Method and system for calculating comprehensive evaluation index and processing data

Abstract

The invention belongs to the technical field of data processing, and particularly relates to a method and system for calculating a comprehensive evaluation index and processing data. The method comprises the following steps: importing original data of an index system needing to be processed, and identifying each index name, a research object and a time range in the original data; the method comprises the following steps: preprocessing original data, carrying out forward processing on all indexes, and carrying out normalization to eliminate dimension and scale differences among different indexes; determining the weight of each index based on the preprocessed data; calculating a comprehensive evaluation index through the weight and the preprocessed index value; and respectively drawing a first graph and a second graph according to the comprehensive evaluation index, the original data or the preprocessed data. The method comprises the steps of data import, automatic identification, data normalization, weight calculation, correlation analysis, drawing and the like, the working process of a user is greatly simplified, and the data processing efficiency is improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of data processing, and in particular relates to a method and system for calculating a comprehensive evaluation index and processing data. Background Art

[0002] In many industries, such as energy, finance, education, public policy, health, etc., the calculation of index systems has become necessary to provide insightful information about complex systems. These indexes provide managers, decision makers, and analysts with a basis for evaluating and optimizing operations, analyzing market trends, predicting outcomes, and making strategic decisions. However, calculating these indexes usually involves a lot of data collection, processing, analysis, and interpretation, which is a complex and time-consuming process.

[0003] Currently, there is no way to automatically calculate the index of the indicator system in the market. Usually, a large amount of data needs to be entered manually, which is not only time-consuming but also may lead to errors. In addition, it is difficult to manually process large amounts of high-dimensional and complex types of data, such as time series data and unstructured data. It usually requires researchers with advanced technical knowledge and experience to complete it more efficiently, which limits the practicality of manual processing of modern business problems. In addition, when analyzing data and drawing, general software only provides rough basic drawing services, or the steps of analyzing data and drawing are carried out separately, without providing one-step customization options. Summary of the invention

[0004] In view of the above problems, the present invention provides a method for calculating a comprehensive evaluation index and processing data, the method comprising:

[0005] Import the original data of the indicator system to be processed, and identify the name, research object, and time range of each indicator in the original data;

[0006] The raw data were preprocessed to make all indicators positive and normalized to eliminate the dimension and scale differences between different indicators;

[0007] Determine the weight of each indicator based on the preprocessed data;

[0008] Calculate the comprehensive evaluation index through weights and preprocessed indicator values;

[0009] The first graph and the second graph are drawn respectively according to the comprehensive evaluation index, the original data or the pre-processed data.

[0010] Preferably, the correlation between the indicators is analyzed based on the original data or the pre-processed data;

[0011] Draw the third graph based on the correlation analysis results.

[0012] Preferably, the first graph includes a total score plot and a horizontal bar graph;

[0013] The second graph includes one graph for each indicator and one graph for each indicator for each country;

[0014] The third graph includes a correlation scatter plot and a correlation coefficient matrix heat value plot.

[0015] Preferably, the preprocessing of the raw data includes: firstly performing positive transformation on the index, and then performing Z-score processing or Min-Max processing on the raw data.

[0016] Preferably, the Z-score processing includes: converting the original data into standard normal distribution data with a mean of 0 and a standard deviation of 1;

[0017] The Min-Max processing includes: scaling the value of each indicator in the original data to between 0 and 1.

[0018] Preferably, the weight determination methods include: principal component analysis method, entropy weight method, equal weight method, subjective plus principal component analysis method, subjective plus entropy weight method, subjective plus equal weight method and expert subjective weighting method.

