Student practical literacy evaluation method based on digital technology
By constructing a profile of instructors' capabilities and fusing multi-source data, the problems of vague identification of instructors' capabilities and inaccurate task matching in existing technologies have been solved, enabling accurate evaluation and dynamic optimization of students' practical skills and improving the training effect.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- CHENGDU IND VOCATIONAL TECHN COLLEGE
- Filing Date
- 2026-02-24
- Publication Date
- 2026-05-29
AI Technical Summary
Existing technologies for evaluating students' practical skills suffer from problems such as unclear identification of instructors' abilities, inaccurate task matching, insufficient data integration, and a disconnect between evaluation and optimization, resulting in poor training outcomes.
By collecting historical data of instructors to build competency profiles, combining student growth data to generate personalized training programs, and generating high-value feature datasets through multi-source data fusion, a two-way iterative evaluation is conducted to optimize instructor capabilities and adjust the practice system.
This approach enables precise identification and personalized training of instructors' abilities, enhances the matching of practical tasks and the comprehensiveness of evaluation, forms a closed loop of training and optimization, and improves the quality of students' practical literacy development.
Smart Images
Figure CN122115161A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of student practical literacy evaluation technology, specifically to a student practical literacy evaluation method based on digital intelligence technology. Background Technology
[0002] Cultivating students' practical literacy has become a core aspect of improving the quality of higher education. Currently, universities generally carry out various practical activities such as social practice, professional practice, and innovative practice, forming a practical education framework that integrates research, training, and mentorship, and is guided by the six virtues of students. They are attempting to promote students' all-round development through diversified forms of practice. Therefore, a student practical literacy evaluation method based on digital technology is needed.
[0003] Existing technologies mostly adopt a uniform and standardized training model, failing to incorporate personalized analysis of instructors' historical practical data. This makes it impossible to accurately identify potential weaknesses in their guidance capabilities and the severity of those weaknesses. Consequently, training programs become disconnected from the actual needs of instructors, and some instructors' core competency gaps remain unaddressed for extended periods, impacting the effectiveness of practical guidance.
[0004] Current practical task allocation relies heavily on human experience and fails to fully integrate student growth data with course requirement data for intelligent matching. Existing methods struggle to generate personalized practical tasks based on individual student differences and course knowledge requirements, often resulting in mismatches between task difficulty and student abilities, and a disconnect between task content and course content, thus reducing student participation and the effectiveness of practical training.
[0005] Current evaluations often rely on single-dimensional practical data, failing to effectively integrate multi-source, cross-scenario data such as student practice execution data, instructor guidance data, and course adaptation data. Data preprocessing and feature extraction methods are simplistic, making it difficult to uncover the underlying correlations and values within the data. This results in evaluations lacking comprehensiveness and depth, failing to accurately reflect students' practical competence levels.
[0006] Existing technologies often focus on a single assessment of students' practical skills, failing to establish a closed-loop system from evaluation and feedback to optimization. On the one hand, precise optimization plans for instructors' capabilities are not developed based on evaluation results, leading to a disconnect between instructor skill improvement and student performance evaluation results. On the other hand, the practical system is not dynamically adjusted based on evaluation data, resulting in long-term rigidity in practical project settings and resource allocation, making it difficult to adapt to the dynamic needs of student growth and curriculum reform. Summary of the Invention
[0007] To address the aforementioned technical shortcomings, the present invention aims to provide a method for evaluating students' practical literacy based on digital intelligence technology.
[0008] To solve the above-mentioned technical problems, the present invention adopts the following technical solution: The present invention provides a student practical literacy evaluation method based on digital intelligence technology, including the following steps: Step 1, empowering instructors: collect the historical practice data of instructors, analyze the historical practice data of instructors through digital intelligence technology, construct an instructor competency profile, and then generate a personalized training plan.
[0009] Step 2, Dynamic Practice Execution: Collect student growth data and course requirement data, analyze the student growth data and course requirement data, and generate intelligently adapted practice tasks.
[0010] Step 3: Multi-source data fusion: Based on the instructor's competency profile, personalized training plan, and intelligent adaptation practice tasks, collect cross-scenario practice data. After preprocessing and feature extraction, generate a high-value feature dataset through cross-domain fusion.
[0011] Step 4, Two-way Iterative Evaluation: Perform fusion feature calculation on the high-value feature dataset to obtain students' practical literacy scores, thereby generating an optimization plan for instructors' abilities and suggestions for adjusting the practical system.
