Integrated method and system for evaluating sub-item development and comprehensive development of modern service industry

Abstract

The invention belongs to the technical field of information, and particularly relates to an integration method and system for evaluating sub-item development and comprehensive development of the modern service industry. The method comprises the following steps: identifying key factors of the modern service industry by adopting a global factor analysis method, solving scores of the sub-class service industries, and evaluating development levels of the eight sub-class service industries according to the scores; distributing weights for the key factors by adopting an entropy weight method; and obtaining a comprehensive score by using the key factor and the weight thereof and adopting a TOPSIS method, and evaluating the overall development level of the modern service industry according to the comprehensive score. Through integration factor analysis, the entropy weight method and the TOPSIS method, the evaluation accuracy is enhanced, the objectivity of the evaluation result is ensured, a more comprehensive and scientific modern service industry evaluation and decision support tool is provided, and the method can be widely applied to evaluation problems in different fields and complex scenes.

Description

Technical Field

[0001] The present invention belongs to the field of information technology, and in particular relates to an integrated method and system for evaluating the sub-item development and comprehensive development of modern service industry. Background Art

[0003] The modern service industry is subdivided into 8 major categories (01 information transmission, software and information technology services, 02 scientific research and technical services, 03 finance, 04 modern logistics services, 05 modern commercial services, 06 modern life services, 07 modern public services, 08 integrated development services), 39 middle categories and 139 small categories, showing the wide range of fields and sub-industries covered by the modern service industry. This detailed classification system can not only better reflect the actual situation of the modern service industry, but also provide an important basis for policy formulation, industry supervision and market analysis. Therefore, the traditional single evaluation method can no longer meet the needs of comprehensive evaluation. It is particularly important to establish a comprehensive and scientific evaluation system and evaluation method, which needs to be able to comprehensively consider the diversity within each sub-category and conduct macro-evaluation and comparison at the industry level.

[0004] Regarding the construction and research of the evaluation system, the existing methods are mainly divided into subjective methods and objective methods. Subjective methods include the analytic hierarchy process and the Delphi method. The analytic hierarchy process (AHP) simplifies and hierarchizes complex decision-making problems by constructing a hierarchical model, and determines the weight of each factor through pairwise comparison and consistency test. Its advantages are simple operation and suitability for multi-objective and multi-criteria decision-making problems. Its disadvantages are that it relies too much on expert judgment and is highly subjective. The Delphi method collects expert opinions through anonymous questionnaires, and gradually reaches a consensus after multiple rounds of feedback and adjustment. It is superior to the method of concentrated face-to-face discussion and can reduce the bias in group dynamics, but the method is time-consuming and depends on the choice and participation of experts.

[0005] Objective methods include factor analysis, entropy weight method, TOPSIS method, etc. Factor analysis is a statistical method used to study the correlation between variables. It simplifies the data structure by extracting key factors from the variables while retaining the information of the original data as much as possible. The main steps include: first, evaluate the applicability of the data, that is, conduct a correlation test; then, extract factors, commonly used methods include principal component analysis and principal axis factor analysis; next, perform factor rotation (such as right-angle rotation or oblique rotation) to improve the interpretability of the solution; finally, interpret the factors and determine the set of variables represented by each factor based on the factor loading. This process helps researchers identify the potential structure behind the variables.

[0006] The entropy weight method-TOPSIS method is a comprehensive evaluation method. First, the entropy weight method is used to objectively determine the weight of each evaluation index, and the information entropy principle is used to measure the discrete degree of the index, thereby reflecting the importance of the index; then, the TOPSIS method is used to calculate the relative closeness of each evaluation object based on the distance between it and the ideal optimal solution and the worst solution, so as to evaluate the comprehensive performance of the object. This method combines the objectivity of the entropy weight method and the intuitiveness of the TOPSIS method. It is suitable for multi-index comprehensive evaluation problems, can effectively overcome the deviation caused by subjective judgment, and improve the scientificity and accuracy of the evaluation.

[0007] Subjective evaluation system methods, such as the analytic hierarchy process (AHP) and the Delphi method, face a series of challenges despite their popularity in decision-making and evaluation processes. First, these methods rely heavily on the judgment and experience of experts, and the personal preferences and background differences among experts may lead to strong subjectivity in the evaluation results, thus affecting their objectivity. In addition, when faced with complex systems containing a large amount of data and variables, subjective evaluation methods may find it difficult to fully consider and evaluate all relevant factors, especially when the amount of information is large or the interactions between variables are complex. The consistency and repeatability of these methods are also a problem, because different experts or expert groups may come to very different conclusions. In addition, methods such as the Delphi method may require multiple rounds of expert consultation and feedback, which is time-consuming and may increase costs. Although dealing with fuzzy and difficult-to-quantify evaluation criteria is one of the advantages of these methods, it also makes it difficult to express the decision-making process and results in precise mathematical language, reducing the transparency and interpretability of the evaluation results. Finally, methods based on collective expert opinions may lead to the phenomenon of "group thinking", that is, reducing the rigor and diversity of the evaluation in order to reach a consensus. Therefore, although subjective evaluation methods are very useful in specific situations, their limitations suggest that the present invention needs to be combined with objective evaluation methods when applied to improve the overall effectiveness and reliability of the evaluation system.

