AHP-based student development evaluation method and system

Abstract

The invention relates to the technical field of data processing, and discloses an AHP-based student development evaluation method and system. The method comprises the steps of performing feature root decomposition, consistency check and hierarchical transitivity analysis on an analytic hierarchy matrix, and performing normalization processing and stability optimization on a weight vector to obtain a hierarchical weight coefficient; performing hierarchical comprehensive calculation, criterion synthesis and multi-level evaluation mapping on the basis of the hierarchical weight coefficient, and combining fuzzy membership analysis and evaluation calibration to obtain a comprehensive evaluation result; and aiming at the comprehensive evaluation result, constructing a multi-criterion decision model and a development path map, and carrying out scheme synthesis and multi-objective optimization analysis to obtain personalized suggestions. According to the method and the device, dynamic optimization and adaptive adjustment of the multi-dimensional evaluation indexes are realized, and personalized development suggestions are generated through a data driving mode, so that scientificity and pertinence of student development evaluation are improved.

Description

Technical Field

[0001] The present application relates to the field of data processing, and in particular to a student development assessment method and system based on AHP. Background Art

[0002] In the existing field of student development assessment technology, traditional evaluation methods mainly rely on single test score evaluation and subjective teacher evaluation. The current evaluation system usually adopts a scoring system or a grade system for evaluation, and the evaluation indicators involve multiple aspects such as morality, intelligence, physical fitness, aesthetics, and labor. Some evaluation methods have begun to introduce the analytic hierarchy process (AHP) to determine the weight of each evaluation indicator, and collect evaluation data through questionnaires, expert scoring, etc. At the same time, some evaluation systems have also begun to try to use data mining and machine learning technology to analyze students' historical performance data and behavioral performance in order to obtain more objective evaluation results.

[0003] However, existing evaluation methods have problems such as single evaluation dimension, limited data processing capabilities, and highly subjective evaluation results. In particular, when faced with a large amount of multi-source heterogeneous data, traditional methods are difficult to effectively handle the correlation and interaction between data. At the same time, existing technologies often lack scientific quantitative basis when allocating weights for multi-dimensional evaluation indicators, and the accuracy and reliability of evaluation results are difficult to guarantee. In addition, existing evaluation methods do not take into account individual differences among students, making it difficult to provide students with personalized development suggestions. Summary of the invention

[0004] This application provides a student development assessment method and system based on AHP, which is used to achieve dynamic optimization and adaptive adjustment of multi-dimensional evaluation indicators, and generate personalized development suggestions in a data-driven manner, thereby improving the scientificity and pertinence of student development assessment.

[0005] In a first aspect, the present application provides a student development assessment method based on AHP, and the student development assessment method based on AHP includes: hierarchical decomposition of evaluation indicators, calculation of inter-layer coupling and cross-impact analysis, and combining dynamic reconstruction and criterion mapping of the hierarchical structure to obtain an evaluation indicator system; for the evaluation indicator system, feature decomposition, dimension alignment and modal fusion of the evaluation data are performed, and standardized through data reconstruction and consistency mapping to obtain a hierarchical data set; based on the hierarchical data set, element construction, relative importance calculation and pairwise comparison correction are performed on the judgment matrix, and matrix optimization and dynamic adjustment are performed in combination with the consistency criterion to obtain a hierarchical analysis matrix; for the hierarchical analysis matrix, characteristic root decomposition, consistency test and hierarchical transitivity analysis are performed, and the weight vector is normalized and stability optimized to obtain a hierarchical weight coefficient; based on the hierarchical weight coefficient, hierarchical comprehensive calculation, criterion synthesis and multi-level evaluation mapping are performed, and fuzzy membership analysis and evaluation calibration are combined to obtain a comprehensive evaluation result; for the comprehensive evaluation result, a multi-criteria decision-making model and a development path map are constructed, and scheme synthesis and multi-objective optimization analysis are performed to obtain personalized suggestions.

[0006] In a second aspect, the present application provides a student development assessment system based on AHP, and the student development assessment system based on AHP includes: The decomposition module is used to perform hierarchical decomposition of evaluation indicators, inter-layer coupling calculation and cross-impact analysis, and obtain the evaluation indicator system by combining the dynamic reconstruction of the hierarchical structure and criterion mapping; The fusion module is used to perform feature decomposition, dimension alignment and modal fusion on the evaluation data according to the evaluation index system, and standardize the data through data reconstruction and consistency mapping to obtain a hierarchical data set; The construction module is used to construct elements, calculate relative importance and perform pairwise comparison correction on the judgment matrix based on the hierarchical data set, and optimize and dynamically adjust the matrix in combination with the consistency criterion to obtain the hierarchical analysis matrix; The analysis module is used to perform characteristic root decomposition, consistency test and hierarchical transitivity analysis on the hierarchical analysis matrix, and normalize and optimize the stability of the weight vector to obtain the hierarchical weight coefficient; The calibration module is used to perform hierarchical comprehensive calculation, criterion synthesis and multi-level evaluation mapping based on the hierarchical weight coefficient, and combine fuzzy membership analysis and evaluation calibration to obtain comprehensive evaluation results; The synthesis module is used to construct a multi-criteria decision-making model and development path map based on the comprehensive evaluation results, and to perform solution synthesis and multi-objective optimization analysis to obtain personalized recommendations.

[0007] In the technical solution provided by the present application, a scientific and reasonable evaluation index system can be established by performing hierarchical decomposition of the evaluation indicators, calculating the inter-layer coupling degree and analyzing the cross-impact, so as to realize the accurate quantification of the relationship between the evaluation indicators; by performing feature decomposition, dimension alignment and modal fusion on the evaluation data, and combining the standardized processing of data reconstruction and consistency mapping, the processing problem of multi-source heterogeneous data is effectively solved, and the quality and reliability of the evaluation data are ensured; based on the hierarchical data set, the judgment matrix is ​​constructed with elements, relative importance is calculated and paired comparison is corrected, and the matrix is ​​optimized and dynamically adjusted in combination with the consistency criterion, so that the evaluation process is more objective and accurate; by performing characteristic root decomposition, consistency test and hierarchical transitivity analysis on the hierarchical analysis matrix, and normalizing and optimizing the stability of the weight vector, the scientificity and stability of the evaluation weight are guaranteed; based on the hierarchical weight coefficient, hierarchical comprehensive calculation, criterion synthesis and multi-level evaluation mapping are performed, and fuzzy membership analysis and evaluation calibration are combined to realize the accurate quantification of the evaluation results; finally, by constructing a multi-criteria decision-making model and a development path map, and performing scheme synthesis and multi-objective optimization analysis, personalized development suggestions are provided for students. The overall solution has established a complete and adaptive evaluation system, which not only realizes the normalization and standardization of the evaluation process, but also provides strong data support for educational decision-making through multi-dimensional data analysis and in-depth mining, effectively improving the scientificity, accuracy and practicality of student development assessment. The various technical features of the solution are closely related and mutually supportive, forming a complete technical solution that can effectively solve the problems of strong subjectivity, low accuracy, and insufficient personalization in traditional evaluation methods. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0009] Figure 1 This is a schematic diagram of an embodiment of the student development assessment method based on AHP in the embodiment of the present application; Figure 2 This is a schematic diagram of an embodiment of a student development assessment system based on AHP in an embodiment of the present application. DETAILED DESCRIPTION