[0012] Preferably, the process of constructing the mentor's competency profile is as follows: the mentor's competency profile is the level of each potential weakness of the mentor.
[0013] The actual scores of each guidance ability dimension of the instructor are obtained from the instructor's baseline capability model. Each practice effect dimension is set, and the actual scores of each guidance ability dimension and each practice effect dimension are standardized to a value in the range [0,1]. The training set and the validation set are divided according to a preset ratio. The training set is combined to obtain the combined dimension features. The model parameters are initialized. The decision tree is iteratively trained with the objective function of minimizing the deviation between the total guidance score and each practice effect dimension to obtain the gradient boosting tree model. The precise scores of each guidance ability dimension of the instructor are output. The precise scores of each guidance ability dimension of the instructor are combined to form the instructor's capability profile, thereby obtaining the weakness level of each potential weakness and generating a personalized training plan.
[0014] Preferably, the generation process of the high-value feature dataset is as follows: cross-scenario practice data includes various types of student practice execution data, various types of instructor guidance data, and various types of course adaptation data. The various types of student practice execution data, various types of instructor guidance data, and various types of course adaptation data are mapped to the [0,1] interval to obtain the normalized values of various types of student practice execution data, various types of instructor guidance data, and various types of course adaptation data.
[0015] By setting weight factors for student practice execution, instructor guidance, and course suitability using the analytic hierarchy process, the normalized values of various types of student practice execution data, instructor guidance data, and course suitability data are multiplied by the corresponding weight factors to obtain corrected values for various types of student practice execution data, instructor guidance data, and course suitability data.
[0016] Combined into a high-value feature dataset: corrected values for various types of student practice execution data, corrected values for various types of instructor guidance data, and corrected values for various types of course adaptation data.
[0017] Preferably, the process of generating the instructor's ability optimization scheme is as follows: the correlation between the instructor's various ability dimensions and the student's comprehensive moral character score is analyzed using the Pearson correlation coefficient method, and each ability dimension with a correlation greater than the preset value is recorded as the core instructor ability dimension.
[0018] Each core guidance capability dimension for which a mentor's weakness level is greater than the first weakness level is recorded as a first-level priority guidance capability dimension; each core guidance capability dimension for which a mentor's weakness level is greater than the second weakness level is recorded as a second-level priority guidance capability dimension; and each core guidance capability dimension for which a mentor's weakness level is greater than the third weakness level is recorded as a third-level priority guidance capability dimension.
[0019] Based on the process of generating personalized training programs, optimization plans are generated for each level 1 priority guidance capability dimension, each level 2 priority guidance capability dimension, and each level 3 priority guidance capability dimension. At the same time, assessment cycles are set for the optimization plans of each priority guidance capability dimension at each level. If the plan fails the assessment for two consecutive cycles, its corresponding priority level is increased. When the optimization plan of a certain level 3 priority guidance capability dimension fails the assessment for two consecutive cycles, an early warning is issued, and the guidance task of the instructor is suspended.
[0020] The beneficial effects of this invention are as follows: 1. This invention first collects historical practice data of instructors, analyzes it using digital technology to construct competency profiles, and generates personalized training plans; secondly, based on student growth data and curriculum demand data, it generates intelligently adapted practice tasks, collects cross-scenario practice data, and through preprocessing, feature extraction, and cross-domain fusion, forms a high-value feature dataset; finally, it obtains student practice literacy scores through feature fusion calculation, and simultaneously generates instructor competency optimization plans and practice system adjustment suggestions. This invention, relying on digital technology, improves the intelligence, accuracy, and dynamism of the evaluation process, effectively enhancing the quality of student practice literacy cultivation and the competency level of instructors.
[0021] 2. This invention collects historical practice data from instructors and uses the analytic hierarchy process (AHP) and a gradient boosting tree model to construct competency profiles. It accurately identifies potential weaknesses and their severity levels, and generates a three-tiered personalized training plan encompassing periodic review, paired assistance, and targeted problem-solving. This solves the problems of standardized training models and ambiguous weakness identification in existing technologies, ensuring that the training plan highly aligns with the actual needs of instructors. It effectively enhances instructors' core guidance capabilities and provides high-quality guidance support for students' practical work.
[0022] 3. This invention, based on core student growth characteristics data and course demand data of historically similar students, generates task suitability through standardized processing and weighted calculation, and pushes practical tasks to students according to the suitability. This avoids the subjectivity and blindness of manual allocation, solves the problems of mismatch between task difficulty and student ability and insufficient course relevance in existing technologies, significantly improves student participation, and ensures the targeted and effective improvement of students' overall quality through practical tasks.