[0019] The present invention also provides a system for calculating a comprehensive evaluation index and processing data, the system comprising an identification module, a preprocessing module, a weight determination module, an index calculation module and a drawing module;

[0020] The identification module is used to import the original data of the indicator system to be processed, and identify the name of each indicator, research object, and research time in the original data;

[0021] The preprocessing module is used to preprocess the raw data, normalize all indicators to eliminate the dimension and scale differences between different indicators;

[0022] The weight determination module is used to determine the weight of each indicator based on the preprocessed data;

[0023] The calculation index module is used to calculate the comprehensive evaluation index through the weight and the preprocessed index value;

[0024] The drawing module is used to draw a first graph and a second graph respectively according to the comprehensive evaluation index, the original data or the pre-processed data.

[0025] Preferably, the system further comprises a correlation analysis module;

[0026] The correlation analysis module is used to analyze the correlation between indicators based on original data or preprocessed data.

[0027] Preferably, the drawing module is used to draw the third graph according to the correlation analysis.

[0028] Preferably, the first graph includes a total score plot and a horizontal bar graph;

[0029] The second graph includes one graph for each indicator and one graph for each indicator for each country;

[0030] The third graph includes a correlation scatter plot and a correlation coefficient matrix heat value plot.

[0031] Preferably, the preprocessing module is used to preprocess the raw data, including:

[0032] The preprocessing module is used to first positively transform the index, and then perform Z-score processing or Min-Max processing on the original data.

[0033] Preferably, the weight determination methods include: principal component analysis method, entropy weight method, equal weight method, subjective plus principal component analysis method, subjective plus entropy weight method, subjective plus equal weight method and expert subjective weighting method.

[0034] The present invention has the following beneficial effects:

[0035] (1) The present invention includes data import, automatic identification, data normalization, weight calculation, correlation analysis, drawing, etc., which greatly simplifies the user's workflow and improves the efficiency of data processing;

[0036] (2) The present invention provides a variety of data normalization and weight calculation methods for users to choose from, so that the present invention can adapt to various data processing requirements;

[0037] (3) The present invention can automatically draw various graphs based on the processed data, which is convenient for users to intuitively view and understand the data, and also saves users the trouble of manual drawing;

[0038] (4) All steps of the present invention are strictly based on scientific principles such as operations research, statistics, and management, and can therefore provide accurate and reliable calculation results.

[0039] Other features and advantages of the present invention will be described in the following description, and partly become obvious from the description, or be understood by implementing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures pointed out in the description and the drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0041] Figure 1 A diagram showing a method for calculating a comprehensive evaluation index and processing data in an embodiment of the present invention;

[0042] Figure 2 A system diagram showing calculation of comprehensive evaluation index and processing of data in an embodiment of the present invention is shown. DETAILED DESCRIPTION

[0043] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in a variety of forms and should not be construed as being limited to the examples set forth herein; on the contrary, these embodiments are provided so that the present disclosure will be more comprehensive and complete, and the concepts of the example embodiments are fully conveyed to those skilled in the art. The described features, structures, or characteristics may be combined in one or more embodiments in any suitable manner. In the following description, many specific details are provided to provide a full understanding of the embodiments of the present disclosure. However, those skilled in the art will appreciate that the technical solutions of the present disclosure may be practiced while omitting one or more of the specific details, or other methods, components, devices, steps, etc. may be adopted. In other cases, known technical solutions are not shown or described in detail to avoid obscuring various aspects of the present disclosure.

[0044] In addition, the accompanying drawings are only schematic illustrations of the present disclosure and are not necessarily drawn to scale. The same reference numerals in the figures represent the same or similar parts, and thus their repeated description will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities and do not necessarily correspond to physically or logically independent entities. These functional entities can be implemented in software form, or implemented in one or more hardware units or integrated circuits, or implemented in different networks and / or processor devices and / or microcontroller devices.

[0045] The flowcharts shown in the accompanying drawings are only exemplary and do not necessarily include all the steps. For example, some steps may be decomposed, while some steps may be combined or partially combined, so the actual execution order may change according to the actual situation.

[0046] The terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the terms used in this way can be interchanged where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein.