[0008] Among objective methods, factor analysis is an effective data dimensionality reduction technique that can help researchers extract several representative factors from a large number of variables, but it has certain limitations in application. First, factor analysis requires a high correlation between data. If the variables in the original data are not highly correlated, the factors extracted by factor analysis may not be representative, thus affecting the accuracy and reliability of the analysis results. Second, when dealing with specific problems, factor analysis often focuses on the overall structural relationship and ignores the specific differences between individual variables or factors. This may not accurately reflect the uniqueness and differences of each subclass or variable when evaluating complex systems such as the modern service industry and its subclasses.

[0009] Especially when evaluating the modern service industry and its eight service subcategories, the challenges faced by using factor analysis alone are more significant. The complexity of the modern service industry lies in the fact that it covers a wide range of fields, and each subcategory contains many variables. The correlations between these variables may be different, and even show completely different characteristics in different subcategories. If we rely solely on factor analysis, although we can identify the main factors affecting the entire service industry from a macro perspective, it is difficult to deeply analyze the specific situation of each subcategory, and provide limited information for guiding specific industry development strategies and decisions. For example, factor analysis may be difficult to reveal the subtle differences in variables within a subcategory, or the unique development patterns between different subcategories, thereby limiting the pertinence and operability of the evaluation system in practical applications. In addition, when it is necessary to conduct in-depth evaluation and analysis of the service industry's market segments or specific service areas, a single factor analysis method may not provide sufficient resolution, and thus cannot meet the needs of refined management and decision-making.

[0010] When directly using the entropy weight method-TOPSIS method to construct the evaluation system of the modern service industry, although this method can provide a relatively objective evaluation framework, it also has some limitations. First, this method is mainly used to calculate the comprehensive score, which may not provide enough information for those who want to have a deep understanding of the development of each service subcategory and compare and analyze each other. Second, when dealing with many variables, the entropy weight method-TOPSIS method often assigns a relatively small weight to each variable, which to a certain extent reflects the average contribution of all variables to the comprehensive score and cannot distinguish the impact of different variables on the development of the modern service industry. Especially in a complex system such as the modern service industry that contains multiple subcategories and involves multiple evaluation indicators, more refined methods are needed to identify and measure the specific contributions of each subcategory and its internal variables to the overall development. Therefore, although the entropy weight method-TOPSIS method provides an initial framework for the evaluation of the modern service industry, it still needs to be combined with other methods or the existing methods need to be appropriately adjusted in practical applications to enhance the evaluation system's ability in subcategory analysis and variable influence distinction. Summary of the invention

[0011] In view of the defects of factor analysis and entropy weight method-TOPSIS method, the present invention proposes an integrated method for processing evaluation systems with many variables, aiming to build a comprehensive modern service industry evaluation system. This system is not only intended to comprehensively measure the development status of modern service industry, but also to reveal the relationship between different service industry categories in various regions. First, considering the large number of variables involved and the strong correlation between variables within each service industry subcategory, the present invention adopts the global factor analysis method to reduce the dimension of the data, simplify the data structure and extract key factors, which effectively solves the problem of many variables. Then, based on the key factors and service industry subcategory scores obtained by factor analysis, the weights of each factor are objectively analyzed by using the entropy weight method. Finally, the TOPSIS method is used to calculate the comprehensive score of modern service industry in each region to ensure the objectivity and accuracy of the evaluation results.

[0012] The advantage of adopting this integrated method is that it combines the data dimensionality reduction capability of factor analysis and the comprehensive evaluation capability of entropy weight method-TOPSIS, thus effectively overcoming the limitations of each method. Through factor analysis, the key factors affecting the development of modern service industry can be effectively identified, the evaluation system can be simplified, and a solid foundation can be laid for the subsequent weight distribution and comprehensive scoring. Subsequently, the introduction of entropy weight method increases the objectivity of the evaluation process and ensures the reasonable distribution of weights; while the application of TOPSIS method further improves the accuracy and reliability of the evaluation results. This integrated method not only overcomes the shortcomings of a single method, but also enhances the comprehensiveness and scientificity of the evaluation system, providing a more objective and accurate analysis tool for the evaluation and development of modern service industry.

[0013] The technical solution adopted by the present invention is as follows:

[0014] An integrated method for evaluating the sub-item and comprehensive development of modern service industries, including the following steps:

[0015] The global factor analysis method is used to identify the key factors of the eight subcategories of modern service industry and obtain the scores of the subcategories of service industry. The development level of the eight subcategories of service industry is evaluated based on the scores.