[0023] 4. This invention uses the Pearson correlation coefficient method to identify core guidance competency dimensions, develops differentiated optimization plans for weaknesses at each level, and sets assessment cycles, forming a closed loop for optimizing instructor capabilities from training and assessment to upgrading. Simultaneously, based on student competency scoring data, it identifies weak character dimensions and assigns more excellent instructors to low-fitness practical tasks, achieving dynamic adjustment of the practical system. This solves the problem of disconnect between evaluation and optimization in existing technologies, enabling a virtuous cycle of instructor capability improvement, student competency cultivation, and practical system refinement, continuously improving the quality of practical education. Attached Figure Description
[0024] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0025] Figure 1 This is a schematic diagram of the implementation steps of the method of the present invention. Detailed Implementation
[0026] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0027] according to Figure 1As shown, the present invention provides a student practical literacy evaluation method based on digital intelligence technology, including the following steps: Step 1, empowering instructors: collect the instructors' historical practical data, analyze the instructors' historical practical data through digital intelligence technology, construct an instructor competency profile, and then generate a personalized training plan.
[0028] In one specific embodiment, the collection process for the historical practice data of the instructors is as follows: the historical practice data of the instructors includes, but is not limited to, the instructors' professional background, practice guidance experience, past student practice achievement conversion rate, and student satisfaction evaluation.
[0029] The professional background data of the mentors includes, but is not limited to, their academic qualifications and certificate types, which are obtained through the school's personnel file management system, the mentor qualification certification upload platform, and the industry work experience record form.
[0030] Practical guidance experience includes, but is not limited to, the number of projects guided, the project level, and the guidance period. This information is obtained through the practical project management platform, the basic information entered by the mentor when initiating a project, and the practical education management system.
[0031] The past student practical achievement conversion rate includes, but is not limited to, the award rate and achievement implementation rate of guided student practical projects, which can be obtained through, but is not limited to, the student practical achievement application system, student-submitted award certificates, achievement implementation cooperation agreements, summary tables of instructors' achievements, and the university's practical achievement certification database.
[0032] Student satisfaction ratings are based on students' scores for guidance response speed, the practicality of guidance content, and communication efficiency. These ratings are obtained through a questionnaire sent out after the project ends and anonymous offline evaluation forms.
[0033] It should be noted that composite data is quantized and encoded before being collected.
[0034] In one specific embodiment, the analysis of the instructor's historical practice data is carried out as follows: the instructor's professional background, practice guidance experience, past student practice achievement conversion rate, and student satisfaction evaluation are obtained from the instructor's historical practice data.
[0035] The weights of each guidance capability dimension are determined by the analytic hierarchy process (AHP). The capability data of outstanding historical mentors are obtained from the database, and the mean of each guidance capability dimension is automatically extracted and recorded as the threshold for each guidance capability dimension.
[0036] It should be noted that the various dimensions of guidance ability include, but are not limited to, professional guidance ability, innovation guidance ability, organizational and coordination ability, communication and feedback ability, problem-solving ability, and curriculum integration ability.
[0037] To establish a baseline model for instructor competence: input the target instructor's various instruction competence dimensions data, divide the target instructor's various instruction competence dimensions data by the target instructor's pass threshold to obtain the target instructor's scores for each instruction competence dimension, multiply the target instructor's scores for each instruction competence dimension by the corresponding weighting factor, and output the target instructor's total instruction score.
[0038] It should be noted that the weighting factors for each guidance capability dimension were set by the staff.
[0039] In one specific embodiment, the construction process of the mentor's competency profile is as follows: the mentor's competency profile is the level of each potential weakness of the mentor.
[0040] The actual scores of each guidance ability dimension of the instructor are obtained from the instructor's baseline capability model. Each practice effect dimension is set, and the actual scores of each guidance ability dimension and each practice effect dimension are standardized to a value in the range [0,1]. The training set and the validation set are divided according to a preset ratio. The training set is combined to obtain the combined dimension features. The model parameters are initialized. The decision tree is iteratively trained with the objective function of minimizing the deviation between the total guidance score and each practice effect dimension to obtain the gradient boosting tree model. The precise scores of each guidance ability dimension of the instructor are output. The precise scores of each guidance ability dimension of the instructor are combined to form the instructor's capability profile, thereby obtaining the weakness level of each potential weakness and generating a personalized training plan.
