Course teaching effect evaluation method

Abstract

The invention relates to the technical field of education evaluation, and discloses a course teaching effect evaluation method, which comprises the steps of constructing a two-dimensional evaluation index system containing student ability and teacher teaching as a basic framework of evaluation; based on the two-dimensional evaluation index system, determining an initial weight by adopting an interval intuitionistic fuzzy analytic hierarchy process, and dynamically correcting the initial weight in combination with an entropy weight method to form a combined weight; collecting multi-source data corresponding to the two-dimensional evaluation index system through a triangular verification method, and preprocessing the multi-source data; combining the combined weight with the preprocessed multi-source data, and performing layered fuzzy comprehensive evaluation, thereby realizing quantitative evaluation and weak link diagnosis of the course teaching effect; according to the method, expert knowledge and objective data are fused, so that the evaluation weight is dynamically changed along with actual teaching data, and the problem that evaluation is separated from reality due to weight solidification is solved.

Description

Technical Field

[0001] The present invention relates to the technical field of education evaluation, and more specifically, to a method for evaluating course teaching effects. Background Art

[0002] Course teaching effectiveness evaluation is a core component of higher education's quality assurance system, playing a vital role in guiding teaching reform, optimizing the teaching process, and improving the quality of talent development. With the deepening development of educational concepts, course teaching evaluation has evolved from single-subject evaluation to multi-subject evaluation, from outcome evaluation to process evaluation, and from qualitative evaluation to quantitative evaluation.

[0003] At present, the evaluation of course teaching effect mainly adopts methods such as fuzzy comprehensive evaluation method and data envelopment analysis method. However, the existing technology has the following technical problems in practical application: Existing technologies often use expert scoring or hierarchical analysis to determine weights. Once determined, they are often fixed for a long time and cannot be adaptively adjusted according to the distribution and changes of actual teaching data. When the course content or teaching methods change, the fixed weights cannot reflect the new teaching focus, thus affecting the accuracy of the evaluation. Traditional evaluation methods have subjective biases in the expert judgment link and lack a quantitative processing mechanism for expert uncertainty judgment and hesitation. When experts find it difficult to make a clear judgment on certain indicators, they are often forced to give a deterministic score, ignoring the uncertainty information in the judgment process, resulting in a reduced reliability of the evaluation results. In addition, differences of opinion among experts are difficult to be reasonably expressed and handled in the model. ; Existing evaluation systems mostly focus on the teacher's teaching process and lack attention to students' ability output; or they only focus on students' course grades and ignore the quality assessment of the teaching process, and are unable to build a complete evaluation closed loop of "teaching implementation → ability generation → industry docking". This single-dimensional evaluation cannot fully reflect the complexity and systematic nature of course teaching, and it is difficult to support scientific judgment of teaching quality; existing technologies mostly rely on a single data source (such as student questionnaires or teaching archives), and lack the collection and fusion processing mechanism of multi-source heterogeneous data, resulting in incomplete information in the evaluation process. At the same time, the data preprocessing method is simple, and there is a lack of scientific processing of outliers and missing values, which reduces the reliability of the evaluation results.

[0004] The above-mentioned technical problems lead to inflated scores, unbalanced indicators, inaccurate diagnosis and other phenomena in the evaluation results of course teaching effectiveness, which cannot provide accurate data support and direction guidance for the continuous improvement of the course. A more scientific, objective and dynamic course teaching effectiveness evaluation method is urgently needed. Summary of the Invention

[0005] The present invention provides a course teaching effect evaluation method to solve technical problems in related technologies such as weight solidification, significant subjective bias and single evaluation dimension.

[0006] The present invention provides a course teaching effect evaluation method, comprising: Construct a two-dimensional evaluation indicator system that includes student ability and teacher teaching as the basic framework for evaluation; Based on the dual-dimensional evaluation index system, the interval-interval intuitionistic fuzzy hierarchical analysis method is used to determine the initial weights, and the entropy weight method is combined to dynamically modify the initial weights to form combined weights; The multi-source data corresponding to the two-dimensional evaluation index system are collected and pre-processed through the triangulation verification method; The combined weights are combined with the preprocessed multi-source data to conduct hierarchical fuzzy comprehensive evaluation, thereby achieving quantitative evaluation of course teaching effects and diagnosis of weak links.

