Student scientific literacy digital tracking evaluation method and system based on reinforcement learning

Abstract

The invention relates to the technical field of digital tracking evaluation, and particularly discloses a student scientific literacy digital tracking evaluation method and system based on reinforcement learning, and the method comprises the steps: obtaining the processing values of all types of verification index data of each student in each week based on all types of verification index data of each student in each week, obtaining evaluation priorities of all types of verification index data of each student in each week and all analysis weeks of each student in each week, and obtaining a verification index graph of each student in each week based on the evaluation priorities of all types of verification index data of each student in each week; and based on the verification index graph of each student in each week and the verification index graphs of all analysis weeks, obtaining a scientific literacy digital tracking evaluation result of each student. According to the invention, the evaluation strategy is adjusted in time according to the learning progress performance of the student in real time, the actual scientific literacy level of the student is quantified more accurately, and too high or too low evaluation on some students is avoided.

Description

Technical Field

[0001] The present invention relates to the technical field of digital tracking and evaluation, and in particular to a digital tracking and evaluation method and system for students' scientific literacy based on reinforcement learning. Background Art

[0002] At present, the digital tracking and evaluation system of students' scientific literacy is a method of evaluating students based on various verification indicator data in the process of cultivating scientific literacy. In the traditional scientific literacy evaluation system, the evaluation is often based on fixed standards and static data, which cannot fully capture the dynamic changes of students in the learning process. This "one-size-fits-all" evaluation method may not only ignore the individual differences of students, but also lead to over- or under-evaluation of some students due to the inability to adjust the evaluation strategy in time. The learning process of each student is a dynamic process. They will continue to accumulate knowledge, improve their abilities, and gradually form their own scientific literacy system in the learning process. Therefore, when evaluating students' scientific literacy, this dynamic change needs to be taken into account. Therefore, how to adjust the evaluation strategy in time according to the learning progress and performance of each student to ensure that the evaluation can always be close to the actual level of the students has become an important direction for the reform of the current education evaluation system. The Q-learning algorithm in reinforcement learning technology is a model-free reinforcement learning method based on the Markov decision process. It can learn a state-action value function (Q function), select the optimal action according to the current state, and update the Q function to gradually optimize the strategy. Combining the digital tracking and evaluation system of students' scientific literacy with reinforcement learning is an effective solution.

[0003] However, there is no digital tracking and evaluation method and system for students' scientific literacy based on reinforcement learning to adjust the evaluation strategy in real time according to students' learning progress, more accurately reflect students' actual scientific literacy level, and avoid over- or under-evaluation of some students.

[0004] Therefore, the present invention proposes a digital tracking and evaluation method and system for students' scientific literacy based on reinforcement learning. Summary of the invention

[0005] The present invention provides a method and system for digital tracking and evaluation of students' scientific literacy based on reinforcement learning, which is used to obtain the processed numerical value of all the verification indicator data of each student in each week according to all the verification indicator data of each student in each week, so as to quantify the score performance of each student in each verification indicator of each week, and then obtain the evaluation priority of all the verification indicator data of each student in each week according to the processed numerical value of all the verification indicator data of each student in each week, accurately quantify the relative importance or priority of each verification indicator of each student in each week to the overall evaluation of students' scientific literacy, obtain the verification indicator matrix of each student in each week according to the evaluation priority of all the verification indicator data of each student in all the reference weeks of each week, so as to facilitate the subsequent screening and analysis of the week, and obtain the evaluation priority of each student in each week according to the verification indicator matrix of each student in each week. For all the analysis weeks in the week, according to the processed values ​​of all types of verification indicator data of each student in each week and the evaluation priority of all types of verification indicator data, each student's weekly verification indicator graph is obtained, and the graphical representation of each student's overall scientific literacy evaluation in each week is accurately drawn. Furthermore, according to each student's weekly verification indicator graph and the verification indicator graphs of all the analysis weeks, each student's weekly digital evaluation score of scientific literacy is obtained, and the overall evaluation of scientific literacy of each student in each week is quantified. Finally, according to the digital evaluation score of scientific literacy of each student in all weeks, the digital tracking evaluation result of scientific literacy of each student is obtained, and the evaluation strategy is adjusted in time according to the students' learning progress in real time, which more accurately reflects the students' actual scientific literacy level and avoids over- or under-evaluation of some students.

[0006] The present invention provides a digital tracking and evaluation method for students' scientific literacy based on reinforcement learning, comprising:

[0007] S1: Based on all the class verification indicator data of each student in each week, obtain the processed values ​​of all the class verification indicator data of each student in each week;

[0008] S2: Based on the processed values ​​of all class verification indicator data of each student in each week and the preset reinforcement learning algorithm, the evaluation priority of all class verification indicator data of each student in each week is obtained;

[0009] S3: Based on the evaluation priority of all the class verification indicator data of each student in all the reference weeks of each week, obtain the verification indicator matrix of each student in each week, and based on the verification indicator matrix of each student in each week, obtain all the analysis weeks of each student in each week, and based on the processed values ​​of all the class verification indicator data of each student in each week and the evaluation priority of all the class verification indicator data, obtain the verification indicator graph of each student in each week;

[0010] S4: Based on each student’s weekly verification indicator graph and the verification indicator graphs of all analysis weeks, obtain each student’s weekly digital scientific literacy evaluation score, and based on each student’s digital scientific literacy evaluation score in all weeks, obtain each student’s digital scientific literacy tracking evaluation result.

[0011] Preferably, the digital tracking and evaluation method for student scientific literacy based on reinforcement learning, S1: based on all the class verification indicator data of each student in each week, obtaining the processed value of all the class verification indicator data of each student in each week, including:

[0012] Obtain all the class verification indicator data of each student in each week, wherein all the class verification indicator data include scientific concept indicator data, scientific thinking indicator data, inquiry practice indicator data, and responsibility attitude indicator data, and the scientific concept indicator data include all science test scores, the scientific thinking indicator data include all experimental design program scores, the inquiry practice indicator data include all experimental report scores, and the responsibility attitude indicator data include the homework completion record sheet;

[0013] Based on each type of verification indicator data of each student in each week, the processed value of each type of verification indicator data of each student in each week is obtained.

[0014] Preferably, the digital tracking and evaluation method for student scientific literacy based on reinforcement learning obtains the processed values ​​of each type of verification indicator data of each student in each week based on each type of verification indicator data of each student in each week, including:

[0015] The quotient between the maximum value of all science test scores of each student in each week and the full score of the science test score is used as the processing value of the science concept indicator data of the corresponding student in the corresponding week;

[0016] The quotient between the maximum value of all experimental design scheme scores of each student in each week and the full score of the experimental design scheme score is used as the processing value of the scientific thinking index data of the corresponding student in the corresponding week;

[0017] The quotient between the maximum value of all the experimental report scores of each student in each week and the full score of the experimental report score is used as the processing value of the corresponding student's inquiry practice indicator data in the corresponding week;

[0018] Each student’s weekly homework completion record sheet and the preset responsibility attitude scoring model are used to obtain each student’s weekly responsibility attitude score. The quotient between each student’s weekly responsibility attitude score and the full score of responsibility attitude is used as the processed value of the corresponding student’s responsibility attitude index data in the corresponding week.