[0016] The entropy weight method is used to assign weights to key factors;

[0017] Using key factors and their weights, the TOPSIS method was used to obtain a comprehensive score, and the overall development level of the modern service industry was evaluated based on the comprehensive score.

[0018] Further, the following steps are used to identify the key factors:

[0019] 1) Assuming that the data of m evaluation indicators of n evaluation objects in T years are considered, that is, a three-dimensional matrix of n×m×T, the data under global factor analysis is expressed as Tn ×m horizontal matrix Where T n =T×n,x ij ′ is the element in row i and column j in X′;

[0020] 2) Standardize the data to obtain a standardized matrix The positive and negative indicator data are standardized respectively:

[0021] j is a positive indicator

[0022] j is a negative indicator

[0023] in, It means that for the jth evaluation index, the x corresponding to all evaluation objects i ij ′, represents the corresponding maximum value;

[0024] 3) Performing a correlation test on the standardized data. If the test is passed, it indicates that there are common factors between the variables in the data set, which is suitable for factor analysis. The correlation test includes KMO test and Bartlett sphericity test. If the data has a higher KMO value or passes the Bartlett sphericity test, it indicates that there are common factors between the variables in the data set, which is suitable for factor analysis.

[0025] 4) Perform eigenvalue decomposition on the sample covariance matrix S and obtain p eigenvalues ​​λ1≥λ2≥…≥λ p ≥0, the corresponding eigenvectors are γ1, γ2, ..., γ p , take the eigenvectors of the first p largest eigenvalues ​​to estimate the factor loading matrix The parameter p is determined by the cumulative variance contribution rate of the common factors, that is:

[0026]

[0027] Among them, r represents the threshold ratio of the cumulative explained variance; when the cumulative variance contribution rate of the current p common factors exceeds 80%, it is considered that their linear combination can restore the original variable information, and the p common factors are the key factors finally identified. Furthermore, based on the p key factors, the potential factor score matrix F, and the variance contribution rate of each potential factor The calculation score of the subcategory service industry is:

[0028] Q i =h1F i1 +h2F i2 +…+h p Fip

[0029] Where Q i is the service subcategory score of the ith research object, F ip is the score of the ith research subject on the kth factor.

[0030] Furthermore, the method of using the entropy weight method to assign weights to key factors includes:

[0031] 1) The i-th sample value x under the k-th key factor ik The proportion of this key factor is Where n is the number of evaluation objects, k = 1, 2, ..., p, then the entropy value and index difference of the kth factor are:

[0032]

[0033] D k =1-E k

[0034] 2) Calculate the weight w of the key factor k :

[0035]

[0036] Furthermore, the TOPSIS method is used to obtain the comprehensive score, including:

[0037] 1) Based on the comprehensive factor score Q obtained by factor analysis, the standardized matrix and weights are obtained according to the entropy weight method, and the weighted normalized matrix Z is established:

[0038]

[0039] Let z ik =w k F ik , where i = 1, ..., T n , k = 1, ..., p;

[0040] 2) Calculate the optimal solution and the worst solution, the optimal solution Z + The worst solution Z is composed of the maximum value in each column of Z. - Constructed from the minimum value in each column of Z:

[0041] 3) Calculate the Euclidean distance D between each solution and the optimal solution i + , and the Euclidean distance D between each solution and the worst solution i - :

[0042]

[0043] 4) Calculate the degree of proximity between each evaluation object and the optimal solution The degree of proximity C i It is a comprehensive score between 0 and 1. The larger the value, the stronger the development of the modern service industry.

[0044] Furthermore, the evaluation of the overall development level of the modern service industry based on the comprehensive score includes: evaluating the comprehensive development of the modern service industry in each region, and analyzing the development of the eight major sub-categories of the modern service industry in each region.

[0045] An integrated system for evaluating the sub-item and comprehensive development of modern service industries, including:

[0046] The key factor identification module is used to identify the key factors of the modern service industry using a global factor analysis method and obtain the scores of the sub-category service industry, and evaluate the development level of the eight sub-category service industries based on the scores;

[0047] The key factor weighting module is used to assign weights to key factors using the entropy weight method;

[0048] The comprehensive score calculation module is used to use key factors and their weights to obtain the comprehensive score using the TOPSIS method, and to evaluate the overall development level of the modern service industry based on the comprehensive score.