[0007] Furthermore, the construction of the dual-dimensional evaluation index system includes: Under the student ability dimension, three first-level indicators are set: engineering practice ability, technical documentation ability, and collaborative innovation ability; Under the teacher teaching dimension, four first-level indicators are set: engineering teaching literacy, technology frontier integration, teaching process execution, and depth of industry-education interaction; Each first-level indicator is further decomposed into second-level indicators, forming a complete evaluation system consisting of 7 first-level indicators and 14 second-level indicators.

[0008] Furthermore, the determination of the initial weights by using the interval-interval intuitionistic fuzzy hierarchical analysis method includes: Quantify the uncertainty of expert judgment through interval-valued intuitionistic fuzzy; The weighted arithmetic average operator was used to aggregate the judgments of multiple experts; pass -The cut-set method converts the fuzzy judgment matrix into an exact numerical matrix; Solve the maximum eigenvalue and its corresponding eigenvector to obtain the initial weight.

[0009] Furthermore, the hierarchical fuzzy comprehensive evaluation includes: Set up an evaluation set and divide it into five levels: excellent, good, average, qualified, and unqualified; Calculate the membership of each indicator to the evaluation set through frequency statistics method; The weighted average fuzzy synthesis operator is used to aggregate the secondary indicator layer and the primary indicator layer; The total fuzzy evaluation vector is defuzzified by the weighted average method to obtain the final quantitative score.

[0010] Furthermore, the dynamic correction of the initial weights by combining the entropy weight method includes: The original data were standardized by range method; Calculation indicators Information entropy of Calculating objective entropy weight based on information entropy nn; The final dynamic weight is obtained by fusion formula: ; in represents the final dynamic weight; represents the subjective initial weight; represents the objective entropy weight; represents the fusion coefficient.

[0011] Furthermore, the value of the fusion coefficient is determined based on the information entropy of the index data. and coefficient of variation Make adaptive adjustments: when and hour, The value is 0.3, which increases the proportion of objective entropy weight; when and hour, The value range is 0.7, increasing the proportion of subjective weight; when and , The value is 0.5.

[0012] Furthermore, the collecting of multi-source data corresponding to the dual-dimensional evaluation index system and pre-processing includes: Obtain subjective evaluation data through anonymous student questionnaires; Obtain objective capability indicator data through experimental reports and platform logs; Obtain teaching process record data through teaching archives and enterprise certificates; Obtain expert observation and evaluation data through expert rating forms; The K nearest neighbor algorithm was used to interpolate missing values; The box plot method was used to identify and handle outliers.

[0013] Furthermore, the weak link diagnosis is achieved by combining information entropy and coefficient of variation: When the information entropy of the secondary indicator And the coefficient of variation When , it is judged as a high dispersion indicator, which is a weak link with significant cognitive differences among students; When the information entropy of the secondary indicator And the coefficient of variation When it is judged as a low-controversy indicator, it belongs to the weak link with high evaluation consistency but insufficient teaching depth; When the secondary indicators are defuzzified by weighted average method, the quantitative scores When the problem is solved, it is identified as a weak link that needs to be improved.

[0014] Furthermore, the secondary indicator layer aggregation adopts a weighted average fuzzy synthesis operator: ; in Indicates the A fuzzy evaluation vector of the first-level indicators; Indicates the The transposition of the weight vector of the second-level indicators under the first-level indicators; Indicates the The membership sub-matrix of the first-level indicators; The calculation formula for the first-level indicator layer aggregation is: ; in represents the total fuzzy evaluation vector of the course; Represents the transpose of the dynamic weight vector of the first-level indicator; Represents the first-level indicator evaluation matrix.