[0047] In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product, or apparatus comprising a series of steps or sub-modules is not necessarily limited to those steps or sub-modules explicitly listed, but may include other steps or sub-modules not explicitly listed or inherent to these processes, methods, products, or apparatuses.

[0048] like Figure 1 As shown, the present invention proposes a method for calculating a comprehensive evaluation index and processing data, the method comprising the following steps:

[0049] S1 imports the original data of the indicator system to be processed, and identifies the name, research object, and time range of each indicator in the original data;

[0050] S2 preprocesses the raw data, normalizes all indicators to eliminate the dimension and scale differences between different indicators;

[0051] The preprocessing of the original data includes: firstly positively converting the indicators, and then performing Z-score processing or Min-Max processing on the original data.

[0052] The Z-score processing includes: converting the original data into standard normal distribution data with a mean of 0 and a standard deviation of 1; the Min-Max processing includes: scaling the characteristic value of each indicator in the original data to a preset fixed interval.

[0053] S3 determines the weight of each indicator based on the preprocessed data;

[0054] The weight determination methods include: principal component analysis method, entropy weight method, equal weight method, subjective plus principal component analysis method, subjective plus entropy weight method, subjective plus equal weight method and expert subjective weighting method.

[0055] S4 calculates the comprehensive evaluation index through the weights and preprocessed indicator values;

[0056] S5 draws a first graph and a second graph respectively according to the comprehensive evaluation index, the original data or the pre-processed data.

[0057] The method further includes: analyzing the correlation between indicators based on the original data or the pre-processed data; and drawing a third graph based on the correlation analysis result.

[0058] The first graph includes a total score plot and a horizontal bar graph; the second graph includes a graph for each indicator and a graph for each indicator for each country; the third graph includes a correlation scatter plot and a correlation coefficient matrix heat value graph.

[0059] like Figure 2 As shown, the present invention also proposes a system for calculating a comprehensive evaluation index and processing data, the system comprising an identification module 10 , a preprocessing module 20 , a weight determination module 30 , an index calculation module 40 and a drawing module 50 .

[0060] The identification module 10 is used to import the original data of the indicator system to be processed, and to identify the name of each indicator, the research object, and the research time in the original data.

[0061] Specifically, the identification module provides a "select folder" function, so users can easily import the data to be processed; the identification module can also automatically identify the name of each indicator and default to the third-level indicator. It can also automatically identify the country and year in the data. If the analysis object is not a country, but other individuals (such as companies, individuals, etc.), or the time unit is not a year, it can also adapt.

[0062] The preprocessing module 20 is used to preprocess the original data, make all indicators positive, and normalize them to eliminate the dimension and scale differences between different indicators.

[0063] Specifically, the preprocessing module provides two normalization methods, including Z-score normalization and Min-Max normalization.

[0064] (1) Z-score normalization

[0065] Z-score normalization, also called standardization, is based on the principle of converting the value of a feature into a distribution with a mean of 0 and a standard deviation of 1. This method can handle the problem of outliers because it is based on the mean and standard deviation of the original data. Z-score normalization has no requirements for the distribution of data and can be used even if it is not a normal distribution. The normalization formula is:

[0066]

[0067] Among them, X is the original data, μ is the mean value of the data, σ is the standard deviation of the data, and x is the normalized result. Through Z-score normalization, the original data can be converted into a standard normal distribution with a mean of 0 and a standard deviation of 1, thereby eliminating the scale differences between different indicators.

[0068] (2) Min-Max Normalization

[0069] Min-Max normalization, also called minimum-maximum scaling, works by scaling the value of a feature to a fixed interval, usually between 0 and 1.

[0070] The normalized formula for the positive indicator is:

[0071]

[0072] The normalized formula for the negative indicator is:

[0073]

[0074] Among them, X represents the original data of the indicator, x represents the normalized data, and X min and X max They are respectively the minimum and maximum values ​​of the original data of the same indicator.