[0041] It should be noted that the actual scores for each dimension of practical effectiveness are obtained as follows: Each dimension of practical effectiveness includes, but is not limited to, the quality score of student practical results, the rate of improvement of student practical ability, and the efficiency of practical project completion. The quality score of student practical results is obtained by the instructors scoring the student's practical results report, works, or patents according to the evaluation criteria corresponding to the "Six Virtues". The average of the student's practical results scores is used to calculate the student's practical results quality score. The corresponding ability scores of the "Six Virtues" are collected one week before the practice and one week after the practice task. The corresponding ability score of the "Six Virtues" one week after the practice task is completed is subtracted from the score one week before the practice task is completed, and then divided by the score one week before the practice task to obtain the student's practical ability improvement rate. The actual completion period from task initiation to result submission is automatically calculated. The planned period of the task is divided by the actual completion period to obtain the practical project completion efficiency. The data for each dimension of practical effectiveness are obtained in this way. After quantification, coding, and normalization, the actual scores for each dimension of practical effectiveness are obtained.
[0042] It should be noted that, firstly, the composite data such as the professional background and practical guidance experience of the instructors are quantitatively encoded, and then the data is uniformly normalized to the [0,1] interval using the min-max standardization method, along with the actual scores of each guidance ability dimension and each practical effect dimension. The number of iterations for iterative training of the decision tree is set by the staff.
[0043] In one specific embodiment, the process of obtaining the weakness level of each potential weakness is as follows: each guidance ability dimension with an accurate score less than a preset accurate score is recorded as a potential weakness; the preset accurate score of each potential weakness is subtracted from the corresponding accurate score, and then divided by the preset accurate score to obtain the weakness percentage of each potential weakness; the weakness percentage range of each level of weakness is obtained from the database; if the weakness percentage of a certain potential weakness belongs to the weakness percentage range of a certain level of weakness, the potential weakness is a weakness of that level; thus, the weakness level of each potential weakness is obtained, and a personalized training plan is generated.
[0044] It should be noted that the upper and lower limits of the proportion of shortcomings at each level are set by the staff.
[0045] In one specific embodiment, the process of generating a personalized training plan is as follows: potential shortcomings with a level greater than the preset first level are recorded as various cyclic review guidance capability dimensions; potential shortcomings with a level greater than the preset second level are recorded as various pairing assistance guidance capability dimensions; and potential shortcomings with a level greater than the preset third level are recorded as various special tackling guidance capability dimensions.
[0046] It should be noted that the first, second, and third shortcoming levels are all preset by the staff.
[0047] We obtain the precise scores of each guidance ability dimension of each historical outstanding guide from the ability data of each historical outstanding guide, and record the outstanding guides whose precise scores for each guidance ability dimension are greater than the preset precise score for assistance as the key personnel.
[0048] It should be noted that the preset precise score for assistance is set by the staff.
[0049] Sort the historical outstanding mentors according to the precise scores of each mentoring ability dimension from largest to smallest to obtain the sequence of historical outstanding mentors corresponding to each mentoring ability dimension. Record the historical outstanding mentors at the first predetermined proportion of the sequence corresponding to each mentoring ability dimension as the mentors of each specialization. In this way, the mentors of each specialization for each mentoring ability dimension are obtained.
[0050] It should be noted that the preset number of participants is set by the staff.
[0051] Provide instructors with online course resources corresponding to the various dimensions of their ability to provide guidance on periodic review processes.
[0052] Provide mentors with support for each of the paired assistance and guidance capabilities dimensions, and assign practical projects that are strongly related to the corresponding weakness dimension to each mentor.
[0053] To provide guidance to personnel on various special tasks, task forces are established, each consisting of personnel responsible for specific tasks within each task force.
[0054] Step 2, Dynamic Practice Execution: Collect student growth data and course requirement data, analyze the student growth data and course requirement data, and generate intelligently adapted practice tasks.
[0055] In one specific embodiment, the collection of student growth data and course requirement data is carried out as follows: student growth data consists of various core characteristics of student growth, and course requirement data consists of various core characteristics of student growth for various tasks of students with similar histories.
[0056] It should be noted that the core characteristics of student growth data include, but are not limited to, academic performance, skills certificates, types of projects participated in, and duration of participation. Academic performance is collected through the academic affairs system, skills certificates are collected through the student's certificate upload port, and students submit scanned copies or electronic versions of certificates. The types of projects participated in and duration of participation are collected through the student practice file system, student-entered practice information, practice project management platform, matching instructor records, or results submission port.