[0015] The present invention provides a computer storage medium comprising a memory and one or more processors, wherein the memory stores executable code, and when the one or more processors execute the executable code, they are used to implement the above-mentioned method for evaluating the teaching effect of a course.

[0016] The beneficial effects of the present invention are as follows: the subjective initial weights are dynamically corrected by the entropy weight method, an adaptive weight adjustment mechanism is established, and by integrating expert knowledge and objective data, the evaluation weights are dynamically changed with the actual teaching data, solving the problem of evaluation being out of touch with reality due to weight solidification. The redistribution is closer to the actual data distribution and can adapt to changes in teaching content and methods. Interval-valued intuitionistic fuzzy numbers are introduced to quantify the uncertainty of expert judgments. The support interval, opposition interval, and hesitation interval are used to comprehensively reflect the fuzziness and uncertainty of expert judgments. The weighted arithmetic average operator is used to integrate the judgments of multiple experts, effectively reducing the randomness and bias of individual expert judgments. This invention constructs a dual-dimensional evaluation index system of "student ability-teacher teaching", combining the teaching process with student ability output, forming a complete evaluation closed loop of "teaching implementation → ability generation → industry connection". This dual-dimensional collaborative evaluation mechanism makes the evaluation results more comprehensive and systematic, and can comprehensively reflect all aspects of teaching quality. The triangulation method was used to collect data from multiple channels, including students' subjective evaluations, objective ability indicators, teaching process records, and expert observation and evaluation, ensuring the comprehensiveness and reliability of the data. At the same time, scientific methods were used to interpolate missing values ​​and identify and process outliers, improving data quality and providing a reliable data foundation for subsequent evaluation. Based on hierarchical fuzzy comprehensive evaluation technology, the present invention can accurately locate weak links in teaching, successfully identify weak indicators of courses, and reveal the causes of weakness through data analysis, providing a clear direction for course improvement. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 It is a flow chart of a method for evaluating course teaching effectiveness in the present invention; Figure 2 It is a display diagram of the two-dimensional course evaluation indicator system; Figure 3 It is a diagram showing the scale of interval-valued intuitionistic fuzzy numbers; Figure 4 It is a display diagram of the initial weight distribution of the first-level indicators; Figure 5 It is a diagram showing the data collection framework for teaching effectiveness evaluation; Figure 6 This is the distribution diagram of the interval distribution of the interval intuitionistic fuzzy numbers of the P1 indicator by 10 experts; Figure 7 It is a bar chart comparing the weights of the first-level indicators before and after entropy weight correction; Figure 8 It is a display diagram of the five-level evaluation criteria; Figure 9 It is a display diagram for weak indicator diagnosis; Figure 10 This is a diagram showing the model comparison and verification results. DETAILED DESCRIPTION

[0018] The subject matter described herein will now be discussed with reference to example embodiments. It should be understood that these embodiments are discussed solely to enable those skilled in the art to better understand and implement the subject matter described herein, and that the functions and arrangements of the elements discussed may be varied without departing from the scope of this specification. Various examples may omit, substitute, or add various processes or components as needed. Furthermore, features described in some examples may be combined in other examples.

[0019] At least one embodiment of the present invention discloses a method for evaluating the teaching effect of a course. Figure 1 As shown, including: Step 1: Construct a two-dimensional evaluation index system that includes student ability and teacher teaching as the basic framework for evaluation; In order to fully reflect the teaching effect of the course, this method constructs a two-dimensional evaluation index system of "student ability-teacher teaching". This system includes primary indicators and secondary indicators. The specific construction process is as follows: Step 1.1, determine the evaluation dimensions; Two top-level dimensions are set: "student ability" and "teacher teaching." The student dimension focuses on the output effect of the course, that is, the various abilities that students acquire through course learning; the teacher dimension focuses on the quality of the teaching implementation process.