[0010] The embodiments of the present application provide a student development assessment method and system based on AHP. The terms "first", "second", "third", "fourth", etc. (if any) in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments described here can be implemented in an order other than that illustrated or described here. In addition, the terms "including" or "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0011] For ease of understanding, the specific process of the embodiment of the present application is described below. Figure 1 In the embodiment of the present application, an embodiment of the student development assessment method based on AHP includes: Step S101, performing hierarchical decomposition, inter-layer coupling degree calculation and cross-impact analysis on the evaluation indicators, and combining dynamic reconstruction of the hierarchical structure and criterion mapping to obtain an evaluation indicator system; Step S102: for the evaluation index system, the evaluation data is subjected to feature decomposition, dimension alignment and modality fusion, and standardized through data reconstruction and consistency mapping to obtain a hierarchical data set; Step S103: Based on the hierarchical data set, construct elements, calculate relative importance, and perform pairwise comparison correction on the judgment matrix, and optimize and dynamically adjust the matrix in combination with the consistency criterion to obtain a hierarchical analysis matrix; Step S104: for the hierarchical analysis matrix, perform characteristic root decomposition, consistency check and hierarchical transitivity analysis, and perform normalization and stability optimization on the weight vector to obtain the hierarchical weight coefficient; Step S105: Based on the hierarchical weight coefficient, hierarchical comprehensive calculation, criterion synthesis and multi-level evaluation mapping are performed, and fuzzy membership analysis and evaluation calibration are combined to obtain a comprehensive evaluation result; Step S106: Based on the comprehensive evaluation results, a multi-criteria decision-making model and a development path map are constructed, and solution synthesis and multi-objective optimization analysis are performed to obtain personalized recommendations.

[0012] It is understandable that the execution subject of the present application may be a student development assessment system based on AHP, or may be a terminal or a server, which is not specifically limited here. The present application embodiment is described by taking a server as the execution subject as an example.

[0013] Specifically, the evaluation indicators are decomposed hierarchically, including dividing the student development evaluation indicators into the target layer, the criterion layer and the indicator layer. The target layer is a comprehensive evaluation of student development, and the criterion layer includes five dimensions: moral development, intellectual development, physical development, aesthetic development and labor education development. The indicator layer is further subdivided into specific evaluation items, such as moral development, which includes ideological and moral character, behavioral habits, social responsibility and other indicators. In the calculation of inter-layer coupling degree, the correlation matrix between indicators is established by calculating the influence degree between different indicators. For cross-impact analysis, the interaction between different dimensions is evaluated, such as the role of physical development in promoting intellectual development. Then, combined with the dynamic reconstruction of the hierarchical structure, the indicator system is dynamically adjusted according to the actual evaluation needs, and the rationality of the adjusted indicator system is ensured through criterion mapping. For the constructed evaluation indicator system, the collected evaluation data is processed in multiple steps. The feature decomposition stage decomposes the original evaluation data into main features and secondary features, and extracts key evaluation information. In the process of dimension alignment, the evaluation data of different dimensions are unified under the same evaluation standard to ensure the comparability of the data. Modal fusion integrates quantitative evaluation data and qualitative evaluation data to form a unified evaluation index. The data reconstruction stage supplements the missing data, and consistency mapping ensures the consistency before and after data processing, and finally obtains a standardized hierarchical data set.

[0014] Based on the hierarchical data set, in the process of constructing the judgment matrix, the elements are first constructed, and the evaluation indicators are compared pairwise to form an initial judgment matrix. The relative importance calculation link quantifies the comparison results and determines the relative importance of the indicators. The paired comparison correction ensures the accuracy of the comparison results through repeated verification and adjustment. The matrix is ​​optimized in combination with the consistency criterion to ensure that the judgment matrix meets the consistency requirements, and the timeliness of the matrix is ​​maintained through dynamic adjustment, and finally a hierarchical analysis matrix is ​​formed.

[0015] The formed hierarchical analysis matrix is ​​decomposed into eigenvalues, and the maximum eigenvalue and its corresponding eigenvector are calculated. The consistency test verifies the rationality of the judgment matrix by calculating the consistency ratio. The hierarchical transitivity analysis examines the transitive relationship between judgments at different levels. The weight vector is normalized so that the sum of the weights of each indicator is 1, and the reliability of the weight is ensured by stability optimization, and finally the hierarchical weight coefficient of each indicator is obtained.

[0016] Based on the obtained hierarchical weight coefficient, hierarchical comprehensive calculation is performed, and the evaluation results of each level are weighted and summarized. The criterion synthesis link integrates the evaluation results of different criterion layers, and the multi-level evaluation mapping establishes the corresponding relationship between the evaluation results and the evaluation level. Combined with fuzzy membership analysis, the degree of membership of the evaluation object to different evaluation levels is determined, and the accuracy of the evaluation results is ensured through evaluation calibration, and finally a comprehensive evaluation result is formed.

[0017] Based on the comprehensive evaluation results, a multi-criteria decision-making model is constructed to comprehensively consider the contributions of multiple evaluation dimensions. The construction of the development path map is based on historical evaluation data to analyze the possible paths of student development. The program synthesis stage integrates multiple development suggestions, and the multi-objective optimization analysis weighs different development goals to ultimately form personalized development suggestions.

[0018] For example, in the hierarchical decomposition of evaluation indicators, intellectual development is subdivided into indicators such as learning ability, subject performance, and innovative thinking. Through the calculation of inter-layer coupling degree, it is found that the coupling coefficient between learning ability and subject performance is 0.85, indicating that the two have a strong correlation. In the data processing process, the student's recent monthly test scores (95 points, 92 points, 88 points) and the completion of regular homework (excellent rate 85%, 90%, 87%) are feature decomposed and dimension aligned. In the judgment matrix construction stage, through pairwise comparison, it is determined that the importance of learning ability relative to subject performance is 1.5, and the importance of innovative thinking relative to subject performance is 1.2. After characteristic root decomposition and consistency test, the weights of learning ability, subject performance, and innovative thinking are calculated to be 0.4, 0.35, and 0.25 respectively. The final comprehensive evaluation results show that the student's intellectual development is at a good level, and it is recommended to improve innovative thinking ability by participating in subject competitions.