[0057] Method for obtaining historically similar students: Obtain academic performance data and skill certificate data of each historical student from the database, quantify and encode the composite data, and then merge them into vectors to obtain the feature vectors of each historical student and the feature vectors of the target student. The similarity of each historical student is obtained through cosine similarity calculation, and historical students with similarity greater than the preset value are recorded as historically similar students.
[0058] In one specific embodiment, the generation process of intelligently adapted practice tasks is as follows: The core growth feature data of various tasks for historically similar students are standardized according to the interval [0,1]. The initial adaptation coefficients for various practice tasks are calculated using a weighted summation method. The core growth feature data of each student are standardized according to the interval [0,1], then multiplied by the corresponding weight factor, and then multiplied again by the initial adaptation coefficients for each practice task. Finally, the results are summed to obtain the student's adaptation degree for each type of task. Practice tasks are then sent to students in descending order of adaptation degree.
[0059] Step 3: Multi-source data fusion: Based on the instructor's competency profile, personalized training plan, and intelligent adaptation practice tasks, collect cross-scenario practice data. After preprocessing and feature extraction, generate a high-value feature dataset through cross-domain fusion.
[0060] In one specific embodiment, the collection of cross-scenario practice data is carried out as follows: cross-scenario practice data includes various types of student practice execution data, various types of instructor guidance data, and various types of course adaptation data. Various types of student practice execution data include, but are not limited to, task completion progress, participation duration, collaboration frequency, and number of problem-solving times. The data is collected through a practice task management platform, results upload port, or team collaboration tools.
[0061] The guidance data of various instructors includes, but is not limited to, guidance frequency, feedback response time, guidance content quality score, and training program implementation rate, which are collected through the instructor work log system and the student feedback module.
[0062] The data for various courses includes, but is not limited to, knowledge point relevance and knowledge point application rate. A list of core knowledge points is extracted from the course syllabus, and the course leader fills in a list of core knowledge points corresponding to the practical tasks. Scores are assigned to each core knowledge point, and the total course knowledge point score and the corresponding practical task knowledge point score are collected. The knowledge point relevance is obtained by dividing the practical task knowledge point score by the total course knowledge point score. The number of knowledge points actually applied is identified through student practice logs or projects. Simultaneously, the total number of core knowledge points is set during task design. The knowledge point application rate is obtained by dividing the number of actually applied knowledge points by the total number of core knowledge points set during task design.
[0063] In one specific embodiment, the generation of the high-value feature dataset is specifically performed as follows: various types of student practice execution data, various types of instructor guidance data, and various types of course adaptation data are mapped to the [0,1] interval to obtain normalized values for various types of student practice execution data, various types of instructor guidance data, and various types of course adaptation data.
[0064] By setting weight factors for student practice execution, instructor guidance, and course suitability using the analytic hierarchy process, the normalized values of various types of student practice execution data, instructor guidance data, and course suitability data are multiplied by the corresponding weight factors to obtain corrected values for various types of student practice execution data, instructor guidance data, and course suitability data.
[0065] Combined into a high-value feature dataset: corrected values for various types of student practice execution data, corrected values for various types of instructor guidance data, and corrected values for various types of course adaptation data.
[0066] Step 4, Two-way Iterative Evaluation: Perform fusion feature calculation on the high-value feature dataset to obtain students' practical literacy scores, thereby generating an optimization plan for instructors' abilities and suggestions for adjusting the practical system.
[0067] In one specific embodiment, the process of obtaining the student's practical literacy score is as follows: setting up secondary indicators for six categories of moral character indicators, obtaining the correction values of each secondary indicator for the six categories of moral character indicators from the high-value feature dataset, and weighting the correction values of each secondary indicator for the six categories of moral character indicators to obtain the student's comprehensive moral character score.
[0068] It should be noted that in the weighted calculation process, staff preset the weighting factors for each secondary indicator and the weighting factors for the six categories of moral character indicators. The correction values of each secondary indicator of the six categories of moral character indicators are multiplied by the corresponding weighting factors and then added together to obtain the correction values of the six categories of moral character indicators. The correction values of the six categories of moral character indicators are then multiplied by the corresponding weighting factors and then added together to obtain the student's comprehensive moral character score.