[0020] Step 1.2, set the first-level indicators; Under the two dimensions, seven first-level indicators (P1-P7) are set; The student dimensions are divided into: engineering practice ability (P1), technical documentation ability (P2), and collaborative innovation ability (P3); The teacher dimension includes: engineering teaching literacy (P4), integration of cutting-edge technology (P5), teaching process execution (P6), and depth of industry-education interaction (P7).

[0021] Step 1.3, decompose the secondary indicators; Each primary indicator is further broken down into 14 quantifiable secondary indicators (b1-b14). For example, the primary indicator "Engineering Practice Capability (P1)" is broken down into two secondary indicators: "Power Supply and Distribution System Commissioning (b1)" and "Lighting Solution Implementation (b2)." Each secondary indicator is linked to a specific engineering capability observation point to ensure operability and objectivity in the evaluation.

[0022] The specific two-dimensional course evaluation indicator system is as follows: Figure 2 As shown, the first-level indicators, second-level indicators, engineering capability observation points and data sources under the student dimension and teacher dimension are displayed.

[0023] Step 2: Based on the two-dimensional evaluation index system, the interval-based intuitionistic fuzzy hierarchical analysis method is used to determine the initial weights, and the entropy weight method is combined to dynamically modify the initial weights to form combined weights; In order to solve the problem of static weights and strong subjectivity in traditional evaluation, this method uses a combination of subjective and objective methods to determine dynamic weights.

[0024] Step 2.1, determine the initial weights based on interval-valued intuitionistic fuzzy analytic hierarchy process; This step aims to reduce the subjective bias of expert scoring by introducing mathematical tools that can handle ambiguity and uncertainty.

[0025] Step 2.1.1, custom term explanation; Interval-Valued Intuitionistic Fuzzy Number (IVIFN): A mathematical form used to describe uncertain information. For an evaluation object, an interval-valued intuitionistic fuzzy number consists of three parts: support interval, opposition interval, and hesitation interval, which can be expressed as: ; in, represents interval-valued intuitionistic fuzzy number; represents the lower bound of support, represents the upper bound of support, Indicates the lower bound of opposition, represents the upper bound of the degree of opposition, and ; The symbol for representing interval-valued intuitionistic fuzzy numbers; The symbol representing the interval; is the support interval, which indicates the extent to which experts believe that one indicator is better than another; The disapproval interval indicates the range of degrees to which experts believe that one indicator is not superior to another; Hesitation interval :Depend on and The decision represents the degree of uncertainty that prevents the expert from making a clear judgment. The introduction of the hesitation interval is one of the keys to overcoming subjective bias in this method.

[0026] Step 2.1.2, construct the interval-valued intuitionistic fuzzy judgment matrix; invite experts (in this embodiment ) Compare the importance of indicators at each level. Figure 3 As shown in the figure, the experts make judgments based on the preset interval intuitionistic fuzzy number scale, forming a personal judgment matrix for each expert: ; in, Indicates the The interval-valued intuitionistic fuzzy judgment matrix of experts; Indicates the Experts on indicators and indicators The comparison results; Indicates the total number of indicators; express OK A matrix of columns.

[0027] In this example, the distribution of support and opposition intervals when 10 experts evaluate the indicators is as follows: Figure 6 As shown in the figure, it intuitively shows the differences in the judgments of different experts on the same indicator (taking P1 engineering practice ability as an example), where the horizontal axis represents the expert number and the vertical axis represents the range of support and opposition. Figure 1 It can be clearly seen that even for the same indicator, there are certain differences in the judgments of different experts, which is why it is necessary to introduce interval-based intuitionistic fuzzy numbers.

[0028] Step 2.1.3, group opinion aggregation; In order to gather the opinions of all experts, the weighted arithmetic averaging (WAA) operator is used to Aggregate the individual judgment matrices to obtain the group judgment matrix ; WAA operator definition: an information fusion operator, whose calculation formula is as follows:

[0029] in, represents the aggregated interval intuitionistic fuzzy number; Indicates the Experts on indicators and indicators The lower bound of support for comparison; Indicates the Experts on indicators and indicators The upper bound of support for comparison; Indicates the Experts on indicators and indicators The lower bound of the degree of opposition for comparison; Indicates the Experts on indicators and indicators The upper bound of the degree of opposition of comparison; Indicates the The weight of the experts; represents the total number of experts; Represents the multiplication operator.

