Questionnaire comprehensive evaluation model construction method and device, medium and product

Abstract

The invention discloses a questionnaire comprehensive evaluation model construction method and device, a medium and a product, and relates to the technical field of data analysis, and the method comprises the steps: obtaining questionnaire result data of a target questionnaire; according to the questionnaire result data, performing evaluation index extraction based on a principal component analysis method, and establishing a comprehensive index system; according to the comprehensive index system, evaluation index weighting is carried out based on an entropy weight method and an analytic hierarchy process, and an index comprehensive weight is obtained; and establishing a comprehensive evaluation model based on the comprehensive index system and the index comprehensive weight to determine a final evaluation result of the target questionnaire. According to the questionnaire comprehensive evaluation model construction method and device, the medium and the product provided by the invention, the difficulty and cost of evaluation analysis can be reduced, and the accuracy and effectiveness of an evaluation result are improved.

Description

Technical Field

[0001] The present application relates to the field of data analysis technology, and in particular to a method, device, medium and product for constructing a comprehensive evaluation model for a questionnaire. Background Art

[0002] In evaluation research, single subjective evaluation and objective evaluation can no longer explain the scientific nature of the evaluation method. The evaluation method that integrates subjective and objective is the mainstream evaluation method today. Although the questionnaire survey method can be applied to both qualitative and quantitative research, when using the questionnaire survey method as a means of data collection, although it is a widely used quantitative research method with advantages such as easy operation, low cost, and wide coverage, it also faces many challenges and problems.

[0003] Specifically, the questions in questionnaire design are often inevitably subjective, which is mainly due to the differences in the understanding, cognition and values ​​of the evaluation object by the question designer. This subjectivity may lead to biased expressions in the questionnaire questions, or fail to fully cover all key aspects of the evaluation object, thus affecting the authenticity and comprehensiveness of the data. In addition, the data obtained by the questionnaire survey method is often redundant, that is, the data volume is huge but the quality is uneven, and contains a large amount of non-critical, redundant or even wrong information. At the same time, due to the lack of systematic considerations in the design of the questionnaire, the number of indicators collected is too large and the correlation is not strong, making it difficult to use them directly for evaluation and analysis. In this case, if the evaluation is directly based on the questionnaire results, it will not only increase the difficulty and cost of analysis, but also reduce the accuracy and effectiveness of the evaluation results due to information overload or misleading information. Summary of the invention

[0004] The purpose of this application is to provide a method, equipment, medium and product for constructing a comprehensive evaluation model for a questionnaire, which can reduce the difficulty and cost of evaluation analysis and improve the accuracy and effectiveness of evaluation results.

[0005] To achieve the above objectives, this application provides the following solutions:

[0006] In a first aspect, the present application provides a method for constructing a comprehensive evaluation model of a questionnaire, comprising:

[0007] Obtain the questionnaire result data of the target questionnaire;

[0008] According to the questionnaire result data, the evaluation indexes are extracted based on the principal component analysis method to establish a comprehensive index system;

[0009] According to the comprehensive indicator system, the evaluation indicators are weighted based on the entropy weight method and the hierarchical analysis method to obtain the comprehensive weight of the indicators;

[0010] A comprehensive evaluation model is established based on the comprehensive indicator system and the comprehensive weights of the indicators to determine the final evaluation results of the target questionnaire.

[0011] Optionally, based on the questionnaire result data, evaluation indicators are extracted based on principal component analysis to establish a comprehensive indicator system, including:

[0012] Establishing an initial value matrix according to the questionnaire result data;

[0013] Screening and standardizing the initial value matrix to obtain a standardized matrix;

[0014] Perform correlation analysis according to the standardized matrix to obtain a correlation coefficient matrix;

[0015] Principal components are extracted based on the standardized matrix and the correlation coefficient matrix, and a comprehensive index system is established.

[0016] Optionally, principal component extraction is performed based on the standardized matrix and the correlation coefficient matrix to establish a comprehensive indicator system, including:

[0017] Calculate the principal component and the variance contribution rate of each principal component according to the standardized matrix and the correlation coefficient matrix;

[0018] Calculating the cumulative variance contribution rate according to the variance contribution rate of each principal component in an accumulative manner until the cumulative variance contribution rate reaches a set value, thereby determining the target principal component;

[0019] Factor analysis is performed on the target principal component to determine the evaluation index corresponding to the target principal component and establish a comprehensive index system.