[0075] The weight determination module 30 is used to determine the weight of each indicator based on the preprocessed data.

[0076] Specifically, when calculating a composite indicator or an indicator system index, it is very important to determine the weight of each indicator. This is because the weight can reflect the importance of each indicator. If the weight is determined appropriately, the calculated composite indicator can better reflect the actual situation; on the contrary, if the weight is set unreasonably, the composite indicator may be distorted.

[0077] The weight determination module provides seven weight determination methods: overall equal weight method, subjective equal weight method, subjective weight method, overall principal component analysis method, overall entropy weight method, subjective plus principal component analysis method, and subjective plus entropy weight method. The objective methods used are principal component analysis method and entropy weight method.

[0078] (1) Principal component analysis

[0079] Principal component analysis is a multivariate statistical method. By studying the internal structure of multiple indicators with certain correlation, adhering to the core idea of ​​"dimensionality reduction", a few independent comprehensive indicators are used to replace the original more variable indicators, and the comprehensive indicators here are the principal components, which are linear combinations of the original variables. The number of principal components depends on the eigenvalue. The larger the eigenvalue, the more information the principal component contains. Since the purpose of principal component analysis is to reduce dimensionality, only principal components with eigenvalues ​​greater than 1 are selected. One of the advantages of principal component analysis is that the weights are determined by statistical analysis of the data, so there is no need for subjective judgment in advance.

[0080] Before principal component analysis, the normalized indicators need to pass Bartlett's sphericity test and KMO (Keiser-Meyer-Olkin) test. Only when both tests are passed can principal component analysis be performed. If they are not passed at the same time, weights need to be assigned by other methods. KMO test is used to compare simple correlation coefficients and partial correlation coefficients between variables. The closer the KMO value is to 1, the stronger the correlation between variables is, and the more suitable it is for principal component analysis; the closer the KMO value is to 0, the weaker the correlation between variables is, and the less suitable it is for principal component analysis. The KMO value suitable for principal component analysis is between 0.5 and 1. Bartlett's test is used to detect the correlation between indicators. When the significance level of the chi-square test is P<0.05, it indicates that the indicators are interrelated and suitable for principal component analysis; when P>0.05, it indicates that these variables may provide some information independently and are not suitable for principal component analysis. After passing KMO and Bartlett's test, grouping and weighting are performed through PCA calculation. In summary, principal component analysis can only be performed when KMO>0.5 and P of Bartlett's test<0.05.

[0081] The calculation steps of principal component analysis are:

[0082] 1) Assuming there are m countries, each country has n energy evaluation indicators, construct an m×n data matrix:

[0083]

[0084] 2) After normalization, we get:

[0085]

[0086] 3) Establish the correlation coefficient matrix of variables:

[0087]

[0088] R=(r ij ) p×p ;

[0089] Among them, r ij is the correlation coefficient between indicator i and indicator j, R is the correlation coefficient matrix, x ki is the normalized value of the ith indicator of the kth country, x kj is the normalized value of the jth indicator of the kth country, is the average value of index i, is the average value of index j.

[0090] 4) Find the eigenvalues ​​of the correlation coefficient matrix λ1≥λ2≥…≥λ n >0 and the corresponding eigenvector:

[0091]

[0092] 5) Calculate the principal component contribution rate and cumulative contribution rate:

[0093] Contribution rate:

[0094]

[0095] Cumulative contribution rate:

[0096]

[0097] 6) Generally, the first, second, ..., mth (m≤p) principal components corresponding to the eigenvalues ​​with cumulative contribution rates of 85% to 95% are taken.

[0098] 7) Calculate the indicator weight based on the eigenvector and contribution rate.