[0030] Step 2.1.4, defuzzification and weight calculation; The aggregated fuzzy judgment matrix Convert to exact numerical matrix , in order to calculate the eigenvalue; - Definition of the cut-set method: A method for converting fuzzy numbers into clear numerical values. The formula used in this embodiment is: ; in, Represents the exact value after defuzzification; represents the optimism coefficient; represents the lower bound of support; represents the upper bound of support; Indicates the lower bound of the degree of opposition; It indicates the upper bound of the degree of opposition; By solving the exact numerical matrix The maximum eigenvalue of and its corresponding eigenvector to obtain the normalized initial weight vector , specifically expressed as: Solving the Matrix The characteristic equation of , get all the eigenvalues, The characteristic equation is: ; in, represents the exact numerical matrix after defuzzification; represents the eigenvalue; Represents the identity matrix.

[0031] Select the largest one from all eigenvalues, which is , then solve for The eigenvector of and normalize it to get the weight vector .

[0032] In this embodiment, the initial weight vector of the first-level indicator is: ; in is the normalized initial weight vector, that is, the initial weight vector of the first-level indicator; Represents the transpose operator.

[0033] The initial weight distribution of the first-level indicators obtained by the above calculation is as follows: Figure 4 As shown, the weight values, rankings and engineering significance of the seven first-level indicators are displayed.

[0034] Step 2.2, weight correction based on entropy weight method; This step aims to use objective data to correct subjective weights and solve the problem of static weight solidification.

[0035] Step 2.2.1, custom term explanation; Information Entropy: Originated from information theory, it is used to measure the uncertainty or amount of information in a system. In this method, the more discrete the scoring data of an indicator, the smaller its information entropy, indicating that the indicator contains more valid information and should be given a higher weight.

[0036] Step 2.2.2, data standardization; collect samples (in this embodiment students) Secondary indicators ( ) score data, forming the original matrix In order to eliminate the dimension effect, the range method is used to standardize the data: ; in, Indicates the data value after range method standardization; Indicates the first Sample pair The score of each indicator; Indicates the The minimum value of an indicator; Indicates the The maximum value of the indicator.

[0037] Step 2.2.3, calculate information entropy and entropy weight; Calculation indicators Information entropy : ; in, Indicates the The information entropy of each indicator; represents the number of samples; Indicates the Samples in the index The percentage of ratings on represents the natural logarithm function; represents the summation operator; Calculation indicators The objective entropy weight : ; in, Indicates the The entropy weight of each indicator; Indicates the The information entropy of each indicator; Indicates the total number of indicators; Indicates that all Sum of indicators.

[0038] Information entropy The smaller the value, the higher the dispersion of the indicator data, the greater the amount of effective information it carries, and the higher the entropy weight. As the value increases, the objectivity of weight correction is not enhanced, and the coefficient of variation is introduced to quantify the relative dispersion of indicator data: ; in, Indicates the The coefficient of variation of each indicator; Indicates the The standard deviation of each indicator; Indicates the The mean of the indicators.

[0039] Step 2.2.4, dynamic weight fusion; The subjective initial weights obtained by IVIF-AHP The objective weight obtained by entropy weight method Perform weighted combination to form the final dynamic weight: ; in represents the final dynamic weight; represents the subjective initial weight; represents the objective entropy weight; Represents the fusion coefficient.

[0040] Fusion coefficient The value of is based on the information entropy of the indicator data and coefficient of variation Make adaptive adjustments: when and hour, The value is 0.3, which increases the proportion of objective entropy weight; when and hour, The value range is 0.7, increasing the proportion of subjective weight; when and , The value is 0.5.