[0020] Optionally, according to the comprehensive indicator system, the evaluation indicators are weighted based on the entropy weight method and the hierarchical analysis method to obtain the comprehensive weights of the indicators, including:

[0021] According to the comprehensive indicator system, the objective weights of the evaluation indicators are weighted based on the entropy weight method to obtain the objective weights of the indicators;

[0022] According to the comprehensive indicator system, subjective weighting of evaluation indicators is performed based on the analytic hierarchy process to obtain subjective weights of indicators;

[0023] The objective weights of the indicators and the subjective weights of the indicators are combined and weighted, and the comprehensive weights of the indicators are determined according to the combined weighting results.

[0024] Optionally, according to the comprehensive indicator system, the objective weights of the evaluation indicators are weighted based on the entropy weight method to obtain objective weights of the indicators, including:

[0025] Establishing a standardized judgment matrix based on the comprehensive indicator system and the questionnaire result data;

[0026] Calculate the information entropy value of the evaluation index according to the standardized judgment matrix;

[0027] The objective weight of the indicator is calculated according to the information entropy value of the evaluation indicator.

[0028] Optionally, according to the comprehensive indicator system, subjective weighting of evaluation indicators is performed based on the analytic hierarchy process to obtain subjective weights of indicators, including:

[0029] Determine the hierarchical structure of evaluation indicators according to the comprehensive indicator system;

[0030] Establishing a judgment matrix for each evaluation indicator at each level according to the hierarchical structure of the evaluation indicators;

[0031] Performing a consistency check on each of the judgment matrices to obtain a consistency check result;

[0032] If the consistency test result is that the test is passed, the subjective weight of the indicator is calculated according to the judgment matrix;

[0033] If the consistency check result is that the check fails, the step of determining the hierarchical structure of the evaluation indicators according to the comprehensive indicator system is returned to adjust the hierarchical structure of the evaluation indicators and the judgment matrix of each evaluation indicator at each level.

[0034] Optionally, a comprehensive evaluation model is established based on the comprehensive indicator system and the comprehensive weight of the indicators to determine the final evaluation result of the target questionnaire, including:

[0035] Establishing an evaluation factor set and a comment set according to the comprehensive indicator system; the evaluation factors in the evaluation factor set include: primary indicators, secondary indicators and tertiary indicators; the comment levels in the comment set include: very high, high, average and low;

[0036] Establishing a judgment matrix according to the evaluation factor set and the comment set; the judgment matrix reflects the fuzzy transformation relationship from the evaluation factor to the comment level;

[0037] Establishing a weight set according to the comprehensive weights of the indicators;

[0038] Establishing a comprehensive evaluation model based on the evaluation factor set, the comment set, the evaluation matrix and the weight set;

[0039] The final evaluation result of the target questionnaire is determined according to the comprehensive evaluation model.

[0040] In a second aspect, the present application provides a computer device, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the method for constructing a comprehensive evaluation model of a questionnaire.

[0041] In a third aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method for constructing a comprehensive evaluation model for a questionnaire.

[0042] In a fourth aspect, the present application provides a computer program product, including a computer program, which, when executed by a processor, implements the method for constructing a comprehensive evaluation model for a questionnaire.

[0043] According to the specific embodiments provided in this application, this application discloses the following technical effects:

[0044] The present application provides a method, equipment, medium and product for constructing a comprehensive evaluation model of a questionnaire. On the basis of the questionnaire survey method, the questionnaire result data and questions are subjected to dimensionality reduction processing by the principal component analysis method, and a comprehensive index system is established. The subjective and objective comprehensive evaluation method combining the entropy weight method and the analytic hierarchy process is used to comprehensively, impartially and accurately weight each evaluation index, obtain the comprehensive weight of the index, and finally establish a comprehensive evaluation model based on the comprehensive index system and the comprehensive weight of the index. The established comprehensive evaluation model can accurately and comprehensively extract key influencing factors from massive data, and take into account both subjective and objective aspects, can reduce the difficulty and cost of evaluation analysis, and improve the accuracy and effectiveness of the evaluation results, and has broad application prospects in data analysis, financial investment, medical services, agricultural construction and other aspects. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0046] Figure 1 A flowchart of the method for constructing a comprehensive evaluation model for the questionnaire provided in this application;

[0047] Figure 2 A diagram of the implementation steps of the method for constructing a comprehensive evaluation model of the questionnaire provided for this application;

[0048] Figure 3 Distribution chart of the evaluation results of the questionnaire on the impact of artificial intelligence on college students' learning provided for this application. DETAILED DESCRIPTION

[0049] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0050] In order to overcome the problems existing in the prior art, this application scientifically processes the questionnaire results on the basis of the questionnaire survey method, screens the evaluation indicators, and adopts a comprehensive evaluation method combining subjective and objective factors to weight the evaluation indicators, thereby obtaining a comprehensive evaluation model for the questionnaire, which can reduce the difficulty and cost of the evaluation analysis and improve the accuracy and effectiveness of the evaluation results.