[0069] The six categories of moral indicators are: 1. Public morality, and the corresponding secondary indicators include, but are not limited to, the rate of completion of social service hours, the satisfaction of service recipients, and the frequency of volunteer service participation.
[0070] 2. Professional ethics, corresponding to various secondary indicators including but not limited to the quality of professional practice completion and the application rate of professional knowledge.
[0071] 3. "Bow Virtue" corresponds to various secondary indicators, including but not limited to skills competition results, professional quality scores, and the number of professional skills certificates obtained before and after practice.
[0072] 4. Moral Character, corresponding to various secondary indicators including but not limited to participation rate in labor practice, hands-on ability score, and frequency of daily labor participation.
[0073] 5. Moral cultivation, corresponding to various secondary indicators including but not limited to the number of innovative achievements, project implementation rate, and frequency of participation in scientific research practice.
[0074] 6. Merit, corresponding to each secondary indicator including but not limited to the coverage rate of publicity practice and the publicity feedback score.
[0075] In one specific embodiment, the process of generating the instructor competency optimization scheme is as follows: Pearson correlation coefficients are calculated for the corrected values of various student practice execution data and various instructor guidance data, to obtain the correlation between the various competency dimensions of instructors and the students' comprehensive moral character score, and the competency dimensions that are greater than the preset correlation are recorded as the core instructor competency dimensions.
[0076] Each core guidance capability dimension for which a mentor's weakness level is greater than the first weakness level is recorded as a first-level priority guidance capability dimension; each core guidance capability dimension for which a mentor's weakness level is greater than the second weakness level is recorded as a second-level priority guidance capability dimension; and each core guidance capability dimension for which a mentor's weakness level is greater than the third weakness level is recorded as a third-level priority guidance capability dimension.
[0077] Based on the process of generating personalized training programs, optimization plans are generated for each level 1 priority guidance capability dimension, each level 2 priority guidance capability dimension, and each level 3 priority guidance capability dimension. At the same time, assessment cycles are set for the optimization plans of each priority guidance capability dimension at each level. If the plan fails the assessment for two consecutive cycles, its corresponding priority level is increased. When the optimization plan of a certain level 3 priority guidance capability dimension fails the assessment for two consecutive cycles, an early warning is issued, and the guidance task of the instructor is suspended.
[0078] In one specific embodiment, the generation process of the proposed adjustment of the practice system is as follows: extract student practice literacy score data within a preset time period, and calculate the average score, standard deviation, and percentage of students with scores below the preset threshold for six categories of moral character indicators. If the average score of a certain category of moral character indicator is less than the preset average score of the moral character indicator, or the standard deviation of that category of moral character indicator is greater than the preset standard deviation of the moral character indicator, or the percentage of students with scores below the preset threshold for that category of moral character indicator is greater than the preset percentage of students, then that category of moral character indicator is marked as a weak moral character dimension.
[0079] It should be noted that the preset scores, the preset average score of moral character indicators, the preset standard deviation of moral character indicators, and the preset percentage of students are all set by the staff.
[0080] Increase the number of practice projects for each weak moral dimension by a predetermined number.
[0081] Collect adaptation data of intelligent adaptation practice tasks within a preset time period, and statistically determine the proportion of each type of practice task whose adaptation is less than the preset adaptation. This proportion is recorded as the low adaptation proportion of each type of practice task. Practice tasks whose low adaptation proportion is greater than the preset low adaptation proportion are recorded as various types of low adaptation practice tasks.
[0082] For each low-fitness practice task, assign an additional historically outstanding mentor to the existing mentors to guide the practice.
[0083] The examples described in this invention are not limited to the specific embodiments listed above. The examples are merely illustrative to facilitate understanding of the invention and do not constitute a limitation on the scope of protection of this invention. Any modifications, equivalent substitutions, etc., made within the spirit and principles of this invention should be included within the scope of protection.
[0084] The above description is merely an example and illustration of the concept of the present invention. Those skilled in the art can make various modifications or additions to the specific embodiments described or use similar methods to replace them, as long as they do not deviate from the concept of the invention or exceed the scope defined in this specification, they should all fall within the protection scope of the present invention.