[0019] In the embodiment of the present application, by performing hierarchical decomposition of the evaluation indicators, calculating the inter-layer coupling degree and analyzing the cross-impact, a scientific and reasonable evaluation indicator system can be established to achieve accurate quantification of the relationship between the evaluation indicators; by performing feature decomposition, dimension alignment and modal fusion on the evaluation data, and combining the standardized processing of data reconstruction and consistency mapping, the processing problem of multi-source heterogeneous data is effectively solved, and the quality and reliability of the evaluation data are ensured; based on the hierarchical data set, the judgment matrix is ​​constructed with elements, relative importance is calculated and paired comparison is corrected, and the matrix is ​​optimized and dynamically adjusted in combination with the consistency criterion, so that the evaluation process is more objective and accurate; by performing characteristic root decomposition, consistency test and hierarchical transitivity analysis on the hierarchical analysis matrix, and normalizing and optimizing the stability of the weight vector, the scientificity and stability of the evaluation weight are guaranteed; based on the hierarchical weight coefficient, hierarchical comprehensive calculation, criterion synthesis and multi-level evaluation mapping are performed, combined with fuzzy membership analysis and evaluation calibration, the accurate quantification of the evaluation results is achieved; finally, by constructing a multi-criteria decision-making model and a development path map, and performing scheme synthesis and multi-objective optimization analysis, personalized development suggestions are provided for students. The overall solution has established a complete and adaptive evaluation system, which not only realizes the normalization and standardization of the evaluation process, but also provides strong data support for educational decision-making through multi-dimensional data analysis and in-depth mining, effectively improving the scientificity, accuracy and practicality of student development assessment. The various technical features of the solution are closely related and mutually supportive, forming a complete technical solution that can effectively solve the problems of strong subjectivity, low accuracy, and insufficient personalization in traditional evaluation methods.

[0020] In a specific embodiment, the process of executing step S101 may specifically include the following steps: (1) The hierarchical structure of the evaluation index is reduced in dimension through principal component analysis to obtain the initial feature set; (2) Based on the initialization feature set, the interlayer relationship parameters are fitted and calculated using the least squares method to obtain the interlayer coupling coefficient; (3) Quantitatively analyze the cross-influence intensity of the evaluation indicators using the interlayer coupling coefficient to obtain the cross-influence matrix; (4) For the cross-influence matrix, the hierarchical structure is dynamically decomposed through singular value decomposition to obtain the reconstructed feature vector; (5) Using the reconstructed feature vector, a hierarchical clustering algorithm is used to adaptively map the criterion layer to obtain a criterion mapping matrix; (6) Based on the criterion mapping matrix, the evaluation index system is reconstructed and integrated through the hierarchical synthesis algorithm to obtain the evaluation index system.

[0021] Specifically, for the indicator system construction process in the student development assessment method based on AHP, the hierarchical structure of the evaluation indicators is firstly reduced in dimension through principal component analysis. Principal component analysis maps high-dimensional data to low-dimensional data space by linearly transforming the original data. During the processing, the evaluation indicator matrix is ​​decomposed by eigenvalue, the main feature directions are extracted, and the principal components that contribute more to the data variance are retained, so as to obtain the initial feature set after dimensionality reduction.

[0022] For the obtained initialization feature set, the least square method is used to fit the interlayer relationship parameters to obtain the interlayer coupling coefficient. The calculation formula of the interlayer coupling coefficient is: ; in, represents the coupling coefficient between the i-th layer index and the j-th layer index, represents the index value of the kth sample in the i-th layer, represents the index value of the kth sample in the jth layer, and Represent the average values ​​of the indicators at the i-th and j-th layers, respectively. represents the inter-layer weight factor, and n represents the number of samples.

[0023] Based on the calculated inter-layer coupling coefficient, the cross-influence intensity of the evaluation indicators is quantitatively analyzed to construct a cross-influence matrix. The cross-influence matrix reflects the interaction relationship between indicators at different levels, and each matrix element represents the influence intensity between indicators. By analyzing the eigenvalue distribution of the cross-influence matrix, the key influencing factors and dominant relationships in the indicator system are revealed. For the cross-influence matrix, the dynamic decomposition of the hierarchical structure is achieved through singular value decomposition. Singular value decomposition decomposes the original matrix into the product of three matrices, in which the singular value matrix reflects the importance of different characteristic components. By retaining the eigenvectors corresponding to the larger singular values, the reconstructed eigenvectors are obtained, and the effective reconstruction of the hierarchical structure is achieved.

[0024] The hierarchical clustering algorithm is used to adaptively map the criterion layer by using the reconstructed feature vector. The hierarchical clustering algorithm is based on the similarity measurement between indicators, and gradually merges similar indicator categories from bottom to top to form a hierarchical clustering structure. By setting a suitable clustering threshold, the final clustering result is determined and the criterion mapping matrix is ​​obtained. The criterion mapping matrix reflects the corresponding relationship between each evaluation indicator and the criterion layer, providing a basis for the subsequent reconstruction of the indicator system.

[0025] Finally, based on the criterion mapping matrix, the evaluation index system is reconstructed and integrated through the hierarchical synthesis algorithm. In the hierarchical synthesis process, the hierarchical relationship, coupling strength and mapping relationship between indicators are comprehensively considered to construct a complete evaluation index system. The reconstructed index system has the characteristics of clear hierarchy and reasonable structure.

[0026] For example, the original evaluation indicators include 30 specific indicators such as moral development (ideological and moral character, behavioral habits, and social responsibility), intellectual development (learning ability, academic performance, innovative thinking), and physical development (physical fitness, sports skills, and sportsmanship). Through principal component analysis, the 30 indicators were reduced to 15 main characteristic indicators, retaining 85% of the information. The inter-layer coupling coefficients of these 15 characteristic indicators were calculated, and it was found that the coupling coefficient between the ideological and moral character in moral development and the learning attitude in intellectual development was 0.72, indicating that the two had a strong correlation. After constructing the cross-influence matrix, the reconstructed eigenvectors obtained by singular value decomposition showed that moral cultivation, learning ability, and physical fitness were the key indicators in the evaluation system. The hierarchical clustering algorithm was used to classify the 15 characteristic indicators into 5 criterion layers (moral education, intellectual education, physical education, aesthetic education, and labor education) to form a criterion mapping matrix. Finally, through hierarchical synthesis, an evaluation indicator system containing 5 criterion layers and 15 sub-item indicators was constructed.