[0051] In order to make the above-mentioned objects, features and advantages of the present application more obvious and easy to understand, the present application is further described in detail below in conjunction with the accompanying drawings and specific implementation methods.

[0052] In an exemplary embodiment, the present application provides a method for constructing a comprehensive evaluation model for a questionnaire, which is executed by a computer device, and can be executed by a computer device such as a terminal or a server alone, or by a terminal and a server together. In the embodiment of the present application, Figure 1 and Figure 2 As shown, the method includes the following steps S1 to S4.

[0053] Step S1: Obtaining questionnaire result data of a target questionnaire. The questionnaire result data includes a number of questionnaire samples, each of which includes a number of questions and corresponding scores.

[0054] Step S2: According to the questionnaire result data, the evaluation index is extracted based on the principal component analysis method to establish a comprehensive index system.

[0055] On the basis of the questionnaire survey method, in order to deal with the problems of too many similar questions and too many similar options in the questionnaire, avoid duplication and redundant processing of data analysis, and increase the scientificity and operability of indicators, it is proposed to use principal component analysis to reduce and optimize data dimension, and recombine multiple variables related to key issues into new comprehensive variables.

[0056] Step 2.1: Preliminary extraction of evaluation indicators.

[0057] Carefully read all the questions in the questionnaire, combine them with the problems to be solved, extract information based on the dimensions that each question focuses on, and refine the evaluation indicators that the question can reflect. Classify the questions and the preliminary refined evaluation indicators accordingly.

[0058] Step 2.2: Priority analysis of evaluation indicators.

[0059] Based on the problems that need to be solved, the priority analysis principle is set, and the preliminary refined evaluation indicators are prioritized and screened according to the principle. The priority can be divided into three levels, namely priority consideration, comprehensive consideration and later consideration. Questions that are not clearly related to the evaluation objectives are classified as later consideration indicators, questions that are directly related to the evaluation objectives are classified as priority consideration indicators, and questions that are not directly related to the evaluation objectives are classified as comprehensive consideration indicators.

[0060] Step 2.3: Scientific analysis of evaluation indicators.

[0061] The scientific analysis of the evaluation indicators is conducted from the perspectives of theoretical basis, data source, and representativeness of the evaluation indicators. Theoretical basis: The theoretical framework for selecting the evaluation indicators is to classify them from the perspective of subject and object, and analyze the logical relationship between the subject and object. Data source: Analyze whether the data on which the evaluation indicators rely is reliable, accurate, authoritative, and representative. Representativeness of evaluation indicators: Representativeness can be determined based on the first two steps.

[0062] Step 2.4: Analysis of the operability of evaluation indicators.

[0063] The questionnaire was coded and processed by data, including labeling and numerical processing. To consider the different degrees of impact, the scores were 5 for "very likely", 4 for "possible", 2 for "unclear", and 1 for "impossible".

[0064] Step 2.5: Principal component analysis.

[0065] Step 2.5.1: Data processing.

[0066] According to the questionnaire result data, an initial value matrix is ​​established. is the initial value of the jth index of sample i, and the initial value matrix established is as follows:

[0067]

[0068] Among them, X * is the initial value matrix, i=1,2,...,n, n is the total number of samples, j=1,2,...,p, p is the total number of indicators.

[0069] In a questionnaire, questions about the basic information of the survey subjects are generally asked first in order to understand the group composition of the survey subjects and verify the validity of the questionnaire. However, this type of information is generally not used as a variable in principal component analysis. Therefore, questions about the basic information of the survey need to be eliminated, and the remaining questions should be included in the scope of principal component analysis.

[0070] The initial value matrix is ​​screened and standardized to obtain a standardized matrix. The screened and standardized data indicators and the standardized matrix are as follows:

[0071]

[0072] X=(x ij ) n×p (3)

[0073] Among them, X is the standardized matrix, x ij To standardize the data index values ​​after screening, and are the sample mean and sample variance of the j-th indicator respectively.

[0074] Step 2.5.2: Correlation analysis.

[0075] Before selecting the principal components, we first need to analyze the correlation between the data indicators, giving priority to indicators with large correlation coefficients with other indicators. Such indicators can characterize the data information to the greatest extent.