Claims
1. A method for evaluating students' practical literacy based on digital intelligence technology, characterized in that, Includes the following steps: Step 1: Empowering Mentors: Collect historical practice data of mentors, analyze the historical practice data through digital technology, construct a mentor competency profile, and then generate personalized training plans; Step 2, Dynamic Practice Execution: Collect student growth data and course requirement data, analyze the student growth data and course requirement data, and generate intelligently adapted practice tasks; Step 3: Multi-source data fusion: Based on the instructor's competency profile, personalized training plan, and intelligent adaptation practice tasks, collect cross-scenario practice data. After preprocessing and feature extraction, generate a high-value feature dataset through cross-domain fusion. Step 4, Two-way Iterative Evaluation: Perform fusion feature calculation on the high-value feature dataset to obtain students' practical literacy scores, thereby generating an optimization plan for instructors' abilities and suggestions for adjusting the practical system.
2. The student practical literacy evaluation method based on digital intelligence technology according to claim 1, characterized in that, The analysis of the historical practice data of the instructors is carried out in the following specific process: We obtain information from the instructors' historical practice data, including their professional background, practical guidance experience, past student practice outcome conversion rate, and student satisfaction evaluations. The weights of each guidance ability dimension are determined by the analytic hierarchy process, the ability data of each historical outstanding mentor are obtained from the database, and the mean of each guidance ability dimension is automatically extracted and recorded as the threshold for each guidance ability dimension. To establish a baseline model for instructor competence: input the target instructor's various instruction competence dimensions data, divide the target instructor's various instruction competence dimensions data by the target instructor's pass threshold to obtain the target instructor's scores for each instruction competence dimension, multiply the target instructor's scores for each instruction competence dimension by the corresponding weighting factor, and output the target instructor's total instruction score.
3. The student practical literacy evaluation method based on digital intelligence technology according to claim 2, characterized in that, The specific process for constructing the capability profile of the guidance personnel is as follows: The mentor's competency profile is a breakdown of the mentor's potential weaknesses. The actual scores of each guidance ability dimension of the instructor are obtained from the instructor's baseline capability model. Each practice effect dimension is set, and the actual scores of each guidance ability dimension and each practice effect dimension are standardized to a value in the range [0,1]. The training set and the validation set are divided according to a preset ratio. The training set is combined to obtain the combined dimension features. The model parameters are initialized. The decision tree is iteratively trained with the objective function of minimizing the deviation between the total guidance score and each practice effect dimension to obtain the gradient boosting tree model. The precise scores of each guidance ability dimension of the instructor are output. The precise scores of each guidance ability dimension of the instructor are combined to form the instructor's capability profile, thereby obtaining the weakness level of each potential weakness and generating a personalized training plan.
4. The student practical literacy evaluation method based on digital intelligence technology according to claim 3, characterized in that, The specific process for obtaining the severity level of each potential weakness is as follows: Each guidance capability dimension with an exact score lower than the preset exact score is recorded as a potential weakness. The preset exact score of each potential weakness is subtracted from the corresponding exact score, and then divided by the preset exact score to obtain the weakness percentage of each potential weakness. The weakness percentage range of each level of weakness is obtained from the database. If the weakness percentage of a certain potential weakness belongs to the weakness percentage range of a certain level of weakness, the potential weakness is a weakness of that level. In this way, the weakness level of each potential weakness is obtained, and then a personalized training plan is generated.
5. The student practical literacy evaluation method based on digital intelligence technology according to claim 4, characterized in that, The specific process for generating the personalized cultivation plan is as follows: Potential shortcomings with a level greater than the preset first level of shortcomings are recorded as the dimensions of guidance capability for periodic review; potential shortcomings with a level greater than the preset second level of shortcomings are recorded as the dimensions of guidance capability for paired assistance; and potential shortcomings with a level greater than the preset third level of shortcomings are recorded as the dimensions of guidance capability for special tackling tasks. The precise scores of each guidance ability dimension of each historical outstanding guide are obtained from the ability data of each historical outstanding guide. Each outstanding guide whose precise score for each guidance ability dimension is greater than the preset precise score for assistance is recorded as a key personnel. The historical outstanding mentors are sorted in descending order of their precise scores for each mentoring ability dimension to obtain a sequence of historical outstanding mentors for each mentoring ability dimension. The historical outstanding mentors who are at the top of the sequence for each mentoring ability dimension are recorded as the mentors for each specialization. This is how the mentors for each specialization of each mentoring ability dimension are obtained. Provide instructors with online course resources corresponding to the various dimensions of their ability to provide guidance on periodic review processes; Provide mentors with support for each of the various paired assistance and guidance capabilities dimensions, and assign practical projects that are strongly related to the corresponding weakness dimensions to each mentor. To provide guidance to personnel on various special tasks, task forces are established, each consisting of personnel responsible for specific tasks within each task force.