[0049] The beneficial effects of the present invention are as follows:

[0050] In terms of the construction of the evaluation system, the evaluation system constructed by the present invention has certain advantages in terms of coverage breadth, in-depth analysis of subdivided industries, comprehensive evaluation capabilities, and improved evaluation accuracy and objectivity. By carefully subdividing the service industry into 8 major categories, this system not only reveals the complexity and diversity of the service industry, but also provides the possibility of in-depth understanding of the development trends of each sub-industry in each region, surpassing the previous evaluation methods that only focus on a single service industry or dimension. The construction of this system allows for a detailed analysis of each sub-industry, while being able to capture the overall development trend of the service industry, solving the limitations of the existing evaluation system in considering both the sub-categories and the overall aspects of the service industry. In addition, by integrating factor analysis, entropy weight method and TOPSIS method, the present invention not only enhances the accuracy of the evaluation, but also ensures the objectivity of the evaluation results, providing a more comprehensive and scientific modern service industry evaluation and decision support tool. In summary, the evaluation system of the present invention demonstrates its superiority in multiple dimensions through its wide industry coverage, in-depth analysis of subdivided industries, and comprehensive analysis of the overall situation of the service industry, marking an important progress in the evaluation and decision-making methodology of the service industry.

[0051] In terms of methods, compared with the existing subjective methods, by integrating factor analysis, entropy weight method and TOPSIS method into a unified evaluation system methodology, the present invention can significantly improve the objectivity, accuracy, consistency and repeatability of the evaluation process, while also improving time and cost efficiency. This integrated methodology uses the statistical characteristics and information content of the data itself to determine the weights of the evaluation indicators, as well as the comprehensive evaluation, reducing the reliance on the subjective judgment of experts, thereby enhancing the objectivity of the evaluation results. It can effectively handle large-scale data sets and complex systems, ensure the comprehensiveness of the evaluation, and its results have high consistency and repeatability, and are not affected by the personal experience and preferences of experts. Compared with subjective evaluation methods that require long expert discussions and multiple rounds of feedback, integrated methodology is more time and cost efficient, especially when data processing and calculation processes can be automated. In addition, this methodology far exceeds the subjective evaluation method in terms of transparency and interpretability of the evaluation process and results, which helps to improve the acceptance and trust of the evaluation results. Its cross-domain application flexibility also shows that this methodology is not only suitable for the evaluation of modern service industries, but can also be flexibly adjusted and applied to multiple fields such as environmental assessment and socio-economic development analysis. In summary, the establishment of this integrated methodology provides significant advantages in providing a scientific and objective evaluation system construction plan, and compared with traditional subjective evaluation methods, it demonstrates its superiority in multiple dimensions.

[0052] The comprehensive evaluation system methodology formed by integrating factor analysis, entropy weight method and TOPSIS method has surpassed the limitations of simple application of factor analysis in multiple dimensions. First, this integrated methodology objectively determines the weight of each evaluation index through the entropy weight method, improves the scientificity and rationality of the evaluation process, and ensures the objectivity and accuracy of the evaluation results. Secondly, after integrating the TOPSIS method, the system can not only identify and simplify the key factors in the data, but also provide a complete decision support mechanism, which can directly sort the evaluation objects and clearly point out the optimal choice, thereby significantly enhancing the decision-making ability. In addition, the integrated methodology has outstanding performance in increasing the comprehensiveness and accuracy of the evaluation system. Each step from factor extraction, weight calculation to comprehensive scoring is based on clear steps and mathematical models, which greatly improves the interpretability and transparency of the evaluation process and facilitates understanding and acceptance. Finally, this methodology has a wide range of applications. It is not only suitable for the simplification of multivariate data and the extraction of key factors, but also for comprehensive evaluation and decision analysis, making it a scientific and comprehensive evaluation tool suitable for a variety of complex decision-making scenarios. In general, the establishment of this integrated methodology provides a more objective, accurate and comprehensive method for dealing with complex decisions, showing its significant advantages in practical applications.

[0053] Compared with the separate application of entropy weight method-TOPSIS analysis, the integrated method provides significant advantages in several key aspects. First, the addition of factor analysis greatly enhances the ability of data dimensionality reduction and simplification, effectively reduces information redundancy, and reveals the potential structure behind the data, providing a meaningful basis for subsequent weight determination and comprehensive evaluation. This not only improves the efficiency of the evaluation process, but also deepens the understanding of the evaluation object. Secondly, the integrated methodology is outstanding in improving the accuracy and objectivity of the evaluation, because it comprehensively considers the complexity and potential dimensions of the data, providing more accurate and objective input for entropy weighting and TOPSIS evaluation. In addition, by combining the advantages of the three methods, the methodology demonstrates excellent applicability and flexibility, enabling it to be widely used in evaluation problems in different fields and complex scenarios. Finally, the key information extracted by factor analysis in the early processing optimizes the quality of decision support information, allowing decision makers to make higher quality and efficient decisions based on more refined and key data. In summary, this integrated methodology, by making full use of the advantages and complementarity of each method, has not only achieved significant improvements in data processing, evaluation accuracy and objectivity, but also demonstrated obvious advantages in applicability and decision support information quality, becoming an efficient and comprehensive evaluation tool. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Figure 1 It is a flow chart of the steps of an integrated method for evaluating the sub-item development and comprehensive development of the modern service industry of the present invention.