[0076] According to the standardized matrix, correlation analysis is performed to obtain a correlation coefficient matrix. The correlation coefficient matrix between indicators is recorded as R and is calculated by the following formula:

[0077]

[0078] And calculate the eigenvalue of the correlation coefficient matrix and the regularized eigenvector, denoted as λ j ,e j , where j = 1, 2, ..., p.

[0079] Step 2.5.3: Principal component extraction.

[0080] After completing steps 2.5.1 and 2.5.2, use the obtained standardized matrix X and correlation coefficient matrix R to calculate the principal components:

[0081] Y j =Xe j (5)

[0082] Calculate the variance contribution of each principal component:

[0083]

[0084] Calculate the cumulative variance contribution rate based on the variance contribution rate of each principal component in an additive manner When α is large enough (reaches a certain set value), it is considered that the q principal components can reflect the information contained in all evaluation indicators, and the target principal component can be determined.

[0085] Step 2.5.4: Factor analysis.

[0086] When there are many research issues and the relationships are complex, multiple variables can be summarized into several common factors while retaining as much information as possible. After extracting the target principal component, each principal component can be screened and intuitively analyzed by calculating the factor loading matrix and common factor variance.

[0087] The elements in the factor loading matrix represent the degree of expression of each factor (i.e., principal component) for each variable. The higher the value, the stronger the correlation between the principal component and the variable. The factor loading matrix is ​​calculated according to the following formula:

[0088]

[0089] Among them, L is the factor loading matrix, q is the total number of principal components. According to the value of each column in the loading matrix, m factors are taken as common factors.

[0090] The common factor variance is the total degree of explanation of all principal components for a certain variable (i.e., indicator). The common factor variance is calculated according to the following formula:

[0091]

[0092] Among them, h j It means that all common factors can explain the variance of the jth indicator.

[0093] If the cumulative variance contribution rate of the principal component is small, the principal component is not representative. You can observe the common factor variance extraction value of each variable. Generally speaking, if the common factor variance is greater than 0.45, it means that the variable can be expressed. According to the processing results, remove the variables with low common factor variance extraction values, and perform correlation analysis on the new variables again. The total variance explained by the components shows which components are more important and which components are less important. When there are many factors to be processed, you can directly remove all factors with eigenvalues ​​less than 1 in the principal components.

[0094] After principal component analysis, if there are m components with initial eigenvalues ​​greater than 1, and the cumulative total variance explanation value is α, it means that m components can better explain all variables. Therefore, the next task is to analyze the meaning of these m components. Based on the factor loading matrix calculated above, the relationship between the principal components and the variables can be written in a table to form a component matrix. According to the component matrix, the variables belonging to components 1 to m are analyzed according to the scores, and the components corresponding to the largest variable values ​​are analyzed row by row to determine the evaluation index corresponding to the target principal component.

[0095] Step 2.6: Establishment of comprehensive indicator system.

[0096] According to the above evaluation index priority analysis and scientific analysis, a preliminary index pool is established. Then, the index pool is screened and selected through operability analysis to obtain a preliminary index system. Reliability analysis and correlation analysis are performed on the obtained index system, and indicators with weak correlation and poor credibility are eliminated from the data layer. Finally, the final evaluation index is obtained through principal component analysis, and a comprehensive index system is established.

[0097] Step S3: According to the comprehensive indicator system, the evaluation indicators are weighted based on the entropy weight method and the hierarchical analysis method to obtain the comprehensive weights of the indicators.

[0098] Step 3.1: Assign weights to the objectivity of evaluation indicators.

[0099] According to the weighting steps of the entropy weight method, it is implemented in Matlab, and the judgment matrix of each sample under each three-level indicator is input to obtain the weight of each evaluation indicator.

[0100] The entropy weight method is an objective evaluation index weight method, which deals with decision-making problems from the perspective of information entropy. The entropy weight method is a synthesis based on information theory and set theory, and is applicable to fields such as multi-attribute evaluation. Its main idea is to integrate multiple indicators into an overall evaluation indicator by calculating the entropy value and weight of each indicator, thereby providing support for decision-making. The entropy weight method can make full use of evaluation information, including the correlation between indicators, to more accurately evaluate the comprehensive performance of each evaluation object; it can avoid errors caused by subjective judgment and empirical methods, thereby ensuring the objectivity and rationality of indicator weights. The steps to evaluate the weight of the indicator system using the entropy weight method are as follows:

[0101] Step 3.1.1: Establish a standardized judgment matrix.

