overall learning effectiveness trend includes: The independent learning performance trends of all students in each exam are averaged to obtain the overall learning performance trend of all students in each exam.

5. The intelligent analysis system for student test scores according to claim 1, characterized in that: The method for obtaining the student clusters includes: The difference between each student's current adjusted score data and the previous original score data is calculated, and K-means clustering is performed on all students to obtain student clusters.

Images