[0027] In a specific embodiment, the process of executing step S102 may specifically include the following steps: (1) Perform feature subspace division, information entropy weight calculation and dimensional feature extraction on the evaluation data to obtain feature subspace data; (2) For the feature subspace data, the time series information of different dimensions is collaboratively decomposed and mapped across dimensions to obtain an aligned dimension matrix; (3) Based on the alignment dimension matrix, complementary information extraction, feature correlation analysis and heterogeneous information fusion are performed on multi-type evaluation data to obtain fused feature data; (4) Based on the fused feature data, the spatial position of the missing data points is inferred, the neighborhood features are completed, and the structural consistency is verified to obtain the reconstructed data vector; (5) Perform local preservation constraints, global consistency mapping, and feature metric calibration on the reconstructed data vector to obtain a mapping data set; (6) For the mapped data set, probability distribution calculation, data standardization transformation and hierarchical structure reorganization are performed to obtain a hierarchical data set.

[0028] Specifically, feature subspace division is first performed to divide the original evaluation data space into multiple subspaces, each of which corresponds to an evaluation dimension. Information entropy weight calculation determines the importance of each dimension by measuring the degree of uncertainty in the data. The larger the information entropy value, the greater the amount of information contained in the dimension. Dimension feature extraction extracts key feature information from each subspace. For example, for students' moral education evaluation data, including ideological and moral character, moral cultivation, behavioral habits and other dimensions, feature extraction is used to retain the most representative evaluation indicators.

[0029] For the characteristic subspace data, the time series information of different dimensions is collaboratively decomposed, and the time series data is decomposed into three parts: trend item, cycle item and random item. Cross-dimensional mapping establishes the corresponding relationship between different evaluation dimensions, such as aligning moral education evaluation with intellectual education evaluation in the time dimension to obtain the alignment dimension matrix. The alignment dimension matrix reflects the changing rules of different evaluation dimensions in the time series and their mutual relationship.

[0030] Based on the obtained alignment dimension matrix, the multi-type evaluation data is deeply analyzed. In the process of complementary information extraction, complementary information is extracted from different types of evaluation data, such as the complementary features of quantitative evaluation data and qualitative evaluation data. Feature correlation analysis identifies feature combinations with significant correlation by calculating the correlation coefficients between different features. Heterogeneous information fusion integrates evaluation information from different sources and in different forms, such as fusing questionnaire survey results with daily performance records to form unified feature data. For the missing values ​​in the fused feature data, spatial position inference is performed, and the possible values ​​of the missing data points are inferred based on the spatial distribution characteristics of the data. Neighborhood feature completion uses the feature information of adjacent data points to reasonably fill in the missing values. Structural consistency verification ensures that the completed data meets the characteristics of the original data structure and maintains the consistency and rationality of the data. Local preservation constraints are imposed on the reconstructed data vector to ensure that the data maintains the original topological structure characteristics within the local range. Global consistency mapping realizes the global optimization mapping of the data on the basis of maintaining local features. In the process of feature metric calibration, the mapping results are standardized to ensure the comparability of different feature dimensions.

[0031] Finally, the probability distribution calculation is performed on the mapped data set to analyze the statistical characteristics of the data. The data standardization transformation unifies the data of different dimensions into the same scale space. The hierarchical reorganization organizes the standardized data according to the hierarchical structure according to the requirements of the AHP method to form the final hierarchical data set.

[0032] For example, the original data includes monthly moral education scores (ideological and moral character, behavioral performance, etc.), weekly test scores (Chinese, mathematics, etc.), and physical assessment scores (physical fitness test, sports skills, etc.). Through feature subspace division, these data are classified into three subspaces: moral education, intellectual education, and physical education. The information entropy of the moral education evaluation data is calculated, and it is found that the information entropy of the ideological and moral dimension is 0.85, and the information entropy of the behavioral performance dimension is 0.72, indicating that the ideological and moral evaluation contains more information. In the process of time series alignment, the monthly moral education score and the weekly test score are converted to the same evaluation period through time interpolation. For the missing physical assessment data, the neighborhood features are supplemented according to the student's two previous and subsequent assessment scores (85 points and 88 points respectively), and the supplement value is 86 points. After local preservation constraints and global consistency mapping, it is ensured that the supplemented data is consistent with the original evaluation system. The final hierarchical data set contains three levels: target layer (comprehensive quality evaluation), criterion layer (three dimensions of morality, intelligence and physical fitness) and indicator layer (specific evaluation indicators). The data at each level are standardized to ensure the scientificity and comparability of the evaluation data.

[0033] In a specific embodiment, the process of executing step S103 may specifically include the following steps: (1) Extract paired indicators, divide hierarchical intervals, and map indicator space to obtain the initial judgment matrix elements. (2) For the elements of the initial judgment matrix, relative importance quantification, interval value mapping and judgment preference analysis are performed to obtain the importance feature matrix; (3) Using the importance feature matrix, the pairwise comparison data are constrained for consistency, the preference relationship is corrected, and the judgment conflict is eliminated to obtain a corrected judgment matrix; (4) Based on the correction judgment matrix, the indicator combination decomposition, matrix element optimization and hierarchical relationship reconstruction are performed to obtain the optimized matrix structure; (5) Perform consistency constraint calculation, boundary threshold calibration and conflict elimination on the optimized matrix structure to obtain a dynamic adjustment matrix; (6) For the dynamically adjusted matrix, matrix feature extraction, hierarchical structure optimization and element relationship reorganization are performed to obtain a hierarchical analysis matrix.

[0034] Specifically, the evaluation indicators of the same level are paired in pairs to form a basic comparison unit. Hierarchical interval division is to divide the value range of the evaluation indicator into several intervals, each interval corresponding to a different level of importance. Indicator space mapping establishes the correspondence between the indicator value and the judgment matrix element, thereby obtaining the initial judgment matrix element. For the obtained initial judgment matrix elements, relative importance is quantified to convert qualitative judgments into quantitative values. Interval value mapping maps discrete judgment values ​​to continuous numerical intervals to make the evaluation results more accurate. Judgment preference analysis analyzes the judgment tendency of the evaluator, identifies and adjusts subjective biases, and finally forms an importance feature matrix.

[0035] For consistency constraint calculation, the following formula is used: ; in, represents the adjusted consistency index, represents the matrix element, n represents the matrix order, represents the largest characteristic root, Represents the adjustment coefficient. The consistency index calculated by this formula provides a basis for the optimization matrix.

[0036] Starting from the importance feature matrix, the pairwise comparison data is constrained for consistency to ensure that the judgment results meet the transitivity requirements. Preference relationship correction corrects the preference inconsistency that occurs during the evaluation process, and judgment conflict elimination deals with the judgment differences between different evaluation subjects to form a correction judgment matrix.