6. The student practical literacy evaluation method based on digital intelligence technology according to claim 5, characterized in that, The specific generation process for the intelligent adaptation practice task is as follows: Student growth data consists of core characteristics of student growth, while course requirement data consists of core characteristics of student growth across various tasks for students with similar past performance. The core growth characteristics of students with similar historical backgrounds are standardized according to the interval [0,1]. The initial fit coefficients of various practical tasks are calculated using a weighted summation method. The core growth characteristics of each student are standardized according to the interval [0,1], then multiplied by the corresponding weight factor, and then multiplied by the initial fit coefficients of various practical tasks. Finally, the results are summed to obtain the fit degree of each student for various tasks. Practical tasks are sent to students in descending order of fit degree.
7. The student practical literacy evaluation method based on digital intelligence technology according to claim 1, characterized in that, The specific process for generating the high-value feature dataset is as follows: Cross-scenario practice data includes various types of student practice execution data, various types of instructor guidance data, and various types of course adaptation data. These data are mapped to the [0,1] interval to obtain normalized values for various types of student practice execution data, various types of instructor guidance data, and various types of course adaptation data. By setting weight factors for student practice execution, instructor guidance, and course adaptation using the analytic hierarchy process, the normalized values of various types of student practice execution data, instructor guidance data, and course adaptation data are multiplied by the corresponding weight factors to obtain the corrected values for various types of student practice execution data, instructor guidance data, and course adaptation data. Combined into a high-value feature dataset: corrected values for various types of student practice execution data, corrected values for various types of instructor guidance data, and corrected values for various types of course adaptation data.
8. The student practical literacy evaluation method based on digital intelligence technology according to claim 1, characterized in that, The specific process for obtaining students' practical skills scores is as follows: We set up secondary indicators for six categories of moral character indicators, obtained correction values for each secondary indicator of the six categories of moral character indicators from the high-value feature dataset, and weighted the correction values of each secondary indicator of the six categories of moral character indicators to obtain the student's comprehensive moral character score.
9. The student practical literacy evaluation method based on digital intelligence technology according to claim 1, characterized in that, The specific process for generating the optimized capability scheme for instructors is as follows: The correlation between various competency dimensions of instructors and students' comprehensive moral character scores was analyzed using the Pearson correlation coefficient method. Each competency dimension with a correlation greater than the preset value was recorded as a core instructor competency dimension. The core guidance ability dimensions of a mentor whose weakness level is greater than the first weakness level are recorded as each first-level priority guidance ability dimension; the core guidance ability dimensions of a mentor whose weakness level is greater than the second weakness level are recorded as each second-level priority guidance ability dimension; and the core guidance ability dimensions of a mentor whose weakness level is greater than the third weakness level are recorded as each third-level priority guidance ability dimension. Based on the process of generating personalized training programs, optimization plans are generated for each level 1 priority guidance capability dimension, each level 2 priority guidance capability dimension, and each level 3 priority guidance capability dimension. At the same time, assessment cycles are set for the optimization plans of each priority guidance capability dimension at each level. If the plan fails the assessment for two consecutive cycles, its corresponding priority level is increased. When the optimization plan of a certain level 3 priority guidance capability dimension fails the assessment for two consecutive cycles, an early warning is issued, and the guidance task of the instructor is suspended.
10. The student practical literacy evaluation method based on digital intelligence technology according to claim 1, characterized in that, The proposed adjustments to the generative practice system are described in the following specific generation process: Extract students' practical literacy scores within a preset time period, and calculate the average score, standard deviation, and percentage of students with scores below the preset threshold for six moral character indicators. If the average score of a certain moral character indicator is less than the preset average score, or the standard deviation of that moral character indicator is greater than the preset standard deviation, or the percentage of students with scores below the preset threshold for that moral character indicator is greater than the preset percentage of students, then that moral character indicator is marked as a weak moral character dimension. Increase the number of practice projects for each weak moral dimension by a predetermined amount; Collect adaptation data of intelligent adaptation practice tasks within a preset time period, and statistically obtain the proportion of each type of practice task whose adaptation is less than the preset adaptation, which is recorded as the low adaptation proportion of each type of practice task. Practice tasks whose low adaptation proportion is greater than the preset low adaptation proportion are recorded as each type of low adaptation practice task. For each low-fitness practice task, assign an additional historically outstanding mentor to the existing mentors to guide the practice.