[0099] (2) Entropy Weight Method

[0100] The entropy weight method is a weight determination method based on the information entropy theory. It uses the information entropy theory to quantify the uncertainty and redundancy of each evaluation indicator in the evaluation process, thereby determining the weight of each indicator. This method scientifically quantifies the indicator weights through the information entropy principle, while fully considering the differences between the indicators. Information entropy is an important concept in information theory, first proposed by Shannon in 1948 to measure the uncertainty of information. In the entropy weight method, the greater the degree of variation of the indicator, the richer its information content and the greater the uncertainty, so its weight should be greater.

[0101] The entropy weight method has its limitations. It assumes that each indicator is completely independent, but in actual applications, there may be a certain correlation between the indicators. Therefore, when using the entropy weight method, it is necessary to select and adjust it according to the actual situation. The calculation process is as follows:

[0102] 1) Assuming there are m countries, each country has n energy evaluation indicators, construct an m×n data matrix:

[0103]

[0104] 2) After normalization, we get:

[0105]

[0106] 3) Calculate the index value proportion p of the i-th country under the j-th index ij :

[0107]

[0108] Among them, x ij is the normalized value of the jth indicator in the ith country.

[0109] 4) Calculate the information entropy e of the jth indicator j :

[0110]

[0111] If p ij =0, then define lim pij→0 p ij ln(p ij )=0. The calculation formula of k is as follows:

[0112]

[0113] 5) The weight of indicator j is:

[0114]

[0115] The specific meanings of the various weighting methods used in the present invention are as follows.

[0116] 1) Overall equal weight method: All three-level indicators adopt equal weights.

[0117] 2) Subjective weight method: The weights of all three-level indicators are given subjectively.

[0118] 3) Subjective plus equal weight method: A combination of the subjective method and the equal weight method. The weight of each first-level indicator is equal, and the weights of the three-level indicators within the same first-level indicator are equally distributed.

[0119] 4) Overall principal component analysis method: All three-level indicators are calculated and weighted using the principal component analysis method at the same time. The weight of the first-level indicator is obtained by adding the weights of the corresponding three-level indicators.

[0120] 5) Overall entropy weight method: The entropy weight method is used to determine the weights of all three-level indicators at the same time, and the weight of the first-level indicator is obtained by adding the weights of the corresponding three-level indicators.

[0121] 6) Subjective plus principal component analysis: A combination of subjective and principal component analysis. Subjective weights are assigned to the first-level indicators, and the principal component analysis method is used to determine a set of weights for the third-level indicators within each first-level indicator. The weight of the third-level indicator multiplied by the subjective weight of the first-level indicator is the weight of the third-level indicator in the entire indicator system.

[0122] 7) Subjective plus entropy weight method: a combination of subjective method and entropy weight method. The specific implementation plan is similar to the subjective plus principal component analysis method. First, the first-level indicators are given subjective weights, and the third-level indicators in each first-level indicator are simultaneously determined using the entropy weight method. The weight of the third-level indicator determined by multiplying the subjective weight of the first-level indicator is the weight of the third-level indicator in the entire indicator system.

[0123] The index calculation module 40 is used to calculate the comprehensive evaluation index through the weights and preprocessed indicator values.

[0124] Specifically, the calculation module can calculate the comprehensive evaluation index (ie, score) of each research object (eg, country, company, individual) at each time point (eg, year, month, day) according to the calculated weights.

[0125] The drawing module 50 is used to draw a first graph and a second graph respectively according to the comprehensive evaluation index, the original data or the pre-processed data. The first graph includes a total score graph and a horizontal bar graph; the second graph includes a graph for each indicator and a graph for each indicator for each country.

[0126] The system further includes a correlation analysis module; the correlation analysis module is used to analyze the correlation between indicators based on the original data or the pre-processed data. The drawing module 50 is used to draw a third graph based on the correlation analysis. The third graph includes a correlation scatter plot and a correlation coefficient matrix heat value graph.