[0037] Based on the correction judgment matrix, the index combination is decomposed to decompose the complex evaluation problem into several simple judgment tasks. The matrix element optimization fine-tunes the matrix elements through numerical calculation methods, and the hierarchical relationship reconstruction re-establishes the hierarchical association between the indicators to obtain the optimized matrix structure.

[0038] The consistency constraint calculation is performed on the optimized matrix structure to verify whether the matrix meets the consistency requirements. The boundary threshold calibration sets the reasonable value range of the judgment value, and the value outside the range needs to be re-evaluated. The conflict elimination process solves the logical conflicts in the matrix and forms a dynamic adjustment matrix. For the dynamic adjustment matrix, matrix feature extraction is performed to obtain the key eigenvalues ​​and eigenvectors of the matrix. The hierarchical structure optimization adjusts and improves the hierarchical relationship, and the element relationship reorganization reconstructs the correlation between indicators, and finally obtains the hierarchical analysis matrix.

[0039] For example, the specific evaluation indicators include ideology and morality, moral cultivation, and behavioral habits. First, the three indicators are compared in pairs to form an initial judgment matrix of 3×3. Through expert judgment, it is determined that the importance of ideology and morality relative to moral cultivation is 2, and the importance relative to behavioral habits is 3; the importance of moral cultivation relative to behavioral habits is 1.5. After consistency constraint calculation, it is found that the initial consistency ratio is 0.08. After the preference relationship correction, the relative importance of ideology and morality and moral cultivation is adjusted to 1.8, and the consistency ratio is reduced to 0.05. The corrected judgment matrix is ​​optimized to obtain the final eigenvector [0.5, 0.3, 0.2], which corresponds to the weights of ideology and morality, moral cultivation, and behavioral habits. Throughout the process, the consistency of the judgment matrix and the rationality of the hierarchical structure are always maintained.

[0040] In a specific embodiment, the process of executing step S104 may specifically include the following steps: (1) Extract eigenvalues ​​of the hierarchical analysis matrix in segments, calculate the main direction properties, and decompose the eigenspace to obtain the eigendecomposition vector; (2) For the feature decomposition vector, consistency ratio calculation, random sampling verification and deviation threshold constraint are performed to obtain a consistency indicator set; (3) Based on the consistency indicator set, the hierarchical structure is verified for transitivity, the inter-layer relationship constraints and the hierarchical judgment analysis are performed to obtain the transitivity matrix; (4) Based on the transitivity matrix, the weight data is subjected to interval scaling, spatial projection mapping and normalization transformation to obtain a normalized vector; (5) For the normalized vector, stability boundary calculation, fluctuation range constraint and threshold calibration are performed to obtain stability parameters; (6) The stability parameters are weighted, hierarchically combined and optimized to obtain hierarchical weight coefficients.

[0041] Specifically, the processing of the hierarchical analysis matrix first extracts the eigenvalues ​​in segments, and divides the eigenvalues ​​into several segments in order of size, each segment representing a different level of importance. The main direction property calculation analyzes the main direction of the eigenvector and determines the dominant factor of the evaluation index. The feature space decomposition decomposes the original space into the main feature space and the secondary feature space to obtain the feature decomposition vector. For the feature decomposition vector, the consistency ratio calculation is used to evaluate the consistency of the judgment matrix. Random sampling verification verifies the stability of the judgment result through multiple random sampling. The deviation threshold constraint sets the acceptable deviation range, and the data beyond the range needs to be re-evaluated to obtain a consistency indicator set.

[0042] For stability boundary calculation, the following formula is used: ; in, represents the stability boundary between the ith and jth weights, and Respectively represent the values ​​of the i-th and j-th weights in the k-th iteration, m represents the number of iterations, represents the correlation coefficient between weight pairs, Represents the time decay factor.

[0043] Starting from the consistency indicator set, the hierarchical structure is transitively verified to ensure the consistency of judgments between upper and lower levels. The inter-layer relationship constraints stipulate the association rules between different levels, and the hierarchical judgment analysis verifies the hierarchical nature of the judgment, and finally forms a transitivity matrix. Based on the transitivity matrix, the weight data is interval-scaled and the judgment value is converted to a unified evaluation scale. The spatial projection mapping projects the multidimensional weight space into the standardized space, and the normalization transformation ensures that the sum of the weights is 1, and a normalized vector is obtained.

[0044] For the normalized vector, stability boundary calculation is performed to determine the reasonable range of weight change. The fluctuation range constraint limits the change range of the weight, and the threshold calibration process adjusts the threshold parameter according to actual needs to obtain the stability parameter. Finally, the stability parameter is weighted and the weights of different levels are combined. The correlation between levels is considered in the process of level combination, and the coefficient optimization process ensures the rationality of the final weight to obtain the level weight coefficient.

[0045] For example, the evaluation indicators include physical fitness (including physical fitness test and physical fitness index), sports skills (including basic skills and special skills) and sportsmanship (including sports attitude and teamwork). First, the eigenvalue of the 3×3 judgment matrix is ​​calculated, and the maximum eigenvalue is 3.05. The main eigenvector is decomposed to obtain [0.42, 0.35, 0.23]. The calculated consistency ratio CR=0.048 is less than the standard threshold of 0.1, indicating that the judgment has acceptable consistency. Through 100 random sampling verifications, it was found that 95% of the sample CR values ​​were less than 0.1. The hierarchical structure was transitively verified to confirm that the judgment between sub-indicators was consistent with the main indicator judgment. The interval scale transformation converted the 1-9 scale to the 0-1 scale and normalized it. The stability analysis showed that the weight fluctuation range was within ±5%, and the hierarchical weight coefficients of physical fitness, sports skills and sportsmanship were finally determined to be 0.40, 0.37 and 0.23 respectively. The whole process maintained the consistency of judgment and the stability of weight.

[0046] In a specific embodiment, the process of executing step S105 may specifically include the following steps: (1) Perform multi-dimensional cross-calculation on the hierarchical weight coefficients, extract the hierarchical dependency relations, and reconstruct the evaluation space to obtain a comprehensive calculation matrix; (2) For the comprehensive calculation matrix, perform criterion layer grouping and aggregation, heterogeneous feature combination, and hierarchical performance synthesis to obtain a synthetic criterion space; (3) Based on the synthetic criterion space, the multi-level structure is subjected to interactive impact analysis, coupling quantification calculation, and bidirectional mapping transformation to obtain a mapping feature set; (4) Based on the mapping feature set, the evaluation data is subjected to fuzzy membership division, boundary fuzzy set decomposition, and interval feature extraction to obtain membership distribution; (5) Based on the membership distribution, hierarchical evaluation constraint processing, cross-influence intensity quantification and dynamic boundary correction are performed to obtain calibrated evaluation parameters; (6) Perform multi-dimensional feature synthesis, evaluation space reorganization and comprehensive performance integration on the calibration evaluation parameters to obtain a comprehensive evaluation result.