[0102] There may be different units in the original data, so the data needs to be dimensionless. Generally, the optimal value of each column can be selected for processing. In order to reduce the impact of too large or too small data on the evaluation results, the dimensionless data results need to be normalized so that the data is in the range of [0,1]. Suppose the number of indicator dimensions for comprehensive evaluation is Z, the number of indicators is N, and the standardized matrix X' constructed by the standardized data is:

[0103] X'=(x' i'j' ) Z×N ,i'=1,2,…,Z; j'=1,2,…,N (9)

[0104] Step 3.1.2: Calculate the information entropy of each indicator:

[0105]

[0106] Step 3.1.3: Indicator weighting:

[0107]

[0108] Among them, H j' is the information entropy value of the j'th indicator, Q is the harmonic parameter, and f i'j' is the element characteristic weight of the j'th indicator under the i'th indicator dimension, v j' represents the objective weight of the j'th indicator, 1-H j' Represents the coefficient of variation of the j'th indicator.

[0109] After using the entropy weight method to assign weights based on the scores in the questionnaire, we can analyze the impact of various factors on the survey subjects. In some cases, the weights of sub-indicators under the same level of indicators may differ too much in value and not meet the actual requirements. Therefore, there is a certain deviation in the objective analysis using the entropy weight method alone, and it is also necessary to combine subjective weights to comprehensively evaluate the final score.

[0110] Step 3.2: Assign subjective weights to evaluation indicators.

[0111] The hierarchical structure of evaluation indicators is determined according to the comprehensive indicator system. Generally speaking, the hierarchical structure is divided into primary indicators, secondary indicators and tertiary indicators, among which the target layer is the final goal of the questionnaire. First, the judgment matrix of each evaluation indicator in each layer of the indicator system is established, and then the consistency of each judgment matrix is ​​tested. If the test result meets the above requirements, the judgment matrix passes the consistency test, indicating that the judgment matrix is ​​valid, thereby obtaining the subjective weight of each indicator.

[0112] The Analytic Hierarchy Process (AHP) is aimed at complex multi-objective decision-making problems. AHP can divide and decompose a problem into several levels layer by layer, and compare the different factors at each level in pairs, gradually extract the most critical criteria and factors, and make a more comprehensive and integrated decision analysis. In AHP, each decision maker can score according to his or her own subjective judgment, so that the differences and subjectivity of different people's opinions can be fully considered. At the same time, AHP also uses methods such as consistency ratio and consistency index to determine whether the decision maker has consistency bias and arbitrariness, so as to improve the accuracy of the decision results. The basic steps of using the analytic hierarchy process to solve multi-objective problems are as follows:

[0113] Step 3.2.1: Hierarchical analysis.

[0114] Split a complex problem into several parts, and regard each part as a hierarchical structure, dividing it into several levels from top to bottom. As the levels go down, the problem becomes more detailed. There is a clear cause-and-effect relationship between each level, and decision makers can analyze problems and make decisions from top to bottom.

[0115] Step 3.2.2: Construction of judgment matrices at each level.

[0116] AHP arranges and combines multiple factors under multiple criteria in a pairwise comparison manner, and quantifies them to construct a pairwise comparison matrix. Each matrix is ​​equivalent to a quantitative description of the level of importance of a certain aspect to other aspects. The nine-level scaling method is usually used to assign values ​​to the elements of the judgment matrix. The relative importance relationship between indicators can be found in Table 1.

[0117] Table 1 Relative importance relationship between indicators

[0118]

[0119]

[0120] Step 3.2.3: Consistency check.

[0121] Due to the complexity of multi-order judgments, some values ​​in the judgment matrix may be inconsistent. The consistency test is used to determine whether the decision maker's choice in the pairwise comparison matrix is ​​reasonable. If the test result does not meet the requirements, it means that the decision maker's estimate of the weight of the factor is biased and needs to be readjusted. The following formula can be used for consistency test:

[0122]

[0123] Among them, l represents the order of the judgment matrix, λ max Indicates the maximum eigenvalue, CI indicates the degree of inconsistency of the judgment matrix, and RI indicates the average random consistency test. Determine whether the CR value is less than 0.10. If it is less than 0.10, it means that the judgment matrix has a certain consistency. If the CR value is greater than or equal to 0.10, it is necessary to check whether the hierarchy determined by the matrix is ​​too complex, or re-collect the information of experts on the judgment matrix to obtain more accurate data.

[0124] Step 3.2.4: Determination of hierarchical weights.

[0125] By using the arithmetic mean method to calculate the judgment matrix, we can get the weight of judging the importance of each factor, thereby determining the main factors affecting the decision.