[0055] Figure 2 It is a module composition diagram of an integrated system for evaluating the sub-item development and comprehensive development of the modern service industry of the present invention. DETAILED DESCRIPTION

[0056] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below through specific embodiments.

[0057] The technical solution proposed in this invention aims to establish a comprehensive modern service industry evaluation system by integrating factor analysis and entropy weight method-TOPSIS method. This system can comprehensively evaluate the development level of modern service industry and explore the interaction and relationship between different major categories of service industry in different regions. This solution is particularly suitable for processing data sets containing a large number of variables and strong correlation between variables within secondary indicators to ensure the comprehensiveness and accuracy of the evaluation. The specific technical solution is as follows Figure 1 As shown, it includes the following key steps:

[0058] 1. Global Factor Analysis to Identify Key Factors

[0059] Factor analysis is an analytical method for exploring multivariate data sets. Its core lies in discovering hidden structures or factors in the data and explaining the interactions between these factors and observed variables. In the context of a comprehensive analysis of the modern service industry, factor analysis is particularly appropriate, considering that this industry is widely classified, involves many variables, and there are often obvious relationships between variables in the same category. Through factor analysis, the present invention aims to screen out key factors from a large number of observed variables to streamline the data structure and reveal the inherent dimensions of the data. This not only helps to reveal the complex connections between variables, but also provides a solid starting point for in-depth research while reducing data complexity.

[0060] Although traditional factor analysis mainly deals with two-dimensional data sets, that is, data tables composed of variables and sample points, and does not directly consider the impact of time series, a challenge faced by the present invention is that if factor analysis is performed on each independent data table separately, differences in factor structures in different data tables may be found, which makes it difficult to effectively compare analysis results across years. In order to ensure the consistency and comparability of analysis results between different years, the present invention adopts a method of merging two-dimensional data sets of multiple years into a three-dimensional time series data set, and uses global factor analysis for processing. This method establishes a unified analysis framework through a comprehensive analysis of the overall data set, ensuring the consistency of analysis results at different time points. The steps of global factor analysis include:

[0061] 1. Construct the data matrix of global factor analysis: Assume that the data of m evaluation indicators of n evaluation objects in T years are considered, that is, a three-dimensional matrix of n×m×T. Then the data under global factor analysis can be expressed as T n ×m horizontal matrix Where T n =T×n. ij ′ is the element in row i and column j in X′.

[0062] 2. Data standardization: The indicators used in this invention are divided into positive and negative indicators, where a positive indicator means that an increase in its value has a positive effect on the evaluation object, while a negative indicator has the opposite effect. The positive and negative indicator data are standardized respectively:

[0063] j is a positive indicator

[0064] j is a negative indicator

[0065] in, It means that for the jth evaluation index, the x corresponding to all evaluation objects i ij ′, Indicates the corresponding maximum value.

[0066] 3. Perform correlation tests on the standardized data, namely KMO test and Bartlett sphericity test. If the data has a high KMO value (generally considered to be greater than 0.6 or higher) or passes the Bartlett sphericity test (p value less than 0.05), it indicates that there are common factors between the variables in the data set and it is suitable for factor analysis.

[0067] 4. Factor analysis: Perform eigenvalue decomposition on the sample covariance matrix S and obtain p eigenvalues ​​λ1≥λ2≥…≥λ p ≥0, the corresponding eigenvectors are γ1, γ2, ..., γ p . Take the eigenvectors of the first p largest eigenvalues ​​to estimate the factor loading matrix The parameter p is determined by the cumulative variance contribution rate of the common factors, that is:

[0068]

[0069] Among them, r represents the threshold proportion of cumulative explained variance. Specifically, r is a number between 0 and 1, which represents the proportion of total variance explained by the factors you want to select. By choosing an appropriate r value (such as 0.8 or 0.9), you can determine how many factors are needed to explain most of the variance in the data, thereby simplifying the model while retaining the main characteristics of the data.

[0070] In the present invention, when the cumulative variance contribution rate of the current p common factors exceeds 80%, it can be considered that their linear combination can basically restore the original variable information, and these p common factors are the "key factors" finally identified. Then the model of factor analysis is:

[0071]

[0072] Where F = (F1, F i , ..., Fp) T is the latent factor score matrix, which represents p independent common influencing factors in the original variables that cannot be directly observed but objectively exist. ε represents the error matrix, which includes the unique variance specific to the variable (also called idiosyncratic variance) and measurement error, that is, the part that the model cannot explain.