[0047] Specifically, multi-dimensional cross-calculations are performed on the hierarchical weight coefficients to calculate the cross-influence values ​​between different evaluation dimensions. Hierarchical dependency extraction analyzes the dependency strength between upper and lower indicators, and evaluation space reconstruction transforms the original evaluation space into a new spatial structure that is more suitable for comprehensive evaluation, thereby obtaining a comprehensive calculation matrix.

[0048] For the comprehensive calculation matrix, the criterion layer is grouped and aggregated to combine the evaluation indicators with strong correlation. In the process of combining different dimensional features, the features of different dimensions are optimized and combined to ensure the complementarity between the features. The hierarchical performance synthesis integrates the evaluation results of each group of indicators to form a synthetic criterion space. Starting from the synthetic criterion space, the multi-level structure is analyzed for interactive impact to identify the interaction relationship between indicators. The coupling quantification calculation quantitatively describes the correlation strength between indicators. The bidirectional mapping transformation establishes the corresponding relationship between the evaluation indicators and the evaluation results to obtain the mapping feature set.

[0049] Based on the mapping feature set, the evaluation data is divided into fuzzy membership, and the evaluation values ​​are mapped to membership spaces of different levels. Boundary fuzzy set decomposition processes the data in the fuzzy boundary area, and interval feature extraction extracts the feature information of each evaluation interval to form a membership distribution. For the membership distribution, hierarchical evaluation constraint processing is performed to ensure that the evaluation results meet the requirements of the hierarchical structure. The cross-influence intensity quantitatively analyzes the degree of cross-influence between different evaluation indicators, and the dynamic boundary correction adjusts the evaluation boundary according to the actual evaluation needs to obtain the calibrated evaluation parameters.

[0050] Finally, the calibration evaluation parameters are synthesized into multi-dimensional features, and the evaluation results of different dimensions are integrated. The evaluation space is reorganized to adjust the spatial structure of the evaluation results, and the comprehensive performance integration forms the final comprehensive evaluation results.

[0051] For example, the evaluation includes three dimensions: moral development (ideological and moral character 0.4, moral behavior 0.3, social morality 0.3), intellectual development (learning ability 0.35, knowledge mastery 0.35, innovative thinking 0.3) and physical development (physical quality 0.4, sports skills 0.3, sports spirit 0.3). Multidimensional cross calculation shows that the cross-influence coefficient between moral education and intellectual education is 0.65, moral education and physical education is 0.45, and intellectual education and physical education is 0.55. After grouping at the criterion level, the moral cultivation group (ideological and moral character, social morality) and the behavioral performance group (moral behavior) are formed. The coupling degree calculation shows that the coupling degree between moral cultivation and learning ability is 0.72. In the fuzzy membership division, the evaluation levels are divided into four levels: excellent, good, medium, and qualified, and the student membership in ideological and moral aspects is [0.6, 0.3, 0.1, 0] respectively. After hierarchical evaluation constraints and comprehensive integration, it was finally determined that the student's moral development was 88 points (excellent), intellectual development was 85 points (good), physical development was 82 points (good), and the comprehensive quality evaluation result was 85 points, reaching an excellent level.

[0052] In a specific embodiment, the process of executing step S106 may specifically include the following steps: (1) Construct a multi-criteria space, analyze the interaction effects of criteria, and divide the decision domain based on the comprehensive evaluation results to obtain a multi-criteria decision structure; (2) For the multi-criteria decision-making structure, development trajectory extraction, path node construction and correlation strength calculation are performed to obtain the development path map; (3) Based on the development path map, the feasibility constraint analysis, plan element decomposition and priority ranking of the development plan are carried out to obtain the initial plan set; (4) Based on the initial solution set, perform solution feature synthesis, multi-dimensional combination optimization and solution integration processing to obtain the solution synthesis result; (5) Based on the scheme synthesis results, multi-objective constraint definition, objective conflict resolution and optimization space construction are performed to obtain multi-objective analysis results; (6) Generate suggestions, match personalized features, and formulate development strategies based on the multi-objective analysis results to obtain personalized suggestions.

[0053] Specifically, when constructing a multi-criteria space for the comprehensive evaluation results, multiple evaluation criterion dimensions are established based on the evaluation results, including learning development, ability training, quality improvement, etc. Criteria interaction effect analysis explores the mutual influence and constraint relationship between different criteria. Decision domain division divides the entire decision space into several subdomains, each of which corresponds to a different development strategy direction, thereby obtaining a multi-criteria decision structure. For the multi-criteria decision structure, development trajectory extraction is performed to analyze the development laws and change trends in historical evaluation data. Path node construction determines the key turning points in the development process. Correlation strength calculation quantifies the degree of correlation between different development paths to form a complete development path map.

[0054] Starting from the development path map, the feasibility constraint analysis of the development plan is carried out, taking into account realistic constraints such as time and resources. The decomposition of the plan elements divides each development plan into specific implementation steps and task items. Priority sorting determines the execution order according to the importance and urgency of the plan to obtain the initial plan set. Based on the initial plan set, the plan feature synthesis is carried out to integrate similar or complementary plans. Multi-dimensional combination optimization optimizes the combination of plans from multiple dimensions, and the plan integration process ensures the coordination between the plans to form a plan synthesis result.

[0055] Based on the results of the scheme synthesis, multi-objective constraints are defined to clarify the specific requirements and restrictions of each development goal. Objective conflict resolution deals with the contradictions and competition between different objectives, and the optimization of spatial construction provides an operational spatial scope for the realization of the objectives, thus obtaining multi-objective analysis results.

[0056] Finally, recommendations are generated based on the results of the multi-objective analysis to form specific improvement measures. Personalized feature matching ensures that the recommendations are consistent with the individual characteristics of the evaluation object, and development strategy formulation plans the long-term development route, ultimately forming personalized recommendations.

[0057] For example, the comprehensive evaluation results show that the student's moral education score is 88 points, intellectual education score is 85 points, and physical education score is 82 points. The multi-criteria space construction divides the evaluation space into three dimensions: knowledge accumulation, ability improvement, and character cultivation. The development trajectory analysis found that the student's intellectual education score showed a stable upward trend (the first three monthly exams were 82, 84, and 85 points, respectively), while the physical education score fluctuated. The correlation strength calculation showed that the correlation coefficient between knowledge accumulation and ability improvement was 0.78, indicating that the two were highly correlated. In the process of program generation, key development directions were identified: strengthening physical training, participating in subject competitions, and cultivating innovative thinking. The feasibility analysis of these programs was carried out. Considering that the student had a strong interest in competitions, participating in subject competitions was listed as a priority development project. The multi-objective analysis showed that there was a time conflict between intellectual education improvement and physical development, and coordination was achieved through scientific planning of training time. The personalized suggestions finally formed included: participating in special physical training twice a week to cultivate a sports specialty; participating in 1-2 subject competition simulation trainings per month; using spare time for innovative thinking training and other specific measures. During the implementation of the recommendations, monthly periodic evaluations should be conducted to adjust development strategies in a timely manner.