[0126]

[0127] Among them, wj' is the subjective weight of the j'th indicator, u q'p' It is the value of the q'th row and p'th column in the judgment matrix, and the row number q' corresponds to the index number j'.

[0128] The subjective weights obtained by establishing the judgment matrix of each indicator through expert experience reduce the difference in unreasonable weights between indicators compared with the weights under the entropy weight method. It can be obviously seen that the experts' scoring of the influencing factors in the questionnaire and the focus of the questionnaire scores are different. If the weights under subjectivity and objectivity are combined, more reasonable and scientific indicator weights can be obtained.

[0129] Step 3.3: Comprehensive weight of evaluation indicators.

[0130] Compared with the analytic hierarchy process, the entropy weight method has the advantages of not needing to establish a hierarchical structure and construct a judgment matrix, being relatively objective, relatively simple to calculate, and requiring little data information. The disadvantage is that it has high requirements for data quality. If there is a large deviation, it will seriously affect the results of weight evaluation. The advantage of the analytic hierarchy process is that it can effectively solve complex decision-making problems, can more intuitively consider the impact of multiple factors on decision-making, and can eliminate the influence of subjective factors by integrating expert opinions. However, the analytic hierarchy process also has its disadvantages, mainly manifested in the need for greater subjective judgment and human preference when constructing a hierarchical structure and a judgment matrix, which may cause the evaluation results to be far from the actual situation. In addition, the evaluation results of different experts may differ, making it difficult to reach a consensus.

[0131] In view of the shortcomings of a single evaluation method, this application proposes to combine the entropy weight method and the hierarchical analysis method to improve the accuracy of the weight evaluation results. The combined weighting results are:

[0132] S j' =αw j' +βv j' (17)

[0133] Among them, S j' is the final weight of the j'th indicator, α and β are the subjective weight and objective weight coefficients, which are determined according to experience.

[0134] By assigning weights to the indicator weight combinations and weighted summing up the indicators at each level, we can get the comprehensive weight of each third-level indicator to the target layer. Finally, by comparing the weights of the three-level indicator combinations, we can analyze the core information reflected in the questionnaire from these high-weighted indicators.

[0135] Step S4: Establish a comprehensive evaluation model based on the comprehensive indicator system and the comprehensive weights of the indicators to determine the final evaluation results of the target questionnaire.

[0136] In view of the impact of artificial intelligence on college students' learning, this application adopts the entropy weight-AHP analysis method, combines the questions in the questionnaire, analyzes the impact of each indicator on it, and uses a weighted model to comprehensively calculate the impact value of each evaluation object. The general steps of comprehensive evaluation are: determine the object to be evaluated; establish an evaluation index system; calculate the weight coefficient corresponding to each indicator; construct a comprehensive evaluation model; and sort the comprehensive evaluation values.

[0137] The mathematical model of fuzzy comprehensive evaluation, that is, the comprehensive evaluation model, consists of four aspects (C, V, R, W), where C is the evaluation factor set, V is the comment set, R is the evaluation matrix, and W is the weight set. The evaluation matrix R is actually a fuzzy relationship matrix from C to V, which represents a fuzzy transformation relationship from evaluation factors to comment levels. According to the fuzzy transformation principle, the weight vector W and the fuzzy relationship matrix R are compounded to obtain the comprehensive evaluation result S.

[0138] The steps of fuzzy comprehensive evaluation algorithm are as follows:

[0139] Step 4.1: Establish the evaluation factor set, comment set, value set and weight set.

[0140] 1) The set of evaluation factors obtained according to the evaluation index system is as follows:

[0141] First-level indicators: A, B, C.

[0142] Secondary indicators: A = {A1, A2, A3, A4}; B = {B1, B2, B3, B4}; C = {C1, C2, C3, C4}.

[0143] Level 3 indicator: A1 = {A 11 ,A 12},A2={A 21 ,A 22},A3={A 31 ,A 32 ,A 33 ,A 34 ,A 35 ,A 36},A4={A 41 ,A 42 ,A 43}; B1 = {B 11 ,B 12 ,B 13 ,B 14},B2={B 21 ,B 22 ,B 23},B3={B 31 ,B 32 ,B 33},B4={B 41,B 42 ,B 43}; C1 = {C 11 ,C 12 ,C 13},C2={C 21 ,C 22}.

[0144] Taking the questionnaire on the impact of artificial intelligence on college students' learning as an example, the comprehensive evaluation index system established is shown in Table 2.

[0145] Table 2 Comprehensive evaluation index system

[0146]

[0147]

[0148]

[0149] 2) Construct comment sets and value sets.