[0019] Preferably, the digital tracking and evaluation method for student scientific literacy based on reinforcement learning, S2: based on the processed values ​​of all class verification indicator data of each student in each week and the preset reinforcement learning algorithm, obtaining the evaluation priority of all class verification indicator data of each student in each week, including:

[0020] The processed values ​​of all class verification indicator data of each student in all weeks before each week are combined with the preset reinforcement learning algorithm to obtain the evaluation model of each student in each week;

[0021] Based on the processed values ​​of all class verification indicator data of each student in each week and the evaluation model of the corresponding student in the corresponding week, the evaluation priority of all class verification indicator data of each student in each week is obtained.

[0022] Preferably, the digital tracking and evaluation method for student scientific literacy based on reinforcement learning obtains the verification indicator matrix for each student in each week based on the evaluation priority of all class verification indicator data of each student in all reference weeks of each week, including:

[0023] The four weeks before each student's week are used as all reference weeks for each student in each week;

[0024] Define the ordinal numbers of each student in all reference weeks of each week in chronological order from the beginning to the end, starting from 1, to obtain the ordinal definition results;

[0025] Based on the evaluation priority and ordinal definition results of all the verification indicator data of each student in all reference weeks of each week, the verification indicator matrix of each student in each week is obtained, which is:

[0026]

[0027] Among them, F is the verification indicator matrix of the current computing student in the current computing week, A 1 is the evaluation priority of the scientific concept indicator data of the current calculation student in the reference week with the ordinal number 1 of the current calculation week, B 1 is the evaluation priority of the scientific thinking index data of the current calculation student in the reference week with the ordinal number 1 in the current calculation week, C 1 D is the evaluation priority of the inquiry practice indicator data of the current calculation student in the reference week with the ordinal number 1 of the current calculation week, 1 is the evaluation priority of the responsibility attitude index data of the current student in the reference week with the ordinal number 1 in the current calculation week, A 2 is the evaluation priority of the scientific concept indicator data of the current calculation student in the reference week with the ordinal number 2 of the current calculation week, B 2 is the evaluation priority of the scientific thinking index data of the current calculation student in the reference week with the ordinal number 2 of the current calculation week, C 2D is the evaluation priority of the inquiry practice indicator data of the current calculation student in the reference week with the ordinal number 2 of the current calculation week. 2 is the evaluation priority of the responsibility attitude index data of the current student in the reference week with the ordinal number 2 in the current calculation week, A 3 is the evaluation priority of the scientific concept indicator data of the current calculation student in the reference week with the ordinal number 3 of the current calculation week, B 3 is the evaluation priority of the scientific thinking index data of the current calculation student in the reference week with the ordinal number 3 of the current calculation week, C 3 D is the evaluation priority of the inquiry practice indicator data of the current calculation student in the reference week with the ordinal number 3 of the current calculation week. 3 is the evaluation priority of the responsibility attitude index data of the current calculation student in the reference week with the ordinal number 3 in the current calculation week, A 4 is the evaluation priority of the scientific concept indicator data of the current calculation student in the reference week with the ordinal number of 4 in the current calculation week, B 4 is the evaluation priority of the scientific thinking index data of the current calculation student in the reference week with the ordinal number of 4 in the current calculation week, C 4 D is the evaluation priority of the inquiry practice indicator data of the current calculation student in the reference week with the ordinal number 4 of the current calculation week. 4 The evaluation priority of the responsibility attitude index data of the current calculation student in the reference week with the ordinal number 4 of the current calculation week.

[0028] Preferably, the digital tracking and evaluation method for students' scientific literacy based on reinforcement learning obtains all analysis weeks for each student in each week based on the verification indicator matrix of each student in each week, including:

[0029] Get all eigenvalues ​​of the validation indicator matrix for each student in each week;

[0030] If each eigenvalue of the verification indicator matrix of each student in each week is a real number, the corresponding eigenvalue is regarded as the preliminary eigennumber of the verification indicator matrix of the corresponding student in the corresponding week; if each eigenvalue of the verification indicator matrix of each student in each week is a complex number, the real part of the corresponding eigenvalue is regarded as the preliminary eigennumber of the verification indicator matrix of the corresponding student in the corresponding week;

[0031] The maximum value of all the preparatory characteristic numbers of each student's validation indicator matrix in each week is regarded as the standard characteristic number of the validation indicator matrix of the corresponding student in the corresponding week;

[0032] Based on the evaluation priority of all class verification indicator data of each student in each week, all column matrices of the verification indicator matrix of each student in each week are replaced in turn to obtain all replacement matrices of each student in each week;

[0033] If each eigenvalue of each replacement matrix of each student in each week is a real number, the corresponding eigenvalue is regarded as the analysis characteristic number of the corresponding replacement matrix of the corresponding student in the corresponding week; if each eigenvalue of each replacement matrix of each student in each week is a complex number, the real part of the corresponding eigenvalue is regarded as the analysis characteristic number of the corresponding replacement matrix of the corresponding student in the corresponding week, and the maximum value of all the analysis characteristic numbers of each replacement matrix of each student in each week is regarded as the comparison characteristic number of the corresponding replacement matrix of the corresponding student in the corresponding week;

[0034] The difference between the standard characteristic number of the validation indicator matrix of each student in each week and the comparison characteristic number of each replacement matrix of the corresponding student in the corresponding week is regarded as the analysis difference of the corresponding replacement matrix of the corresponding student in the corresponding week;

[0035] The replacement matrix corresponding to the smallest analysis difference among all replacement matrices of each student in each week is regarded as the elimination matrix of the corresponding student in the corresponding week, and all replacement matrices except the elimination matrix among all replacement matrices of each student in each week are regarded as the retention matrix of the corresponding student in the corresponding week;

[0036] The reference week corresponding to the replaced column matrix of each retained matrix of each student in each week is regarded as the analysis week of each student in each week, and all the analysis weeks of each student in each week are obtained.

[0037] Preferably, the digital tracking and evaluation method for student scientific literacy based on reinforcement learning obtains the verification indicator graph of each student in each week based on the processed values ​​of all class verification indicator data of each student in each week and the evaluation priority of all class verification indicator data, including:

[0038] The processed value of each student's verification indicator data of each category in each week is used as the horizontal axis value, and the evaluation priority of the corresponding student's verification indicator data of the corresponding category in the corresponding week is used as the vertical axis value to obtain the positioning point of each student's verification indicator data of each category in each week;

[0039] Based on the positioning points of all class verification indicator data of each student in each week and the preset graph drawing model, the verification indicator graph of each student in each week is obtained.