[0058] The above describes the student development assessment method based on AHP in the embodiment of the present application. The following describes the student development assessment system based on AHP in the embodiment of the present application. Figure 2 In the embodiment of the present application, an embodiment of the student development assessment system based on AHP includes: Decomposition module 201, used to perform hierarchical decomposition of evaluation indicators, inter-layer coupling degree calculation and cross-impact analysis, and obtain an evaluation indicator system by combining dynamic reconstruction of the hierarchical structure and criterion mapping; The fusion module 202 is used to perform feature decomposition, dimension alignment and modality fusion on the evaluation data according to the evaluation index system, and perform standardization processing through data reconstruction and consistency mapping to obtain a hierarchical data set; The construction module 203 is used to construct elements, calculate relative importance and perform pairwise comparison correction on the judgment matrix based on the hierarchical data set, and optimize and dynamically adjust the matrix in combination with the consistency criterion to obtain a hierarchical analysis matrix; The analysis module 204 is used to perform characteristic root decomposition, consistency test and hierarchical transitivity analysis on the hierarchical analysis matrix, and normalize and optimize the stability of the weight vector to obtain the hierarchical weight coefficient; The calibration module 205 is used to perform hierarchical comprehensive calculation, criterion synthesis and multi-level evaluation mapping based on the hierarchical weight coefficient, and combine fuzzy membership analysis and evaluation calibration to obtain a comprehensive evaluation result; The synthesis module 206 is used to construct a multi-criteria decision-making model and a development path map based on the comprehensive evaluation results, and to perform solution synthesis and multi-objective optimization analysis to obtain personalized recommendations.

[0059] Through the collaborative cooperation of the above-mentioned components, a scientific and reasonable evaluation index system can be established by performing hierarchical decomposition of evaluation indicators, calculating inter-layer coupling and analyzing cross-impact, and realizing the accurate quantification of the relationship between evaluation indicators; by performing feature decomposition, dimension alignment and modal fusion on evaluation data, and combining with the standardized processing of data reconstruction and consistency mapping, the processing problem of multi-source heterogeneous data is effectively solved, and the quality and reliability of evaluation data are ensured; based on the hierarchical data set, the judgment matrix is ​​constructed with elements, relative importance is calculated and paired comparison is corrected, and the matrix is ​​optimized and dynamically adjusted in combination with the consistency criterion, making the evaluation process more objective and accurate; by performing characteristic root decomposition, consistency test and hierarchical transitivity analysis on the hierarchical analysis matrix, and normalizing and optimizing the stability of the weight vector, the scientificity and stability of the evaluation weight are guaranteed; based on the hierarchical weight coefficient, hierarchical comprehensive calculation, criterion synthesis and multi-level evaluation mapping are carried out, combined with fuzzy membership analysis and evaluation calibration, the accurate quantification of evaluation results is realized; finally, by constructing a multi-criteria decision-making model and a development path map, and performing scheme synthesis and multi-objective optimization analysis, personalized development suggestions are provided for students. The overall solution has established a complete and adaptive evaluation system, which not only realizes the normalization and standardization of the evaluation process, but also provides strong data support for educational decision-making through multi-dimensional data analysis and in-depth mining, effectively improving the scientificity, accuracy and practicality of student development assessment. The various technical features of the solution are closely related and mutually supportive, forming a complete technical solution that can effectively solve the problems of strong subjectivity, low accuracy, and insufficient personalization in traditional evaluation methods.

[0060] As described above, the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A student development assessment method based on AHP, characterized by: The student development assessment method based on AHP includes: The evaluation index system is obtained by performing hierarchical decomposition, calculating the inter-layer coupling degree and analyzing the cross-impact, and combining the dynamic reconstruction of the hierarchical structure and the criterion mapping. According to the evaluation index system, the evaluation data is subjected to feature decomposition, dimension alignment and modal fusion, and standardized through data reconstruction and consistency mapping to obtain a hierarchical data set; Based on the hierarchical data set, the judgment matrix is ​​constructed, the relative importance is calculated, and the pairwise comparison is corrected. The matrix is ​​optimized and dynamically adjusted in combination with the consistency criterion to obtain the hierarchical analysis matrix. For the hierarchical analysis matrix, characteristic root decomposition, consistency test and hierarchical transitivity analysis are performed, and the weight vector is normalized and stability optimized to obtain the hierarchical weight coefficient; Based on the hierarchical weight coefficient, hierarchical comprehensive calculation, criterion synthesis and multi-level evaluation mapping are carried out, and combined with fuzzy membership analysis and evaluation calibration, the comprehensive evaluation results are obtained; Based on the comprehensive evaluation results, a multi-criteria decision-making model and development path map are constructed, and scheme synthesis and multi-objective optimization analysis are carried out to obtain personalized recommendations.

2. The student development assessment method based on AHP according to claim 1 is characterized in that: The evaluation index system is obtained by performing hierarchical decomposition, inter-layer coupling degree calculation and cross-impact analysis on the evaluation index, and combining dynamic reconstruction of the hierarchical structure and criterion mapping, including: Performing dimensionality reduction processing on the hierarchical structure of the evaluation index through principal component analysis to obtain an initialization feature set; According to the initialization feature set, the interlayer relationship parameters are fitted and calculated using the least square method to obtain the interlayer coupling coefficient; The cross-influence intensity of the evaluation index is quantitatively analyzed by the interlayer coupling coefficient to obtain a cross-influence matrix; For the cross-influence matrix, dynamically decompose the hierarchical structure through singular value decomposition to obtain a reconstructed feature vector; Using the reconstructed feature vector, a hierarchical clustering algorithm is used to adaptively map the criterion layer to obtain a criterion mapping matrix; Based on the criterion mapping matrix, the evaluation index system is reconstructed and integrated through a hierarchical synthesis algorithm to obtain the evaluation index system.