[0127] Specifically, according to the comprehensive evaluation index, correlation analysis and original data or pre-processed data, the drawing module provides a total score drawing, a horizontal bar graph, a graph for each indicator, a graph for each indicator of each country, a correlation scatter plot and a correlation coefficient matrix heat value graph. The picture will be directly displayed on the software page and automatically saved to the same level folder as the original data. In this embodiment, the steps of the calculation module 40 and the drawing module 50 are combined together, which saves a lot of time for the user.

[0128] The total score plot refers to a line graph of the scores of each research object at each time point. Each research object is a line graph. It can directly compare the changes in the index of the research object over time, and can also directly compare the differences in score performance of different research objects at the same time point.

[0129] A horizontal bar chart is produced at each time point, and a horizontal bar is a research object composed of stacked scores of different indicators.

[0130] One graph for each indicator and one graph for each indicator in each country can both represent the changes in each indicator of each research object over time.

[0131] The correlation scatter plot uses two different indicators as the horizontal and vertical axes to intuitively display the correlation between the two indicators.

[0132] The correlation coefficient matrix heat value diagram intuitively displays the correlation coefficients between each pair of indicators in the form of a matrix heat value diagram.

[0133] Based on the above system, a comprehensive evaluation index and data processing software is provided, which integrates data import, data normalization, weight calculation, score drawing and data graphic display. The whole program is completed based on Python language. The specific contents are as follows:

[0134] 1. Data import

[0135] The software provides a "select folder" function, which allows users to easily import data that needs to be processed.

[0136] 2. Automatic identification

[0137] The software can automatically identify the names of each indicator and default to the third-level indicator. It can also automatically identify the country and year in the data. If the analysis object is not a country, but other individuals (such as companies, individuals, etc.), or the time unit is not a year, it can also adapt.

[0138] 3. Data Normalization

[0139] Data normalization is an important data preprocessing technique, which is mainly used to eliminate the dimension and scale differences between different indicators.

[0140] This software provides two normalization methods for users to choose from, including Z-score normalization and Min-Max normalization.

[0141] (1) Z-score normalization

[0142] Z-score normalization, also called standardization, is based on the principle of converting the value of a feature into a distribution with a mean of 0 and a standard deviation of 1. This method can handle the problem of outliers because it is based on the mean and standard deviation of the original data. Z-score normalization has no requirements for the distribution of data and can be used even if it is not a normal distribution. The normalization formula is:

[0143]

[0144] Among them, X is the original data, μ is the mean value of the data, σ is the standard deviation of the data, and x is the normalized result. Through Z-score normalization, the original data can be converted into a standard normal distribution with a mean of 0 and a standard deviation of 1, thereby eliminating the scale differences between different indicators.

[0145] (2) Min-Max Normalization

[0146] Min-Max normalization, also called minimum-maximum scaling, works by scaling the value of a feature to a fixed interval, usually between 0 and 1.

[0147] The normalized formula for the positive indicator is:

[0148]

[0149] The normalized formula for the negative indicator is:

[0150]

[0151] Among them, X represents the original data of the indicator, x represents the normalized data, and X min and X max They are respectively the minimum and maximum values ​​of the original data of the same indicator.

[0152] 4. Weight calculation

[0153] When calculating a composite indicator or an indicator system index, it is very important to determine the weight of each indicator. This is because the weight can reflect the importance of each indicator. If the weight is determined appropriately, the calculated composite indicator can better reflect the actual situation; on the contrary, if the weight is set unreasonably, the composite indicator may be distorted.

[0154] This software provides seven weight determination methods for users to choose from: overall equal weight method, subjective equal weight method, subjective weight method, overall principal component analysis method, overall entropy weight method, subjective plus principal component analysis method, and subjective plus entropy weight method. The objective methods used are principal component analysis method and entropy weight method.