[0150] V={V1,V2,V3,V4}={very high, high, average, low}, the numerical set is the corresponding score of each indicator after secondary encoding, and the corresponding comment set scores can be found in Table 3.

[0151] Table 3 Comment set scores

[0152] Score range Comments 80-100 points Very high 65-80 points high 50-65 minutes generally 0~50 points Low

[0153] 3) Establish a target allocation weight set.

[0154] In step S3, the comprehensive weight of each indicator in the evaluation system may be calculated, which will not be described in detail here.

[0155] Step 4.2: Construct membership subset R e , R e =(r e1 ,r e2 ,r e3 ,r e4 ), where R e Indicates the hth evaluation level V of the comment set corresponding to the eth indicator in the evaluation factor h The membership degree of

[0156] Based on the above steps, the final evaluation result is obtained. The embodiment of this application finally obtains the degree of influence of artificial intelligence on each college student participating in the questionnaire. According to the score division of the comment set, the number of people under different comments is obtained, see Figure 3 and Table 4.

[0157] Table 4 Number of people with different comments (total number: 4244)

[0158]

[0159] Among them, there are 610 students with scores below 50, accounting for 14%. The largest number of students scored 60-70, accounting for 28%. There are only 97 students with scores above 90, accounting for 2%. In general, this score situation shows that artificial intelligence technology has a high impact on the learning of only a very small number of people, and most of them are at a generally high level, indicating that the current college students' learning is greatly affected by the development of artificial intelligence, and the positive impact is relatively large. At the same time, the proportion of students with scores below 60 reached 36%, indicating that the current technical bottlenecks of artificial intelligence have a significant negative impact on college students' learning. There is still a lot of room for improvement in the positive effect of college students' learning.

[0160] This application is based on the questionnaire survey method. The questionnaire result data and questions are processed by principal component analysis to reduce the dimension and extract the principal component index. After constructing the evaluation index, the subjective and objective comprehensive evaluation method combined with entropy weight-AHP is adopted to fully, impartially and accurately weight each evaluation index. The comprehensive evaluation model established can accurately and comprehensively extract key influencing factors from massive data, and take into account both subjective and objective aspects, that is, on the basis of factual data, experts' judgment and tendency on social development trends (such as artificial intelligence) are also considered, providing decision makers with strong data support and decision-making basis, and has broad application prospects in data analysis, financial investment, medical services, agricultural construction, etc. Among them, the advantage of principal component analysis in processing questionnaire data is that it simplifies the complexity of data, reduces the dimension of data, and makes subsequent data processing and analysis more efficient. The comprehensive variables extracted by principal component analysis can better represent the essential characteristics of the original data, are more intuitive and concise in interpretation, and help researchers better understand the relationship and law between data. The advantage of the subjective and objective comprehensive evaluation method combining entropy weight and AHP is that it improves the credibility, comprehensiveness and complementarity of the indicator weight basis. The entropy weight method and AHP have different focuses in determining the weight. The entropy weight method focuses on the objective laws of data, while AHP emphasizes the subjective judgment of experts. The combination of the two can complement each other and make the weight distribution more reasonable and accurate.

[0161] In an exemplary embodiment, the present application further provides a computer device, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the steps in the above-mentioned method embodiments when executing the computer program.

[0162] In an exemplary embodiment, the present application further provides a computer-readable storage medium storing a computer program, which implements the steps in the above-mentioned method embodiments when executed by a processor.

[0163] In an exemplary embodiment, the present application further provides a computer program product, including a computer program, which implements the steps in the above-mentioned method embodiments when executed by a processor.

[0164] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant legal provisions.

[0165] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM may be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM).

[0166] The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. The non-relational database may include a distributed database based on blockchain, etc., but is not limited thereto. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., but is not limited thereto.

[0167] The technical features of the above embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0168] This article uses specific examples to illustrate the principles and implementation methods of this application. The description of the above embodiments is only used to help understand the method and core ideas of this application. At the same time, for those skilled in the art, according to the ideas of this application, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting this application.

Claims

1. A method for constructing a comprehensive evaluation model of a questionnaire, characterized in that: include: Obtain the questionnaire result data of the target questionnaire; According to the questionnaire result data, the evaluation indexes are extracted based on the principal component analysis method to establish a comprehensive index system; According to the comprehensive indicator system, the evaluation indicators are weighted based on the entropy weight method and the hierarchical analysis method to obtain the comprehensive weight of the indicators; A comprehensive evaluation model is established based on the comprehensive indicator system and the comprehensive weights of the indicators to determine the final evaluation results of the target questionnaire.