[0073] Based on the obtained p key factors, the potential factor score matrix F, and the variance contribution rate of each potential factor The subcategory service industry score Qi can be expressed as:

[0074] Q i =h1F i1 +h2Fi2 +…+h p F ip

[0075] Where Q i is the i-th (i=1, 2, ..., T n ) of the research objects’ service subcategory scores, F ip is the score of the ith research subject on the kth factor.

[0076] 2. Entropy Weight Method to Weight Key Factors

[0077] After mining and identifying the key factors through factor analysis, the present invention then uses the entropy weight method to accurately assign the weights of these key factors in the evaluation system, laying the foundation for the subsequent steps of comprehensive scoring. The entropy weight method, as an objective weight assignment method based on the principle of information entropy, is mainly used to quantify the contribution of each variable in providing information. Information entropy, an index that measures the degree of disorder of data, can reflect the distribution divergence of variable data and the validity of its information.

[0078] Specifically, if the information entropy of a variable is low, it means that it shows a high variability in the data set and can provide more effective information, so it should be given a higher weight. Conversely, variables with high information entropy have a relatively low weight in the entire evaluation system because of their low variability. The weight allocation by the entropy weight method ensures the objectivity and rationality of the weight allocation in the evaluation system, which lays a scientific weight foundation for calculating the comprehensive score and helps to accurately reveal the multidimensional characteristics of the development of the modern service industry and its comprehensive performance. The operation steps of the entropy weight method mainly include:

[0079] 1. Determine the factor entropy value: Since the i-th sample value x under the k-th (k=1, 2, ..., p) key factor ik The proportion of this key factor is Where n is the number of evaluation objects, then for the kth factor, its entropy value and index difference are

[0080]

[0081] D k =1-E k

[0082] 2. Calculate the weight W of the key factor k :

[0083]

[0084] 3. TOPSIS method to obtain comprehensive score

[0085] TOPSIS (Technique for Order of Preference by Similarity to IdealSolution), or ideal solution ranking method, is a widely recognized multi-criteria decision analysis tool. It is based on the core principle that among a series of alternative solutions, the best solution should be the one that is closest to the ideal optimal solution and farthest from the ideal worst solution. The ideal optimal solution refers to the hypothetical solution that performs best in all considered evaluation indicators, while the ideal worst solution is the hypothetical solution that performs worst in all evaluation indicators. The advantage of the TOPSIS method is that it provides an intuitive and effective way to compare the similarity and difference between each solution and the ideal state in a quantitative way, so as to determine the relative advantages and disadvantages of each solution. The execution steps of the TOPSIS method usually include:

[0086] 1. Establish a weighted normalized matrix: Based on the comprehensive factor score Q obtained by factor analysis, the standardized matrix and weights are obtained according to the entropy weight method, and the weighted normalized matrix Z is established:

[0087]

[0088] Order k =W k F ik , where i = 1, ..., T n , k=1,…,p.

[0089] 2. Calculate the optimal solution and the worst solution: the optimal solution Z + The worst solution Z is composed of the maximum value in each column of Z. - Constructed from the minimum value in each column of Z:

[0090]

[0091] 3. Calculate the Euclidean distances between each solution and the optimal solution and the worst solution:

[0092]

[0093] Among them, D i + represents the Euclidean distance between each solution and the optimal solution, D i - Represents the Euclidean distance between each solution and the worst solution.

[0094] Among them, each plan refers to a different entity or decision-making unit to be evaluated, specifically, the comprehensive situation of the service industry in each province, city or economic region.

[0095] 4. Calculate the degree of proximity between each evaluation object and the optimal solution The “degree of closeness” is a comprehensive score, which is between 0 and 1. The larger the value, the stronger the development of the modern service industry.

[0096] In summary, by integrating factor analysis, entropy weight method and TOPSIS method, the present invention successfully constructs a comprehensive methodological framework for modern service industry evaluation. The establishment of this integrated methodology not only optimizes the construction process of the evaluation system, but also significantly improves the accuracy and objectivity of the evaluation results. The main advantages of this integrated methodology are summarized as follows:

[0097] (1) Comprehensiveness and depth: The proposed method system is particularly applicable when the evaluation object covers a wide range and involves many variables. Factor analysis, as a data dimensionality reduction tool, effectively reveals the inherent structure and key factors of the data, providing a solid foundation for the evaluation system. This step ensures that the evaluation system can comprehensively cover and deeply reflect the core elements of the modern service industry.

[0098] (2) Objectivity and scientificity: By automatically determining the weight of each factor through the entropy weight method, the method of the present invention abandons the traditional subjective weighting method and improves the objectivity and scientificity of the evaluation system. This process ensures the rationality of the weight of each evaluation indicator, thereby enhancing the reliability of the evaluation results.

[0099] (3) Accuracy and practicality: The application of the TOPSIS method provides an intuitive and effective way to evaluate and rank the development level of modern service industries by comparing with the ideal solution. This distance-based decision analysis makes the evaluation results not only accurate but also easy to understand and implement, and has high practical value.