[0040] Preferably, the digital tracking and evaluation method for student scientific literacy based on reinforcement learning obtains the digital evaluation score of each student's scientific literacy in each week based on the verification indicator graph of each student in each week and the verification indicator graph of all analysis weeks, including:

[0041] Obtain the verification indicator graphs and digital scientific literacy assessment scores for each student in all analysis weeks of each week;

[0042] Based on the verification indicator graphs and digital evaluation scores of scientific literacy of each student in all analysis weeks of each week, the digital evaluation scores of scientific literacy of each student in each week are obtained, which are:

[0043]

[0044] Among them, τ is the digital evaluation score of the current computing student’s scientific literacy in the current computing week, S 1 is the numerical value of the area of ​​the verification indicator graph of the current student in the first analysis week of the current calculation week, δ 1 Assign a score to the current computing student's digital evaluation of scientific literacy in the first analysis week of the current computing week, S 2 is the numerical value of the area of ​​the verification indicator graph of the current student in the second analysis week of the current calculation week, δ 2 Assign a score to the current computing student's digital evaluation of scientific literacy in the second analysis week of the current computing week, S 3 is the numerical value of the area of ​​the verification indicator graph of the current student in the third analysis week of the current calculation week, δ 3 Assign a score to the current computing student's digital evaluation of scientific literacy in the third analysis week of the current computing week, S 0 is the numerical value of the area of ​​the verification indicator graph of the current calculation student in the current calculation week, σ is the standard deviation of the numerical values ​​of the area of ​​the verification indicator graph of the current calculation student in all analysis weeks of the current calculation week, ln is the natural logarithm, and the value of the natural constant e is 2.718.

[0045] Preferably, the digital tracking and evaluation method for students' scientific literacy based on reinforcement learning obtains the digital tracking and evaluation results of each student's scientific literacy based on the digital evaluation scores of each student in all weeks, including:

[0046] A visualization result is generated based on the digital evaluation scores of each student's scientific literacy in all weeks before the current moment, and the generated visualization result is used as the digital tracking evaluation result of the scientific literacy of the corresponding student.

[0047] The present invention provides a digital tracking and evaluation system for student scientific literacy based on reinforcement learning, which is used to execute any one of the digital tracking and evaluation methods for student scientific literacy based on reinforcement learning in embodiments 1 to 9, including:

[0048] A processing module, used for obtaining a processed value of all class verification indicator data of each student in each week based on all class verification indicator data of each student in each week;

[0049] An evaluation module, used to obtain the evaluation priority of all class verification indicator data of each student in each week based on the processed values ​​of all class verification indicator data of each student in each week and a preset reinforcement learning algorithm;

[0050] A graphics drawing module, for obtaining a verification indicator matrix for each student in each week based on the evaluation priority of all class verification indicator data of each student in all reference weeks of each week, and obtaining all analysis weeks of each student in each week based on the verification indicator matrix of each student in each week, and obtaining a verification indicator graph for each student in each week based on the processed values ​​of all class verification indicator data of each student in each week and the evaluation priority of all class verification indicator data;

[0051] The evaluation and scoring module is used to obtain each student's weekly digital scientific literacy evaluation score based on each student's weekly verification indicator graph and the verification indicator graphs of all analysis weeks, and to obtain each student's scientific literacy digital tracking evaluation results based on each student's scientific literacy digital evaluation score in all weeks.

[0052] The beneficial effects of the present invention compared with the prior art are as follows: based on all the verification indicator data of each student in each week, the processed values ​​of all the verification indicator data of each student in each week are obtained, so that the score performance of each student in each verification indicator in each week can be quantified, and then based on the processed values ​​of all the verification indicator data of each student in each week, the evaluation priority of all the verification indicator data of each student in each week can be obtained, and the relative importance or priority of each verification indicator of each student in each week to the overall evaluation of students' scientific literacy can be accurately quantified. According to the evaluation priority of all the verification indicator data of each student in all the reference weeks of each week, the verification indicator matrix of each student in each week is obtained, which is convenient for the subsequent screening of analysis weeks, and according to the verification indicator matrix of each student in each week, the scores of each student in all the analysis weeks of each week are obtained. According to the processed values ​​of all kinds of verification indicator data of each student in each week and the evaluation priority of all kinds of verification indicator data, the verification indicator graph of each student in each week is obtained, and the graphical representation of the overall evaluation of each student's scientific literacy in each week is accurately drawn. Then, according to the verification indicator graph of each student in each week and the verification indicator graph of all analysis weeks, the digital evaluation score of each student's scientific literacy in each week is obtained, and the overall evaluation of each student's scientific literacy in each week is quantified. Finally, according to the digital evaluation score of each student's scientific literacy in all weeks, the digital tracking evaluation result of each student's scientific literacy is obtained. The evaluation strategy is adjusted in time according to the students' learning progress in real time, which more accurately reflects the students' actual scientific literacy level and avoids over- or under-evaluation of some students.

[0053] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or be understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structure specifically pointed out in the written application document.

[0054] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0056] Figure 1 It is a flow chart of a digital tracking and evaluation method for students' scientific literacy based on reinforcement learning in an embodiment of the present invention;

[0057] Figure 2 Schematic diagram of a digital tracking and evaluation system for students' scientific literacy based on reinforcement learning in an embodiment of the present invention. DETAILED DESCRIPTION

[0058] The preferred embodiments of the present invention are described below in conjunction with the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0059] Embodiment 1:

[0060] The present invention provides a digital tracking and evaluation method for students' scientific literacy based on reinforcement learning. Figure 1 ,include:

[0061] S1: Based on all the class verification indicator data of each student in each week, obtain the processed values ​​of all the class verification indicator data of each student in each week;

[0062] S2: Based on the processed values ​​of all class verification indicator data of each student in each week and the preset reinforcement learning algorithm, the evaluation priority of all class verification indicator data of each student in each week is obtained;

[0063] S3: Based on the evaluation priority of all the class verification indicator data of each student in all the reference weeks of each week, obtain the verification indicator matrix of each student in each week, and based on the verification indicator matrix of each student in each week, obtain all the analysis weeks of each student in each week, and based on the processed values ​​of all the class verification indicator data of each student in each week and the evaluation priority of all the class verification indicator data, obtain the verification indicator graph of each student in each week;

[0064] S4: Based on each student’s weekly verification indicator graph and the verification indicator graphs of all analysis weeks, obtain each student’s weekly digital scientific literacy evaluation score, and based on each student’s digital scientific literacy evaluation score in all weeks, obtain each student’s digital scientific literacy tracking evaluation result.

[0065] In this embodiment, each week is a study week (Monday to Sunday) during the period when students study at school, and the rest week during the holidays is not included in the analysis (for example, if there is a rest week between two study weeks, the week before the second study week is the study week, and the intervening rest week can be ignored), and tracking students' scientific literacy through the digital tracking and evaluation system for students' scientific literacy based on reinforcement learning requires ensuring that students have completed at least 10 weeks of study, that is, at least 10 study weeks have been completed.

[0066] In this embodiment, all types of verification indicator data are all types of data reflecting students' scientific literacy, including scientific concept indicator data, scientific thinking indicator data, inquiry practice indicator data, and responsibility attitude indicator data.

[0067] In this embodiment, the processed value of the verification indicator data is a value obtained based on each type of verification indicator data of each student each week, which can characterize the score performance of each student on each type of verification indicator each week.