2. The method for constructing a comprehensive evaluation model of a questionnaire according to claim 1, characterized in that: According to the questionnaire results, the evaluation indicators are extracted based on the principal component analysis method, and a comprehensive indicator system is established, including: Establishing an initial value matrix according to the questionnaire result data; Screening and standardizing the initial value matrix to obtain a standardized matrix; Perform correlation analysis according to the standardized matrix to obtain a correlation coefficient matrix; Principal components are extracted based on the standardized matrix and the correlation coefficient matrix, and a comprehensive index system is established.

3. The method for constructing a comprehensive evaluation model of a questionnaire according to claim 2, characterized in that: Based on the standardized matrix and the correlation coefficient matrix, principal component extraction is performed to establish a comprehensive indicator system, including: Calculate the principal component and the variance contribution rate of each principal component according to the standardized matrix and the correlation coefficient matrix; Calculating the cumulative variance contribution rate according to the variance contribution rate of each principal component in an accumulative manner until the cumulative variance contribution rate reaches a set value, thereby determining the target principal component; Factor analysis is performed on the target principal component to determine the evaluation index corresponding to the target principal component and establish a comprehensive index system.

4. The method for constructing a comprehensive evaluation model of a questionnaire according to claim 1, characterized in that: According to the comprehensive indicator system, the evaluation indicators are weighted based on the entropy weight method and the hierarchical analysis method to obtain the comprehensive weights of the indicators, including: According to the comprehensive indicator system, the objective weights of the evaluation indicators are weighted based on the entropy weight method to obtain the objective weights of the indicators; According to the comprehensive indicator system, subjective weighting of evaluation indicators is performed based on the analytic hierarchy process to obtain subjective weights of indicators; The objective weights of the indicators and the subjective weights of the indicators are combined and weighted, and the comprehensive weights of the indicators are determined according to the combined weighting results.

5. The method for constructing a comprehensive evaluation model of a questionnaire according to claim 4, characterized in that: According to the comprehensive indicator system, the objective weights of the evaluation indicators are weighted based on the entropy weight method, and the objective weights of the indicators are obtained, including: Establishing a standardized judgment matrix based on the comprehensive indicator system and the questionnaire result data; Calculate the information entropy value of the evaluation index according to the standardized judgment matrix; The objective weight of the indicator is calculated according to the information entropy value of the evaluation indicator.

6. The method for constructing a comprehensive evaluation model of a questionnaire according to claim 4, characterized in that: According to the comprehensive indicator system, the subjective weighting of the evaluation indicators is carried out based on the hierarchical analysis method to obtain the subjective weights of the indicators, including: Determine the hierarchical structure of evaluation indicators according to the comprehensive indicator system; Establishing a judgment matrix for each evaluation indicator at each level according to the hierarchical structure of the evaluation indicators; Performing a consistency check on each of the judgment matrices to obtain a consistency check result; If the consistency test result is that the test is passed, the subjective weight of the indicator is calculated according to the judgment matrix; If the consistency check result is that the check fails, the step of determining the hierarchical structure of the evaluation indicators according to the comprehensive indicator system is returned to adjust the hierarchical structure of the evaluation indicators and the judgment matrix of each evaluation indicator at each level.

7. The method for constructing a comprehensive evaluation model of a questionnaire according to claim 1, characterized in that: A comprehensive evaluation model is established based on the comprehensive indicator system and the comprehensive weights of the indicators to determine the final evaluation results of the target questionnaire, including: Establishing an evaluation factor set and a comment set according to the comprehensive indicator system; the evaluation factors in the evaluation factor set include: primary indicators, secondary indicators and tertiary indicators; the comment levels in the comment set include: very high, high, average and low; Establishing a judgment matrix according to the evaluation factor set and the comment set; the judgment matrix reflects the fuzzy transformation relationship from the evaluation factor to the comment level; Establishing a weight set according to the comprehensive weights of the indicators; Establishing a comprehensive evaluation model based on the evaluation factor set, the comment set, the evaluation matrix and the weight set; The final evaluation result of the target questionnaire is determined according to the comprehensive evaluation model.

8. A computer device comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the method for constructing a comprehensive evaluation model for a questionnaire as described in any one of claims 1 to 7.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method for constructing a comprehensive evaluation model for a questionnaire described in any one of claims 1 to 7 is implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the method for constructing a comprehensive evaluation model for a questionnaire described in any one of claims 1 to 7 is implemented.

Images