[0041] The effect of weight adjustment is as follows Figure 7As shown in the figure, the changes in the weights of the first-level indicators before and after the entropy weight correction are shown. The weight of the technology frontier integration (P5) indicator increased from 0.087 to 0.142, an increase of 63.2%; while the weight of the industry-education interaction depth (P7) indicator dropped sharply from 0.060 to 0.018, a decrease of 70.0%. The weight adjustment reflects the effective adjustment of the static initial weight by the entropy weight correction mechanism based on actual data, making the final weight more in line with the actual teaching situation.

[0042] Step 3: Collect and pre-process multi-source data corresponding to the two-dimensional evaluation index system through triangulation verification method; Step 3.1, data collection; Use triangulation to collect data from multiple sources, including: Students’ subjective evaluation: obtained through anonymous questionnaire; Objective capability indicators: obtained through experimental reports, platform logs, etc. Teaching process records: obtained through teaching archives, enterprise certificates, etc.; Expert observation and evaluation: obtained through expert scoring form.

[0043] Step 3.2, data preprocessing; Correction of reverse indicators in the questionnaire; K-NearestNeighbor (KNN) algorithm is used to interpolate missing values; Use box plot method to identify and handle outliers; All data were normalized and a standardized data matrix was constructed.

[0044] The complete data collection framework is as follows Figure 5 As shown, the four core data types, acquisition methods, sample sizes, engineering capability mapping, and data standardization methods are listed in detail.

[0045] Step 4: Combine the combined weights with the pre-processed multi-source data to conduct a hierarchical fuzzy comprehensive evaluation, thereby achieving a quantitative evaluation of the course teaching effect and diagnosis of weak links; This step is used to convert qualitative evaluation language into quantitative scores and perform hierarchical aggregation.

[0046] Step 4.1, custom term explanation; Evaluation Set: A set of all possible evaluation levels. In this embodiment, the evaluation set is: ; And corresponding quantitative scores: ; in, represents the rating set; 、 、 、 、 Represents five evaluation levels: excellent, good, average, qualified, and unqualified; The specific five-level evaluation criteria are as follows: Figure 8 As shown, the score range, score assignment and mapping relationship with engineering capabilities corresponding to each evaluation level are listed in detail.

[0047] For example, the "Excellent" grade has a score range of 90 to 100, with a score of 95 indicating full compliance with industry-leading standards. The "Unsatisfactory" grade has a score range of 0 to 49, with a score of 30 indicating a significant deviation from engineering education quality requirements. This quantitative evaluation standard ensures objectivity and comparability of evaluation results.

[0048] Step 4.2, construct the membership matrix; based on The standardized data of samples are used to calculate each secondary index using frequency statistics. Rating set Membership : ; in, Indicator Evaluation level The degree of membership; Indicator Rated The number of samples; Indicates the total number of samples.

[0049] Combine the membership vectors of all secondary indicators under the same primary indicator to form the membership submatrix of the primary indicator .

[0050] Step 4.3, hierarchical aggregation calculation; Secondary indicator layer aggregation: Use the secondary indicator dynamic weight vector calculated in the second step The corresponding membership submatrix Perform fuzzy synthesis operation to obtain the fuzzy evaluation vector of the first-level index .

[0051] ; in Indicates the A fuzzy evaluation vector of the first-level indicators; Indicates the The transposition of the weight vector of the second-level indicators under the first-level indicators; Indicates the The membership matrix of the first-level indicators.

[0052] This operation adopts the weighted average fuzzy synthesis operator.

[0053] Calculation example: Taking the first-level indicator (Engineering practice ability) as an example, assuming that its subordinate indicators The dynamic weight of is: ; The membership matrix is ​​(for simplicity, only three levels are listed): ; but: ; in Represents the weight vector of the secondary indicators under the P1 indicator; represents the transpose operator; represents the membership matrix of the P1 indicator; represents the fuzzy evaluation vector of the P1 index; First-level indicator layer aggregation: the fuzzy evaluation vector of all first-level indicators Combined into a first-level indicator evaluation matrix , and then use the dynamic weight vector of the first-level indicator Perform secondary aggregation to obtain the total fuzzy evaluation vector of the course .