[0068] In this embodiment, the preset reinforcement learning algorithm is a reinforcement learning algorithm that is preset to construct an evaluation model for each student every week, such as a Q-learning algorithm.

[0069] In this embodiment, the evaluation priority of the verification indicator data is obtained based on the processed values ​​of all types of verification indicator data of each student each week and a preset reinforcement learning algorithm, which can reflect the relative importance or priority of each type of verification indicator of each student each week to the overall evaluation of the student's scientific literacy.

[0070] In this embodiment, the verification indicator matrix of each student in each week is constructed by using the evaluation priority of all types of verification indicator data of each student in all reference weeks of each week as matrix elements, and can analyze all analysis weeks of each student in each week.

[0071] In this embodiment, all analysis weeks for each student in each week are partial weeks that can be used as references and are screened out from all reference weeks for each student in each week based on the verification indicator matrix for each student in each week.

[0072] In this embodiment, the verification indicator graph of each student each week is a graphical representation of the overall evaluation of each student's scientific literacy each week, obtained based on the processed values ​​of all categories of verification indicator data of each student each week and the evaluation priority of all categories of verification indicator data.

[0073] In this embodiment, each student's weekly digital evaluation score for scientific literacy is a numerical value obtained based on each student's weekly verification indicator graph and the verification indicator graphs of all analysis weeks, which can characterize each student's overall weekly scientific literacy evaluation. The student's digital evaluation score for scientific literacy in the first few weeks is manually marked and assigned, and is subsequently assigned by a digital tracking and evaluation system for student scientific literacy based on reinforcement learning.

[0074] In this embodiment, the digital tracking evaluation result of each student's scientific literacy is based on the digital evaluation score of each student's scientific literacy in all weeks (all weeks before the current moment), and a visualization result (such as a continuous line graph, bar graph, etc.) containing the digital evaluation score of each student's scientific literacy in all weeks before the current moment is obtained.

[0075] The beneficial effects of the above technology are: according to all the verification indicator data of each student in each week, the processed value of all the verification indicator data of each student in each week is obtained, and the score performance of each student in each verification indicator in each week is quantified, and then according to the processed value of all the verification indicator data of each student in each week, the evaluation priority of all the verification indicator data of each student in each week is obtained, and the relative importance or priority of each student's each week to the overall evaluation of students' scientific literacy is accurately quantified. According to the evaluation priority of all the verification indicator data of each student in all the reference weeks of each week, the verification indicator matrix of each student in each week is obtained, which is convenient for the subsequent screening of analysis weeks. According to the verification indicator matrix of each student in each week, all the analysis weeks of each student in each week are obtained. The processed values ​​of all types of verification indicator data of each student in each week and the evaluation priority of all types of verification indicator data are obtained to obtain the verification indicator graph of each student in each week, and accurately draw a graphical representation of the overall evaluation of each student's scientific literacy in each week. Then, based on the verification indicator graph of each student in each week and the verification indicator graph of all analysis weeks, the digital evaluation score of each student's scientific literacy in each week is obtained, and the overall evaluation of each student's scientific literacy in each week is quantified. Finally, based on the digital evaluation score of each student's scientific literacy in all weeks, the digital tracking evaluation result of each student's scientific literacy is obtained, and the evaluation strategy is adjusted in real time according to the students' learning progress performance, which more accurately reflects the students' actual scientific literacy level and avoids over- or under-evaluation of some students.

[0076] Embodiment 2:

[0077] Based on Example 1, a digital tracking and evaluation method for student scientific literacy based on reinforcement learning, S1: based on all class verification indicator data of each student in each week, obtaining the processed value of all class verification indicator data of each student in each week, including:

[0078] Obtain all the class verification indicator data of each student in each week, wherein all the class verification indicator data include scientific concept indicator data, scientific thinking indicator data, inquiry practice indicator data, and responsibility attitude indicator data, and the scientific concept indicator data include all science test scores, the scientific thinking indicator data include all experimental design program scores, the inquiry practice indicator data include all experimental report scores, and the responsibility attitude indicator data include the homework completion record sheet;

[0079] Based on each type of verification indicator data of each student in each week, the processed value of each type of verification indicator data of each student in each week is obtained.

[0080] In this embodiment, the science test scores are the scores of students in weekly physics tests, chemistry tests, and biology tests.

[0081] In this embodiment, the experimental design scheme score is the score of the experimental design scheme submitted by the student in the weekly experimental class.

[0082] In this embodiment, the lab report grade is the grade of the lab report written by the student in the weekly lab class.

[0083] In this embodiment, the homework completion record table is a table filled with all homework submission records of students in each week, and the table contains records of whether the students have submitted all homework.

[0084] The beneficial effect of the above technology is: it clarifies the specific types of all types of verification indicator data for each student each week, which makes it easier to obtain the processed values ​​of each type of verification indicator data for each student each week based on each type of verification indicator data for each student each week.

[0085] Embodiment 3:

[0086] On the basis of Example 2, a digital tracking and evaluation method for student scientific literacy based on reinforcement learning obtains a processed value of each type of verification indicator data of each student in each week based on each type of verification indicator data of each student in each week, including:

[0087] The quotient between the maximum value of all science test scores of each student in each week and the full score of the science test score is used as the processing value of the science concept indicator data of the corresponding student in the corresponding week;

[0088] The quotient between the maximum value of all experimental design scheme scores of each student in each week and the full score of the experimental design scheme score is used as the processing value of the scientific thinking index data of the corresponding student in the corresponding week;

[0089] The quotient between the maximum value of all the experimental report scores of each student in each week and the full score of the experimental report score is used as the processing value of the corresponding student's inquiry practice indicator data in the corresponding week;

[0090] Each student’s weekly homework completion record sheet and the preset responsibility attitude scoring model are used to obtain each student’s weekly responsibility attitude score. The quotient between each student’s weekly responsibility attitude score and the full score of responsibility attitude is used as the processed value of the corresponding student’s responsibility attitude index data in the corresponding week.

[0091] In this embodiment, the maximum score of the science test is the value of the maximum science test score corresponding to the full score of the test subject among all science test scores of each student in each week, for example, 100.

[0092] In this embodiment, the maximum score of the experimental design scheme is the maximum score of the experimental design scheme, for example, 100.

[0093] In this embodiment, the maximum score of the experiment report is the maximum score of the experiment report, for example, 100.

[0094] In this embodiment, the preset responsibility attitude scoring model is based on a large number of pre-collected students' homework completion record sheets for multiple weeks as training input, and the manually labeled responsibility attitude scoring values ​​of the corresponding students in the corresponding weeks as training output. The trained neural network model can input each student's weekly homework completion record sheet and output the responsibility attitude scoring values ​​of the corresponding students in the corresponding weeks.

[0095] In this embodiment, the weekly responsibility attitude score of each student is a numerical value assigned to each student's weekly responsibility attitude.

[0096] In this embodiment, the full score value of the responsibility attitude is the full score value of the responsibility attitude scoring value, for example, 10.