[0100] (4) Flexibility and adaptability: The design of this integrated methodology takes into account the diversity and complexity of modern service industry data, so it has strong flexibility and adaptability. It can be applied to the evaluation of modern service industries of different scales and types, providing customized solutions for decision makers.

[0101] (5) Facilitate decision making: Ultimately, this integrated methodology provides a powerful decision support tool for policy makers and managers by providing a clearly structured, scientifically sound evaluation system. It can help decision makers better identify the development trends of the modern service industry, formulate corresponding strategies and measures, and promote the sustainable and healthy development of the service industry.

[0102] In summary, the integrated methodology proposed by the present invention is not only innovative in theory, but also highly feasible in practice, providing a new idea and tool for the evaluation and management of modern service industry. The establishment of this comprehensive evaluation methodology marks an important step forward in the understanding and promotion of the development of modern service industry.

[0103] In addition, in addition to its application in the evaluation of modern service industries, the integrated methodology constructed by the present invention also has a wide range of cross-domain application potential. The integrated application of factor analysis, entropy weight method and TOPSIS method is not limited to the service industry. Its principles and framework are also applicable to the evaluation and decision analysis in multiple fields such as industry, environmental protection, and social development. This methodology can provide customized, scientific and reasonable evaluation and decision support for different industries or fields by appropriately adjusting the evaluation indicators and parameters. Its flexibility and adaptability ensure that the methodology can meet the evaluation needs of various complex situations, thus having high practical value and promotion potential. This additional advantage further expands the scope of application of the integrated methodology, proving that it is not only innovative in theoretical research, but also shows great flexibility and wide applicability in practical applications.

[0104] The key points of the present invention are:

[0105] 1. Since the new classification of modern service industry will not be released until the end of 2023 by the National Bureau of Statistics, there is no research on building an evaluation system for the development of modern service industry from this dimension. In addition, the existing evaluation systems are all single-dimensional, and only evaluate the development of a certain type of service industry. There is no methodology for analyzing the sub-categories and the overall service industry at the same time.

[0106] 2. The existing evaluation systems for modern service industries are not comprehensive, so they cover fewer variables and use a single method (only factor analysis, or only entropy weight-TOPSIS method), which is relatively simple.

[0107] 3. The evaluation system constructed by the integrated method of the present invention (first applying global factor analysis to extract key factors to obtain comprehensive factor scores, and then using the entropy weight-TOPSIS method based on the results to calculate the comprehensive score of the development level of modern service industry) can not only evaluate the comprehensive situation of the development of modern service industry in various regions, but also analyze the development of eight major modern service industry subcategories in various regions. Such an integrated method provides a solution for evaluation objects with a wide range and many variables.

[0108] Among them, the method for analyzing the development of eight modern service subcategories in each region is the factor analysis method. By conducting factor analysis on multiple indicators within each service category, the main factors were extracted, and their scores reflect the comprehensive development level of each subcategory of service industry. This step allows for accurate measurement of the performance of each service subcategory in each region, thereby identifying the strengths and weaknesses of each subcategory. The method for evaluating the comprehensive development of modern service industries in various regions is: entropy weight method combined with TOPSIS method. Based on the comprehensive factor scores obtained from the above factor analysis, the entropy weight method combined with TOPSIS method was further applied. The entropy weight method objectively determines the weights of each factor in the comprehensive evaluation system, while the TOPSIS method helps us calculate a comprehensive score representing the overall development level of the service industry based on these weights.

[0109] Another embodiment of the present invention provides an integrated system for evaluating the sub-item development and comprehensive development of modern service industries, such as Figure 2 As shown, it includes:

[0110] The key factor identification module is used to identify the key factors of the modern service industry using a global factor analysis method and obtain the scores of the sub-category service industry, and evaluate the development level of the eight sub-category service industries based on the scores;

[0111] The key factor weighting module is used to assign weights to key factors using the entropy weight method;

[0112] The comprehensive score calculation module is used to use key factors and their weights to obtain the comprehensive score using the TOPSIS method, and to evaluate the overall development level of the modern service industry based on the comprehensive score.

[0113] The specific implementation process of each module refers to the above description of the method of the present invention.

[0114] Another embodiment of the present invention provides a computer device (computer, server, smart phone, etc.), which includes a memory and a processor, the memory stores a computer program, the computer program is configured to be executed by the processor, and the computer program includes instructions for executing each step in the method of the present invention.

[0115] Another embodiment of the present invention provides a computer-readable storage medium (such as ROM / RAM, magnetic disk, optical disk), wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a computer, the steps of the method of the present invention are implemented.

[0116] The specific embodiments of the present invention disclosed above are intended to help understand the content of the present invention and implement it accordingly. It can be understood by those skilled in the art that various replacements, changes and modifications are possible without departing from the spirit and scope of the present invention. The present invention should not be limited to the contents disclosed in the embodiments of this specification, and the scope of protection of the present invention shall be subject to the scope defined in the claims.