[0054] ; in, represents the total fuzzy evaluation vector of the course; Represents the transpose of the dynamic weight vector of the first-level indicator; Represents the first-level indicator evaluation matrix.

[0055] Step 4.4, defuzzification calculation; In order to obtain the final single quantitative score, the weighted average method is used to calculate the total fuzzy evaluation vector Perform defuzzification: ; in, represents the final quantitative score; represents the total fuzzy evaluation vector of the course; represents the transpose of the rating set score vector; represents the summation operator; Indicates the total fuzzy evaluation vector The value of Indicates the assessment concentration Corresponding score.

[0056] Weak link diagnosis by combining information entropy and coefficient of variation , Comprehensively judge the sensitivity of indicators: When the information entropy of the secondary indicator And the coefficient of variation When , it is judged as a high dispersion indicator, which is a weak link with significant cognitive differences among students; When the information entropy of the secondary indicator And the coefficient of variation When it is judged as a low-controversy indicator, it belongs to the weak link with high evaluation consistency but insufficient teaching depth; When the secondary indicators are defuzzified by weighted average method, the quantitative scores When the problem is solved, it is identified as a weak link that needs to be improved.

[0057] Through the analysis of fuzzy comprehensive evaluation results and dynamic correction of indicator weights, the key shortcomings in the teaching process can be accurately located, such as Figure 9 As shown, the secondary indicators with scores below 85 points are concentrated in the fields of technological frontiers, ethical education and integration of industry and education. The scoring situation of each secondary indicator is shown, among which smart lighting technology (b9) scored 78.5 points, engineering ethics care (b14) scored 76.8 points, and industry case update (b7) scored 82.3 points. These indicators are identified as weak links in course teaching and need to be improved.

[0058] In order to verify the superiority of the IVIF-AHP-EWFCE model proposed in this example, a comparative analysis was conducted between it and the traditional expert scoring method. The results are as follows: Figure 10 The figure shows the differences between the two models in terms of comprehensive scoring, industry-education interaction (P7) scores, and uncertainty quantification capabilities. The model's comprehensive score of 90.38 is more objective than the traditional expert scoring method's 94.25. In terms of industry-education interaction (P7) scores, the model scored 76.2, while the traditional method scored 88.7, a difference of 12.5 points. This indicates that the model effectively reduces the phenomenon of inflated evaluations. Furthermore, the model supports the calculation of hesitation intervals, while the traditional method ignores expert disagreement, further demonstrating the scientific nature and superiority of the model.

[0059] A computer storage medium includes a memory and one or more processors, wherein the memory stores executable code, and when the one or more processors execute the executable code, it is used to implement the above-mentioned course teaching effect evaluation method.

[0060] The above describes an embodiment of the present invention, but this embodiment is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Ordinary technicians in this field can also make more forms of equivalent embodiments based on the inspiration of this embodiment, all of which are protected by this embodiment.

Claims

1. A method for evaluating course teaching effectiveness, characterized in that: include: Construct a two-dimensional evaluation indicator system that includes student ability and teacher teaching as the basic framework for evaluation; Based on the dual-dimensional evaluation index system, the interval-interval intuitionistic fuzzy hierarchical analysis method is used to determine the initial weights, and the entropy weight method is combined to dynamically modify the initial weights to form combined weights; The multi-source data corresponding to the two-dimensional evaluation index system are collected and pre-processed through the triangulation verification method; The combined weights are combined with the preprocessed multi-source data to conduct hierarchical fuzzy comprehensive evaluation, thereby achieving quantitative evaluation of course teaching effects and diagnosis of weak links.