[0097] The beneficial effects of the above technology are as follows: This embodiment provides in detail a specific method for obtaining the processed numerical value of each type of verification indicator data of each student in each week based on each type of verification indicator data of each student in each week.

[0098] Embodiment 4:

[0099] Based on Example 1, a digital tracking and evaluation method for student scientific literacy based on reinforcement learning, S2: based on the processed values ​​of all class verification indicator data of each student in each week and a preset reinforcement learning algorithm, obtaining the evaluation priority of all class verification indicator data of each student in each week, including:

[0100] The processed values ​​of all class verification indicator data of each student in all weeks before each week are combined with the preset reinforcement learning algorithm to obtain the evaluation model of each student in each week;

[0101] Based on the processed values ​​of all class verification indicator data of each student in each week and the evaluation model of the corresponding student in the corresponding week, the evaluation priority of all class verification indicator data of each student in each week is obtained.

[0102] In this embodiment, the weekly evaluation model for each student is constructed according to a preset reinforcement learning algorithm (the Q-learning algorithm is used in this application document, and the processed values ​​of all types of verification indicator data of students in different weeks are regarded as state space, and the adjustment of the evaluation priority of all types of verification indicator data is regarded as action space, and the accuracy and fairness of the evaluation results are regarded as rewards. The Q value table is initialized and updated based on historical experience, that is, the actual situation of adjusting the evaluation priority of different categories of verification indicator data after observing the student performance in the past, gradually approaching the optimal evaluation strategy) to quantify the relative importance or priority of each type of verification indicator of each student in each week to the overall evaluation of the student's scientific literacy.

[0103] The beneficial effects of the above technology are: based on the processed values ​​of all class verification indicator data of each student each week and the preset reinforcement learning algorithm, the evaluation priority of all class verification indicator data of each student each week is obtained, and the relative importance or priority of each class verification indicator of each student each week to the overall evaluation of students' scientific literacy is accurately quantified.

[0104] Embodiment 5:

[0105] On the basis of Example 1, the digital tracking and evaluation method for student scientific literacy based on reinforcement learning obtains the verification indicator matrix of each student in each week based on the evaluation priority of all class verification indicator data of each student in all reference weeks of each week, including:

[0106] The four weeks before each student's week are used as all reference weeks for each student in each week;

[0107] Define the ordinal numbers of each student in all reference weeks of each week in chronological order from the beginning to the end, starting from 1, to obtain the ordinal definition results;

[0108] Based on the evaluation priority and ordinal definition results of all the verification indicator data of each student in all reference weeks of each week, the verification indicator matrix of each student in each week is obtained, which is:

[0109]

[0110] Among them, F is the verification indicator matrix of the current computing student in the current computing week, A 1 is the evaluation priority of the scientific concept indicator data of the current calculation student in the reference week with the ordinal number 1 of the current calculation week, B 1 is the evaluation priority of the scientific thinking index data of the current calculation student in the reference week with the ordinal number 1 in the current calculation week, C 1 D is the evaluation priority of the inquiry practice indicator data of the current calculation student in the reference week with the ordinal number 1 of the current calculation week, 1 is the evaluation priority of the responsibility attitude index data of the current student in the reference week with the ordinal number 1 in the current calculation week, A 2 is the evaluation priority of the scientific concept indicator data of the current calculation student in the reference week with the ordinal number 2 of the current calculation week, B 2 is the evaluation priority of the scientific thinking index data of the current calculation student in the reference week with the ordinal number 2 of the current calculation week, C 2 D is the evaluation priority of the inquiry practice indicator data of the current calculation student in the reference week with the ordinal number 2 of the current calculation week. 2 is the evaluation priority of the responsibility attitude index data of the current student in the reference week with the ordinal number 2 in the current calculation week, A 3 is the evaluation priority of the scientific concept indicator data of the current calculation student in the reference week with the ordinal number 3 of the current calculation week, B 3 is the evaluation priority of the scientific thinking index data of the current calculation student in the reference week with the ordinal number 3 of the current calculation week, C 3 D is the evaluation priority of the inquiry practice indicator data of the current calculation student in the reference week with the ordinal number 3 of the current calculation week. 3 is the evaluation priority of the responsibility attitude index data of the current calculation student in the reference week with the ordinal number 3 in the current calculation week, A 4 is the evaluation priority of the scientific concept indicator data of the current calculation student in the reference week with the ordinal number of 4 in the current calculation week, B 4 is the evaluation priority of the scientific thinking index data of the current calculation student in the reference week with the ordinal number of 4 in the current calculation week, C 4 D is the evaluation priority of the inquiry practice indicator data of the current calculation student in the reference week with the ordinal number 4 of the current calculation week. 4 The evaluation priority of the responsibility attitude index data of the current calculation student in the reference week with the ordinal number 4 of the current calculation week.

[0111] The beneficial effect of the above technology is: according to the evaluation priority of all types of verification indicator data of each student in all reference weeks of each week, the verification indicator matrix of each student in each week is obtained, which is convenient for subsequent screening and analysis weeks. This embodiment gives in detail a specific method for constructing the verification indicator matrix of each student in each week.

[0112] Embodiment 6:

[0113] On the basis of Example 1, the digital tracking and evaluation method for student scientific literacy based on reinforcement learning obtains all analysis weeks of each student in each week based on the verification indicator matrix of each student in each week, including:

[0114] Get all eigenvalues ​​of the validation indicator matrix for each student in each week;

[0115] If each eigenvalue of the verification indicator matrix of each student in each week is a real number, the corresponding eigenvalue is regarded as the preliminary eigennumber of the verification indicator matrix of the corresponding student in the corresponding week; if each eigenvalue of the verification indicator matrix of each student in each week is a complex number, the real part of the corresponding eigenvalue is regarded as the preliminary eigennumber of the verification indicator matrix of the corresponding student in the corresponding week;

[0116] The maximum value of all the preparatory characteristic numbers of each student's validation indicator matrix in each week is regarded as the standard characteristic number of the validation indicator matrix of the corresponding student in the corresponding week;

[0117] Based on the evaluation priority of all class verification indicator data of each student in each week, all column matrices of the verification indicator matrix of each student in each week are replaced in turn to obtain all replacement matrices of each student in each week;

[0118] If each eigenvalue of each replacement matrix of each student in each week is a real number, the corresponding eigenvalue is regarded as the analysis characteristic number of the corresponding replacement matrix of the corresponding student in the corresponding week; if each eigenvalue of each replacement matrix of each student in each week is a complex number, the real part of the corresponding eigenvalue is regarded as the analysis characteristic number of the corresponding replacement matrix of the corresponding student in the corresponding week, and the maximum value of all the analysis characteristic numbers of each replacement matrix of each student in each week is regarded as the comparison characteristic number of the corresponding replacement matrix of the corresponding student in the corresponding week;

[0119] The difference between the standard characteristic number of the validation indicator matrix of each student in each week and the comparison characteristic number of each replacement matrix of the corresponding student in the corresponding week is regarded as the analysis difference of the corresponding replacement matrix of the corresponding student in the corresponding week;

[0120] The replacement matrix corresponding to the smallest analysis difference among all replacement matrices of each student in each week is regarded as the elimination matrix of the corresponding student in the corresponding week, and all replacement matrices except the elimination matrix among all replacement matrices of each student in each week are regarded as the retention matrix of the corresponding student in the corresponding week;

[0121] The reference week corresponding to the replaced column matrix of each retained matrix of each student in each week is regarded as the analysis week of each student in each week, and all the analysis weeks of each student in each week are obtained.