Claims

1. An integrated method for evaluating the sub-item and comprehensive development of modern service industries, characterized in that: The following steps are involved: The global factor analysis method is used to identify the key factors of the eight subcategories of modern service industry and obtain the scores of the subcategories of service industry. The development level of the eight subcategories of service industry is evaluated based on the scores. The entropy weight method is used to assign weights to key factors; Using key factors and their weights, the TOPSIS method was used to obtain a comprehensive score, and the overall development level of the modern service industry was evaluated based on the comprehensive score.

2. The method according to claim 1, characterized in that: The following steps were used to identify the key factors: 1) Assuming that the data of m evaluation indicators of n evaluation objects in T years are considered, that is, a three-dimensional matrix of n×m×T, the data under global factor analysis is expressed as T n ×m horizontal matrix Where T n =T×n,x ij ′ is the element in row i and column j in X′; 2) Standardize the data to obtain a standardized matrix The positive and negative indicator data are standardized respectively: j is a positive indicator j is a negative indicator in, It means that for the jth evaluation index, the x corresponding to all evaluation objects i ij ′, represents the corresponding maximum value; 3) Performing a correlation test on the standardized data. If the test is passed, it indicates that there are common factors between the variables in the data set, which is suitable for factor analysis. The correlation test includes KMO test and Bartlett sphericity test. If the data has a higher KMO value or passes the Bartlett sphericity test, it indicates that there are common factors between the variables in the data set, which is suitable for factor analysis. 4) Perform eigenvalue decomposition on the sample covariance matrix S and obtain p eigenvalues ​​λ1≥λ2≥…≥λ p ≥0, the corresponding eigenvectors are γ1,γ2,…,γ p , take the eigenvectors of the first p largest eigenvalues ​​to estimate the factor loading matrix The parameter p is determined by the cumulative variance contribution rate of the common factors, that is: Among them, r represents the threshold ratio of the cumulative explained variance; when the cumulative variance contribution rate of the current p common factors exceeds 80%, it is considered that their linear combination can restore the original variable information, and the p common factors are the key factors finally identified.

3. The method according to claim 2, characterized in that Based on p key factors, the potential factor score matrix F, and the variance contribution rate of each potential factor Calculate the subcategory service industry score Q i for: Q i =h1F i1 +h2F i2 +…+h p F ip Where Q i is the service subcategory score of the ith research object, F ip is the score of the ith research subject on the kth factor.

4. The method according to claim 1, characterized in that: The entropy weight method is used to assign weights to key factors, including: 1) The i-th sample value x under the k-th key factor ik The proportion of this key factor is Where n is the number of evaluation objects, k = 1, 2, ..., p, then the entropy value and index difference of the kth factor are: D k =1-E k 2) Calculate the weight w of the key factor k :

5. The method according to claim 1, characterized in that The TOPSIS method is used to obtain a comprehensive score, including: 1) Based on the comprehensive factor score Q obtained by factor analysis, the standardized matrix and weights are obtained according to the entropy weight method, and the weighted normalized matrix Z is established: Let z ik =w k F ik , where i = 1,…,T n ,k=1,…,p; 2) Calculate the optimal solution and the worst solution, the optimal solution Z + The worst solution Z is composed of the maximum value in each column of Z. - Constructed from the minimum value in each column of Z: 3) Calculate the Euclidean distance D between each solution and the optimal solution i + , and the Euclidean distance D between each solution and the worst solution i - : 4) Calculate the degree of proximity between each evaluation object and the optimal solution The degree of proximity C i It is a comprehensive score between 0 and 1. The larger the value, the stronger the development of the modern service industry.

6. The method according to claim 1, characterized in that The evaluation of the overall development level of the modern service industry based on the comprehensive score includes: evaluating the comprehensive development of the modern service industry in various regions, and analyzing the development of the eight major sub-categories of the modern service industry in various regions.

7. An integrated system for evaluating the sub-item and comprehensive development of modern service industries, characterized in that: include: The key factor identification module is used to identify the key factors of the modern service industry using the global factor analysis method and obtain the scores of the sub-category service industry, and evaluate the development level of the eight sub-category service industries based on the scores; The key factor weighting module is used to assign weights to key factors using the entropy weight method; The comprehensive score calculation module is used to use key factors and their weights to obtain the comprehensive score using the TOPSIS method, and to evaluate the overall development level of the modern service industry based on the comprehensive score.

8. A computer device, characterized in that: The method comprises a memory and a processor, wherein the memory stores a computer program, the computer program is configured to be executed by the processor, and the computer program comprises instructions for executing the method according to any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a computer, the method according to any one of claims 1 to 6 is implemented.

Images