2. A course teaching effect evaluation method according to claim 1, characterized in that: The construction of the two-dimensional evaluation index system includes: Under the student ability dimension, three first-level indicators are set: engineering practice ability, technical documentation ability, and collaborative innovation ability; Under the teacher teaching dimension, four first-level indicators are set: engineering teaching literacy, technology frontier integration, teaching process execution, and depth of industry-education interaction; Each first-level indicator is further decomposed into second-level indicators, forming a complete evaluation system consisting of 7 first-level indicators and 14 second-level indicators.

3. A course teaching effect evaluation method according to claim 1, characterized in that: The method of determining the initial weight by using the interval-interval intuitionistic fuzzy hierarchical analysis method includes: Quantify the uncertainty of expert judgment through interval-valued intuitionistic fuzzy; The weighted arithmetic average operator was used to aggregate the judgments of multiple experts; pass -The cut-set method converts the fuzzy judgment matrix into an exact numerical matrix; Solve the maximum eigenvalue and its corresponding eigenvector to obtain the initial weight.

4. A course teaching effect evaluation method according to claim 1, characterized in that: The hierarchical fuzzy comprehensive evaluation includes: Set up an evaluation set and divide it into five levels: excellent, good, average, qualified, and unqualified; Calculate the membership of each indicator to the evaluation set through frequency statistics method; The weighted average fuzzy synthesis operator is used to aggregate the secondary indicator layer and the primary indicator layer; The total fuzzy evaluation vector is defuzzified by the weighted average method to obtain the final quantitative score.

5. A course teaching effect evaluation method according to claim 1, characterized in that: The dynamic correction of the initial weight by combining the entropy weight method includes: The original data were standardized by range method; Calculation indicators Information entropy of Calculating objective entropy weight based on information entropy nn; The final dynamic weight is obtained by fusion formula: ； in represents the final dynamic weight; represents the subjective initial weight; represents the objective entropy weight; represents the fusion coefficient.

6. A course teaching effect evaluation method according to claim 5, characterized in that: The value of the fusion coefficient is determined based on the information entropy of the indicator data. and coefficient of variation Make adaptive adjustments: when and hour, The value is 0.3, which increases the proportion of objective entropy weight; when and hour, The value range is 0.7, increasing the proportion of subjective weight; when and , The value is 0.

5.

7. A course teaching effect evaluation method according to claim 1, characterized in that: The collecting and preprocessing of multi-source data corresponding to the dual-dimensional evaluation index system includes: Obtain subjective evaluation data through anonymous student questionnaires; Obtain objective capability indicator data through experimental reports and platform logs; Obtain teaching process record data through teaching archives and enterprise certificates; Obtain expert observation and evaluation data through expert rating forms; The K nearest neighbor algorithm was used to interpolate missing values; The box plot method was used to identify and handle outliers.

8. A course teaching effect evaluation method according to claim 1, characterized in that: The weak link diagnosis is achieved by combining information entropy and coefficient of variation: When the information entropy of the secondary indicator And the coefficient of variation When , it is judged as a high dispersion indicator, which is a weak link with significant cognitive differences among students; When the information entropy of the secondary indicator And the coefficient of variation When it is judged as a low-controversy indicator, it belongs to the weak link with high evaluation consistency but insufficient teaching depth; When the secondary indicators are defuzzified by weighted average method, the quantitative scores When the problem is solved, it is identified as a weak link that needs to be improved.

9. A method for evaluating course teaching effectiveness according to claim 4, characterized in that: The secondary indicator layer aggregation adopts weighted average fuzzy synthesis operator: ； in Indicates the A fuzzy evaluation vector of the first-level indicators; Indicates the The transposition of the weight vector of the second-level indicators under the first-level indicators; Indicates the The membership sub-matrix of the first-level indicators; The calculation formula for the first-level indicator layer aggregation is: ； in represents the total fuzzy evaluation vector of the course; Represents the transpose of the dynamic weight vector of the first-level indicator; Represents the first-level indicator evaluation matrix.

10. A computer storage medium, characterized in that The method comprises a memory and one or more processors, wherein the memory stores executable code, and when the one or more processors execute the executable code, the method is used to implement a course teaching effect evaluation method according to any one of claims 1 to 9.

Images