[0122] In this embodiment, the evaluation priority of each student's verification indicator data of all classes in each week is replaced in sequence, and each column matrix of each student's verification indicator matrix in each week is replaced, and only one column matrix is ​​replaced each time, and the other column matrices remain unchanged. The relative position of each element in the replaced column matrix is ​​the same as the relative position of each element in the replaced column matrix.

[0123] The beneficial effects of the above technology are: based on the verification indicator matrix of each student in each week and the evaluation priority of all class verification indicator data of each student in each week, all replacement matrices of each student in each week are obtained, and then based on all replacement matrices of each student in each week, all analysis weeks of each student in each week are obtained. This embodiment gives in detail a specific method for obtaining all analysis weeks of each student in each week based on the verification indicator matrix of each student in each week.

[0124] Embodiment 7:

[0125] On the basis of Example 1, the digital tracking and evaluation method for student scientific literacy based on reinforcement learning obtains the verification indicator graph of each student in each week based on the processed values ​​of all class verification indicator data of each student in each week and the evaluation priority of all class verification indicator data, including:

[0126] The processed value of each student's verification indicator data of each category in each week is used as the horizontal axis value, and the evaluation priority of the corresponding student's verification indicator data of the corresponding category in the corresponding week is used as the vertical axis value to obtain the positioning point of each student's verification indicator data of each category in each week;

[0127] Based on the positioning points of all class verification indicator data of each student in each week and the preset graph drawing model, the verification indicator graph of each student in each week is obtained.

[0128] In this embodiment, the preset graphics drawing model uses the positioning points of all class verification indicator data of a large number of students in multiple weeks collected in advance as training input, and the manually labeled verification indicator graphics of the corresponding students in the corresponding weeks as training output. The trained neural network model can input the positioning points of all class verification indicator data of each student in each week, and can output the verification indicator graphics of each student in each week.

[0129] In this embodiment, the verification indicator graph of each student each week is a plane graph with the largest area formed by the positioning points of all class verification indicator data of each student each week in a plane rectangular coordinate system using a preset graphic drawing model.

[0130] The beneficial effects of the above technology are: according to the processed values ​​of all class verification indicator data of each student in each week and the evaluation priority of all class verification indicator data, the positioning points of all class verification indicator data of each student in each week are obtained, and then according to the positioning points of all class verification indicator data of each student in each week and the preset graphic drawing model, the verification indicator graph of each student in each week is obtained, and the graphical representation of the overall evaluation of each student's scientific literacy in each week is accurately drawn.

[0131] Embodiment 8:

[0132] On the basis of Example 1, the digital tracking and evaluation method for student scientific literacy based on reinforcement learning obtains the digital evaluation score of each student's scientific literacy in each week based on the verification indicator graph of each student in each week and the verification indicator graph of all analysis weeks, including:

[0133] Obtain the verification indicator graphs and digital scientific literacy assessment scores for each student in all analysis weeks of each week;

[0134] Based on the verification indicator graphs and digital evaluation scores of scientific literacy of each student in all analysis weeks of each week, the digital evaluation scores of scientific literacy of each student in each week are obtained, which are:

[0135]

[0136] Among them, τ is the digital evaluation score of the current computing student’s scientific literacy in the current computing week, S 1 is the numerical value of the area of ​​the verification indicator graph of the current student in the first analysis week of the current calculation week, δ 1 Assign a score to the current computing student's digital evaluation of scientific literacy in the first analysis week of the current computing week, S 2 is the numerical value of the area of ​​the verification indicator graph of the current student in the second analysis week of the current calculation week, δ 2Assign a score to the current computing student's digital evaluation of scientific literacy in the second analysis week of the current computing week, S 3 is the numerical value of the area of ​​the verification indicator graph of the current student in the third analysis week of the current calculation week, δ 3 Assign a score to the current computing student's digital evaluation of scientific literacy in the third analysis week of the current computing week, S 0 is the numerical value of the area of ​​the verification indicator graph of the current calculation student in the current calculation week, σ is the standard deviation of the numerical values ​​of the area of ​​the verification indicator graph of the current calculation student in all analysis weeks of the current calculation week, ln is the natural logarithm, and the value of the natural constant e is 2.718.

[0137] The beneficial effect of the above technology is: according to the verification indicator graph of each student in each week and the verification indicator graph of all analysis weeks, the digital evaluation score of each student's scientific literacy in each week is obtained. This embodiment gives in detail a specific method for quantifying the overall evaluation of each student's scientific literacy in each week.

[0138] Embodiment 9:

[0139] On the basis of Example 1, the digital tracking and evaluation method for students' scientific literacy based on reinforcement learning, based on the digital evaluation score of each student's scientific literacy in all weeks, obtains the digital tracking and evaluation result of each student's scientific literacy, including:

[0140] A visualization result is generated based on the digital evaluation scores of each student's scientific literacy in all weeks before the current moment, and the generated visualization result is used as the digital tracking evaluation result of the scientific literacy of the corresponding student.

[0141] In this embodiment, the visualization result is a line chart or a bar chart containing the digital evaluation scores of scientific literacy of each student in all weeks before the current moment.

[0142] The beneficial effects of the above technology are: according to the digital evaluation scores of each student's scientific literacy in all weeks, the digital tracking evaluation results of each student's scientific literacy are obtained, and the evaluation strategy is adjusted in time according to the student's learning progress in real time, which more accurately reflects the students' actual scientific literacy level and avoids over- or under-evaluation of some students.

[0143] Embodiment 10:

[0144] The present invention provides a digital tracking and evaluation system for student scientific literacy based on reinforcement learning, which is used to implement any one of the digital tracking and evaluation methods for student scientific literacy based on reinforcement learning in embodiments 1 to 9, referring to Figure 2 ,include:

[0145] A processing module, used for obtaining a processed value of all class verification indicator data of each student in each week based on all class verification indicator data of each student in each week;

[0146] An evaluation module, used to obtain the evaluation priority of all class verification indicator data of each student in each week based on the processed values ​​of all class verification indicator data of each student in each week and a preset reinforcement learning algorithm;

[0147] A graphics drawing module, for obtaining a verification indicator matrix for each student in each week based on the evaluation priority of all class verification indicator data of each student in all reference weeks of each week, and obtaining all analysis weeks of each student in each week based on the verification indicator matrix of each student in each week, and obtaining a verification indicator graph for each student in each week based on the processed values ​​of all class verification indicator data of each student in each week and the evaluation priority of all class verification indicator data;

[0148] The evaluation and scoring module is used to obtain each student's weekly digital scientific literacy evaluation score based on each student's weekly verification indicator graph and the verification indicator graphs of all analysis weeks, and to obtain each student's scientific literacy digital tracking evaluation results based on each student's scientific literacy digital evaluation score in all weeks.