3. The student development assessment method based on AHP according to claim 1 is characterized in that: The evaluation index system is used to perform feature decomposition, dimension alignment and modality fusion on the evaluation data, and standardize the data through data reconstruction and consistency mapping to obtain a hierarchical data set, including: Performing feature subspace division, information entropy weight calculation and dimensional feature extraction on the evaluation data to obtain feature subspace data; For the feature subspace data, collaboratively decompose and cross-dimensionally map the time series information of different dimensions to obtain an aligned dimension matrix; Based on the alignment dimension matrix, complementary information extraction, feature correlation analysis and heterogeneous information fusion are performed on the multi-type evaluation data to obtain fused feature data; Based on the fused feature data, the missing data points are spatially inferred, the neighborhood features are completed, and the structural consistency is verified to obtain a reconstructed data vector; Performing local preservation constraints, global consistency mapping and feature metric calibration on the reconstructed data vector to obtain a mapping data set; For the mapping data set, probability distribution calculation, data standardization transformation and hierarchical structure reorganization are performed to obtain the hierarchical data set.

4. The student development assessment method based on AHP according to claim 1 is characterized in that: Based on the hierarchical data set, the judgment matrix is ​​constructed with elements, relative importance is calculated, and pairwise comparison is corrected. The matrix is ​​optimized and dynamically adjusted in combination with the consistency criterion to obtain a hierarchical analysis matrix, including: Extracting paired indicators, dividing hierarchical intervals, and mapping indicator spaces on the hierarchical data set to obtain initial judgment matrix elements; For the elements of the initial judgment matrix, relative importance quantification, interval value mapping and judgment preference analysis are performed to obtain an importance feature matrix; Using the importance feature matrix, pairwise comparison data are subjected to consistency constraints, preference relationship corrections, and judgment conflict eliminations to obtain a correction judgment matrix; Based on the correction judgment matrix, index combination decomposition, matrix element optimization and hierarchical relationship reconstruction are performed to obtain an optimized matrix structure; Performing consistency constraint calculation, boundary threshold calibration and conflict elimination processing on the optimized matrix structure to obtain a dynamic adjustment matrix; For the dynamic adjustment matrix, matrix feature extraction, hierarchical structure optimization and element relationship reorganization are performed to obtain the hierarchical analysis matrix.

5. The student development assessment method based on AHP according to claim 1 is characterized in that: The hierarchical analysis matrix is ​​subjected to characteristic root decomposition, consistency test and hierarchical transitivity analysis, and the weight vector is normalized and stability optimized to obtain the hierarchical weight coefficient, including: Performing segmented eigenvalue extraction, main direction property calculation and eigenspace decomposition on the hierarchical analysis matrix to obtain an eigendecomposition vector; For the feature decomposition vector, consistency ratio calculation, random sampling verification and deviation threshold constraint are performed to obtain a consistency indicator set; Based on the consistency index set, the hierarchical structure is subjected to transitivity verification, inter-layer relationship constraints and hierarchical judgment analysis to obtain a transitivity matrix; Based on the transitivity matrix, performing interval scaling, spatial projection mapping and normalization transformation on the weight data to obtain a normalized vector; For the normalized vector, stability boundary calculation, fluctuation range constraint and threshold calibration processing are performed to obtain stability parameters; The stability parameters are subjected to weight synthesis, hierarchical combination and coefficient optimization processing to obtain the hierarchical weight coefficients.

6. The student development assessment method based on AHP according to claim 1 is characterized in that: Based on the hierarchical weight coefficient, hierarchical comprehensive calculation, criterion synthesis and multi-level evaluation mapping are performed, and combined with fuzzy membership analysis and evaluation calibration, a comprehensive evaluation result is obtained, including: Performing multi-dimensional cross calculation, hierarchical dependency extraction and evaluation space reconstruction on the hierarchical weight coefficients to obtain a comprehensive calculation matrix; For the comprehensive calculation matrix, grouping and aggregating criteria layers, combining different dimensional features and synthesizing hierarchical performance are performed to obtain a synthetic criteria space; Using the synthetic criterion space, interactive impact analysis, coupling quantization calculation and bidirectional mapping transformation are performed on the multi-level structure to obtain a mapping feature set; Based on the mapping feature set, fuzzy membership division, boundary fuzzy set decomposition and interval feature extraction are performed on the evaluation data to obtain membership distribution; For the membership distribution, hierarchical evaluation constraint processing, cross-influence intensity quantification and dynamic boundary correction are performed to obtain calibration evaluation parameters; The calibration evaluation parameters are subjected to multi-dimensional feature synthesis, evaluation space reorganization and comprehensive performance integration to obtain the comprehensive evaluation result.

7. The student development assessment method based on AHP according to claim 1 is characterized in that: Based on the comprehensive evaluation results, a multi-criteria decision-making model and a development path map are constructed, and program synthesis and multi-objective optimization analysis are performed to obtain personalized suggestions, including: The comprehensive evaluation results are subjected to multi-criteria space construction, criterion interaction effect analysis and decision domain division to obtain a multi-criteria decision structure; For the multi-criteria decision-making structure, development trajectory extraction, path node construction and correlation strength calculation are performed to obtain a development path map; Based on the development path map, feasibility constraint analysis, plan element decomposition and priority sorting are performed on the development plan to obtain an initial plan set; Based on the initial solution set, performing solution feature synthesis, multi-dimensional combination optimization and solution integration processing to obtain a solution synthesis result; Based on the synthesis results of the scheme, multi-objective constraint definition, objective conflict resolution and optimization space construction are performed to obtain multi-objective analysis results; Proposal generation, personalized feature matching and development strategy formulation are performed on the multi-objective analysis results to obtain the personalized suggestions.

8. A student development assessment method system based on AHP, used to implement the student development assessment method based on AHP as described in any one of claims 1 to 7, characterized in that: The student development assessment method system based on AHP includes: The decomposition module is used to perform hierarchical decomposition of evaluation indicators, inter-layer coupling calculation and cross-impact analysis, and obtain the evaluation indicator system by combining the dynamic reconstruction of the hierarchical structure and criterion mapping; The fusion module is used to perform feature decomposition, dimension alignment and modal fusion on the evaluation data according to the evaluation index system, and standardize the data through data reconstruction and consistency mapping to obtain a hierarchical data set; The construction module is used to construct elements, calculate relative importance and perform pairwise comparison correction on the judgment matrix based on the hierarchical data set, and optimize and dynamically adjust the matrix in combination with the consistency criterion to obtain the hierarchical analysis matrix; The analysis module is used to perform characteristic root decomposition, consistency test and hierarchical transitivity analysis on the hierarchical analysis matrix, and normalize and optimize the stability of the weight vector to obtain the hierarchical weight coefficient; The calibration module is used to perform hierarchical comprehensive calculation, criterion synthesis and multi-level evaluation mapping based on the hierarchical weight coefficient, and combine fuzzy membership analysis and evaluation calibration to obtain comprehensive evaluation results; The synthesis module is used to construct a multi-criteria decision-making model and development path map based on the comprehensive evaluation results, and to perform solution synthesis and multi-objective optimization analysis to obtain personalized recommendations.

Images