[0155] (1) Principal component analysis

[0156] Principal component analysis is a multivariate statistical method. By studying the internal structure of multiple indicators with certain correlation, adhering to the core idea of ​​"dimensionality reduction", a few independent comprehensive indicators are used to replace the original more variable indicators, and the comprehensive indicators here are the principal components, which are linear combinations of the original variables. The number of principal components depends on the eigenvalue. The larger the eigenvalue, the more information the principal component contains. Since the purpose of principal component analysis is to reduce dimensionality, only principal components with eigenvalues ​​greater than 1 are selected. One of the advantages of principal component analysis is that the weights are determined by statistical analysis of the data, so there is no need for subjective judgment in advance.

[0157] Before principal component analysis, the normalized indicators need to pass Bartlett's sphericity test and KMO (Keiser-Meyer-Olkin) test. Only when both tests are passed can principal component analysis be performed. If they are not passed at the same time, weights need to be assigned by other methods. KMO test is used to compare simple correlation coefficients and partial correlation coefficients between variables. The closer the KMO value is to 1, the stronger the correlation between variables is, and the more suitable it is for principal component analysis; the closer the KMO value is to 0, the weaker the correlation between variables is, and the less suitable it is for principal component analysis. The KMO value suitable for principal component analysis is between 0.5 and 1. Bartlett's test is used to detect the correlation between indicators. When the significance level of the chi-square test is P<0.05, it indicates that the indicators are interrelated and suitable for principal component analysis; when P>0.05, it indicates that these variables may provide some information independently and are not suitable for principal component analysis. After passing KMO and Bartlett's test, grouping and weighting are performed through PCA calculation. In summary, principal component analysis can only be performed when KMO>0.5 and P of Bartlett's test<0.05.

[0158] The calculation steps of principal component analysis are:

[0159] 1) Assuming there are m countries, each country has n energy evaluation indicators, construct an m×n data matrix:

[0160]

[0161] 2) After normalization, we get:

[0162]

[0163] 3) Establish the correlation coefficient matrix of variables:

[0164]

[0165] R=(r ij ) p×p ;

[0166] 4) Find the eigenvalues ​​of the correlation coefficient matrix λ1≥λ2≥…≥λ n >0 and the corresponding eigenvector:

[0167]

[0168] 5) Calculate the principal component contribution rate and cumulative contribution rate:

[0169] Contribution rate:

[0170]

[0171] Cumulative contribution rate:

[0172]

[0173] 6) Generally, the first, second, ..., mth (m≤p) principal components corresponding to the eigenvalues ​​with cumulative contribution rates of 85% to 95% are taken.

[0174] 7) Calculate the indicator weight based on the eigenvector and contribution rate.

[0175] (2) Entropy Weight Method

[0176] The entropy weight method is a weight determination method based on the information entropy theory. It uses the information entropy theory to quantify the uncertainty and redundancy of each evaluation indicator in the evaluation process, thereby determining the weight of each indicator. This method scientifically quantifies the indicator weights through the information entropy principle, while fully considering the differences between the indicators. Information entropy is an important concept in information theory, first proposed by Shannon in 1948 to measure the uncertainty of information. In the entropy weight method, the greater the degree of variation of the indicator, the richer its information content and the greater the uncertainty, so its weight should be greater.

[0177] The entropy weight method has its limitations. It assumes that each indicator is completely independent, but in actual applications, there may be a certain correlation between the indicators. Therefore, when using the entropy weight method, it is necessary to select and adjust it according to the actual situation. The calculation process is as follows:

[0178] 1) Assuming there are m countries, each country has n energy evaluation indicators, construct an m×n data matrix:

[0179]

[0180] 2) After normalization, we get:

[0181]

[0182] 3) Calculate the index value proportion p of the i-th country under the j-th index ij :

[0183]

[0184] Among them, x ij is the normalized value of the jth indicator in the ith country.

[0185] 4) Calculate the information entropy e of the jth indicator j :

[0186]

[0187] If pij = 0, then define The calculation formula of k is as follows:

[0188]

[0189] 5) The weight of indicator j is:

[0190]

[0191] The specific meanings of each weighting method used in this software are as follows.