[0149] The beneficial effects of the above technology are: according to all the verification indicator data of each student in each week, the processed value of all the verification indicator data of each student in each week is obtained, and the score performance of each student in each verification indicator in each week is quantified, and then according to the processed value of all the verification indicator data of each student in each week, the evaluation priority of all the verification indicator data of each student in each week is obtained, and the relative importance or priority of each student's each week to the overall evaluation of students' scientific literacy is accurately quantified. According to the evaluation priority of all the verification indicator data of each student in all the reference weeks of each week, the verification indicator matrix of each student in each week is obtained, which is convenient for the subsequent screening of analysis weeks. According to the verification indicator matrix of each student in each week, all the analysis weeks of each student in each week are obtained. The processed values ​​of all types of verification indicator data of each student in each week and the evaluation priority of all types of verification indicator data are obtained to obtain the verification indicator graph of each student in each week, and accurately draw a graphical representation of the overall evaluation of each student's scientific literacy in each week. Then, based on the verification indicator graph of each student in each week and the verification indicator graph of all analysis weeks, the digital evaluation score of each student's scientific literacy in each week is obtained, and the overall evaluation of each student's scientific literacy in each week is quantified. Finally, based on the digital evaluation score of each student's scientific literacy in all weeks, the digital tracking evaluation result of each student's scientific literacy is obtained, and the evaluation strategy is adjusted in real time according to the students' learning progress performance, which more accurately reflects the students' actual scientific literacy level and avoids over- or under-evaluation of some students.

[0150] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention, and the present invention is also intended to include these changes and modifications.

Claims

1. A digital tracking and evaluation method for students' scientific literacy based on reinforcement learning, characterized by: include: S1: Based on all the class verification indicator data of each student in each week, obtain the processed values ​​of all the class verification indicator data of each student in each week; S2: Based on the processed values ​​of all class verification indicator data of each student in each week and the preset reinforcement learning algorithm, the evaluation priority of all class verification indicator data of each student in each week is obtained; S3: Based on the evaluation priority of all the class verification indicator data of each student in all the reference weeks of each week, obtain the verification indicator matrix of each student in each week, and based on the verification indicator matrix of each student in each week, obtain all the analysis weeks of each student in each week, and based on the processed values ​​of all the class verification indicator data of each student in each week and the evaluation priority of all the class verification indicator data, obtain the verification indicator graph of each student in each week; S4: Based on each student’s weekly verification indicator graph and the verification indicator graphs of all analysis weeks, obtain each student’s weekly digital scientific literacy evaluation score, and based on each student’s digital scientific literacy evaluation score in all weeks, obtain each student’s digital scientific literacy tracking evaluation result.

2. The digital tracking and evaluation method for students' scientific literacy based on reinforcement learning according to claim 1 is characterized in that: S1: Based on all the class verification indicator data of each student in each week, obtain the processed values ​​of all the class verification indicator data of each student in each week, including: Obtain all the class verification indicator data of each student in each week, wherein all the class verification indicator data include scientific concept indicator data, scientific thinking indicator data, inquiry practice indicator data, and responsibility attitude indicator data, and the scientific concept indicator data include all science test scores, the scientific thinking indicator data include all experimental design program scores, the inquiry practice indicator data include all experimental report scores, and the responsibility attitude indicator data include the homework completion record sheet; Based on each type of verification indicator data of each student in each week, the processed value of each type of verification indicator data of each student in each week is obtained.

3. The digital tracking and evaluation method for students' scientific literacy based on reinforcement learning according to claim 2 is characterized in that: Based on each student's weekly verification indicator data of each category, the processed values ​​of each student's weekly verification indicator data of each category are obtained, including: The quotient between the maximum value of all science test scores of each student in each week and the full score of the science test score is used as the processing value of the science concept indicator data of the corresponding student in the corresponding week; The quotient between the maximum value of all experimental design scheme scores of each student in each week and the full score of the experimental design scheme score is used as the processing value of the scientific thinking index data of the corresponding student in the corresponding week; The quotient between the maximum value of all the experimental report scores of each student in each week and the full score of the experimental report score is used as the processing value of the corresponding student's inquiry practice indicator data in the corresponding week; Each student’s weekly homework completion record sheet and the preset responsibility attitude scoring model are used to obtain each student’s weekly responsibility attitude score. The quotient between each student’s weekly responsibility attitude score and the full score of responsibility attitude is used as the processed value of the corresponding student’s responsibility attitude index data in the corresponding week.

4. The digital tracking and evaluation method for students' scientific literacy based on reinforcement learning according to claim 1 is characterized in that: S2: Based on the processed values ​​of all class verification indicator data of each student in each week and the preset reinforcement learning algorithm, the evaluation priority of all class verification indicator data of each student in each week is obtained, including: The processed values ​​of all class verification indicator data of each student in all weeks before each week are combined with the preset reinforcement learning algorithm to obtain the evaluation model of each student in each week; Based on the processed values ​​of all class verification indicator data of each student in each week and the evaluation model of the corresponding student in the corresponding week, the evaluation priority of all class verification indicator data of each student in each week is obtained.

5. The digital tracking and evaluation method for students' scientific literacy based on reinforcement learning according to claim 1 is characterized in that: Based on the evaluation priority of all the verification indicator data of each student in all the reference weeks in each week, the verification indicator matrix of each student in each week is obtained, including: The four weeks before each student's week are used as all reference weeks for each student in each week; Define the ordinal numbers of each student in all reference weeks of each week in chronological order from the beginning to the end, starting from 1, to obtain the ordinal definition results; Based on the evaluation priority and ordinal definition results of all the verification indicator data of each student in all reference weeks of each week, the verification indicator matrix of each student in each week is obtained, which is: Among them, F is the verification indicator matrix of the current calculation student in the current calculation week, A1 is the evaluation priority of the scientific concept indicator data of the current calculation student in the reference week with the ordinal number 1 of the current calculation week, B1 is the evaluation priority of the scientific thinking indicator data of the current calculation student in the reference week with the ordinal number 1 of the current calculation week, C1 is the evaluation priority of the inquiry practice indicator data of the current calculation student in the reference week with the ordinal number 1 of the current calculation week, D1 is the evaluation priority of the responsibility attitude indicator data of the current calculation student in the reference week with the ordinal number 1 of the current calculation week, A2 is the evaluation priority of the scientific concept indicator data of the current calculation student in the reference week with the ordinal number 2 of the current calculation week, B2 is the evaluation priority of the scientific thinking indicator data of the current calculation student in the reference week with the ordinal number 2 of the current calculation week, C2 is the evaluation priority of the inquiry practice indicator data of the current calculation student in the reference week with the ordinal number 2 of the current calculation week, and D2 is the responsibility attitude indicator data of the current calculation student in the reference week with the ordinal number 2 of the current calculation week. The evaluation priority of attitude indicator data, A3 is the evaluation priority of the scientific concept indicator data of the current student in the reference week of 3 for the current calculation week, B3 is the evaluation priority of the scientific thinking indicator data of the current student in the reference week of 3 for the current calculation week, C3 is the evaluation priority of the inquiry practice indicator data of the current student in the reference week of 3 for the current calculation week, D3 is the evaluation priority of the responsibility attitude indicator data of the current student in the reference week of 3 for the current calculation week, A4 is the evaluation priority of the scientific concept indicator data of the current student in the reference week of 4 for the current calculation week, B4 is the evaluation priority of the scientific thinking indicator data of the current student in the reference week of 4 for the current calculation week, C4 is the evaluation priority of the inquiry practice indicator data of the current student in the reference week of 4 for the current calculation week, and D4 is the evaluation priority of the responsibility attitude indicator data of the current student in the reference week of 4 for the current calculation week.

6. The digital tracking and evaluation method for students' scientific literacy based on reinforcement learning according to claim 1 is characterized in that: Based on the validation indicator matrix for each student in each week, all analysis weeks for each student in each week are obtained, including: Get all eigenvalues ​​of the validation indicator matrix for each student in each week; If each eigenvalue of the verification indicator matrix of each student in each week is a real number, the corresponding eigenvalue is regarded as the preliminary eigennumber of the verification indicator matrix of the corresponding student in the corresponding week; if each eigenvalue of the verification indicator matrix of each student in each week is a complex number, the real part of the corresponding eigenvalue is regarded as the preliminary eigennumber of the verification indicator matrix of the corresponding student in the corresponding week; The maximum value of all the preparatory characteristic numbers of each student's validation indicator matrix in each week is regarded as the standard characteristic number of the validation indicator matrix of the corresponding student in the corresponding week; Based on the evaluation priority of all class verification indicator data of each student in each week, all column matrices of the verification indicator matrix of each student in each week are replaced in turn to obtain all replacement matrices of each student in each week; If each eigenvalue of each replacement matrix of each student in each week is a real number, the corresponding eigenvalue is regarded as the analysis characteristic number of the corresponding replacement matrix of the corresponding student in the corresponding week; if each eigenvalue of each replacement matrix of each student in each week is a complex number, the real part of the corresponding eigenvalue is regarded as the analysis characteristic number of the corresponding replacement matrix of the corresponding student in the corresponding week, and the maximum value of all the analysis characteristic numbers of each replacement matrix of each student in each week is regarded as the comparison characteristic number of the corresponding replacement matrix of the corresponding student in the corresponding week; The difference between the standard characteristic number of the validation indicator matrix of each student in each week and the comparison characteristic number of each replacement matrix of the corresponding student in the corresponding week is regarded as the analysis difference of the corresponding replacement matrix of the corresponding student in the corresponding week; The replacement matrix corresponding to the smallest analysis difference among all replacement matrices of each student in each week is regarded as the elimination matrix of the corresponding student in the corresponding week, and all replacement matrices except the elimination matrix among all replacement matrices of each student in each week are regarded as the retention matrix of the corresponding student in the corresponding week; The reference week corresponding to the replaced column matrix of each retained matrix of each student in each week is regarded as the analysis week of each student in each week, and all the analysis weeks of each student in each week are obtained.

7. The digital tracking and evaluation method for students' scientific literacy based on reinforcement learning according to claim 1 is characterized in that: Based on the processing values ​​of all the verification indicator data of each student in each week and the evaluation priority of all the verification indicator data of each class, the verification indicator graph of each student in each week is obtained, including: The processed value of each student's verification indicator data of each category in each week is used as the horizontal axis value, and the evaluation priority of the corresponding student's verification indicator data of the corresponding category in the corresponding week is used as the vertical axis value to obtain the positioning point of each student's verification indicator data of each category in each week; Based on the positioning points of all class verification indicator data of each student in each week and the preset graph drawing model, the verification indicator graph of each student in each week is obtained.

8. The digital tracking and evaluation method for students' scientific literacy based on reinforcement learning according to claim 1 is characterized in that: Based on the verification indicator graphs of each student in each week and the verification indicator graphs of all analysis weeks, the digital evaluation score of scientific literacy of each student in each week is obtained, including: Obtain the verification indicator graphs and digital scientific literacy assessment scores for each student in all analysis weeks of each week; Based on the verification indicator graphs and digital evaluation scores of scientific literacy of each student in all analysis weeks of each week, the digital evaluation scores of scientific literacy of each student in each week are obtained, which are: Among them, τ is the digital evaluation score of the scientific literacy of the current calculation student in the current calculation week, S1 is the numerical value of the area of ​​the verification index graph of the current calculation student in the first analysis week of the current calculation week, δ1 is the digital evaluation score of the scientific literacy of the current calculation student in the first analysis week of the current calculation week, S2 is the numerical value of the area of ​​the verification index graph of the current calculation student in the second analysis week of the current calculation week, δ2 is the digital evaluation score of the scientific literacy of the current calculation student in the second analysis week of the current calculation week, S3 is the numerical value of the area of ​​the verification index graph of the current calculation student in the third analysis week of the current calculation week, δ3 is the digital evaluation score of the scientific literacy of the current calculation student in the third analysis week of the current calculation week, S0 is the numerical value of the area of ​​the verification index graph of the current calculation student in the current calculation week, σ is the standard deviation of the numerical values ​​of the area of ​​the verification index graph of the current calculation student in all analysis weeks of the current calculation week, ln is the natural logarithm, and the value of the natural constant e is 2.

718.

9. The digital tracking and evaluation method for students' scientific literacy based on reinforcement learning according to claim 1 is characterized in that: Based on the digital evaluation scores of each student's scientific literacy in all weeks, the digital tracking evaluation results of each student's scientific literacy are obtained, including: A visualization result is generated based on the digital evaluation scores of each student's scientific literacy in all weeks before the current moment, and the generated visualization result is used as the digital tracking evaluation result of the scientific literacy of the corresponding student.

10. A digital tracking and evaluation system for students' scientific literacy based on reinforcement learning, characterized by: A method for digital tracking and evaluation of students' scientific literacy based on reinforcement learning for executing any one of claims 1 to 9, comprising: A processing module, used for obtaining a processed value of all class verification indicator data of each student in each week based on all class verification indicator data of each student in each week; An evaluation module, used to obtain the evaluation priority of all class verification indicator data of each student in each week based on the processed values ​​of all class verification indicator data of each student in each week and a preset reinforcement learning algorithm; A graphics drawing module, for obtaining a verification indicator matrix for each student in each week based on the evaluation priority of all class verification indicator data of each student in all reference weeks of each week, and obtaining all analysis weeks of each student in each week based on the verification indicator matrix of each student in each week, and obtaining a verification indicator graph for each student in each week based on the processed values ​​of all class verification indicator data of each student in each week and the evaluation priority of all class verification indicator data; The evaluation and scoring module is used to obtain each student's weekly digital scientific literacy evaluation score based on each student's weekly verification indicator graph and the verification indicator graphs of all analysis weeks, and to obtain each student's scientific literacy digital tracking evaluation results based on each student's scientific literacy digital evaluation score in all weeks.

Images