[0192] 1) Overall equal weight method: All three-level indicators adopt equal weights.

[0193] 2) Subjective weight method: The weights of all three-level indicators are given subjectively.

[0194] 3) Subjective plus equal weight method: A combination of the subjective method and the equal weight method. The weight of each first-level indicator is equal, and the weights of the three-level indicators within the same first-level indicator are equally distributed.

[0195] 4) Overall principal component analysis method: All three-level indicators are calculated and weighted using the principal component analysis method at the same time. The weight of the first-level indicator is obtained by adding the weights of the corresponding three-level indicators.

[0196] 5) Overall entropy weight method: The entropy weight method is used to determine the weights of all three-level indicators at the same time, and the weight of the first-level indicator is obtained by adding the weights of the corresponding three-level indicators.

[0197] 6) Subjective plus principal component analysis: The subjective method and principal component analysis are combined to assign subjective weights to the first-level indicators. The third-level indicators within each first-level indicator are also determined using principal component analysis to determine a set of weights. The weight of the third-level indicator determined multiplied by the subjective weight of the first-level indicator is the weight of the third-level indicator in the entire indicator system.

[0198] 7) Subjective plus entropy weight method: The combination of subjective method and entropy weight method. The specific implementation plan is similar to that of . First, the subjective weight is assigned to the first-level indicator. The third-level indicators in each first-level indicator use the entropy weight method to determine a set of weights. The weight of the third-level indicator determined by multiplying the subjective weight of the first-level indicator is the weight of the third-level indicator in the entire indicator system.

[0199] 5. Index drawing

[0200] Users can calculate the comprehensive evaluation index (i.e. score) of each research object (such as country, company, individual) at each time point (such as year, month, day) based on the calculated weight. According to the comprehensive evaluation index, the software provides two drawing methods: "total score drawing" and "horizontal bar graph". After selecting these two options, the picture will be directly displayed on the software page and automatically saved to the same folder as the original data. This function combines the steps of calculating scores and drawing, saving users a lot of time.

[0201] The total score plot is a line graph of the scores of each research object at each time point. Each research object is a line graph, which can directly compare the changes of the index of the research object over time, and can also directly compare the differences in the scores of different research objects at the same time point. The horizontal bar graph produces a graph at each time point, and a horizontal bar is a research object stacked by different indicator scores.

[0202] 6. Graphical display of raw data

[0203] The software has designed three graphical display methods for the original data, including "one graph for each indicator", "one graph for each indicator per country" and "correlation scatter plot".

[0204] One graph for each indicator and one graph for each country can both represent the changes of each indicator of each research object over time. The correlation scatter plot uses two different indicators as the horizontal and vertical axes to intuitively display the correlation between the two indicators.

[0205] 7. Graphical display of normalized data

[0206] The software can also draw four types of graphs based on the normalized data, including "correlation coefficient matrix heat map", "one map for each indicator", "correlation scatter plot" and "one map for each indicator for each country".

[0207] The correlation coefficient matrix heat value diagram displays the correlation coefficients between two indicators in the form of a matrix heat value diagram. One diagram for each indicator, a correlation scatter diagram, and one diagram for each indicator in each country are basically consistent with the purpose and function of the original data graphic display.

[0208] Those skilled in the art should understand that although the present invention has been described in detail with reference to the aforementioned embodiments, it is still possible to modify the technical solutions described in the aforementioned embodiments, or to make equivalent replacements for some of the technical features therein; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for calculating a comprehensive evaluation index and processing data, characterized in that: The method comprises: Import the original data of the indicator system to be processed, and identify the name, research object, and time range of each indicator in the original data; The raw data were preprocessed to make all indicators positive and normalized to eliminate the dimension and scale differences between different indicators; Determine the weight of each indicator based on the preprocessed data; Calculate the comprehensive evaluation index through weights and preprocessed indicator values; The first graph and the second graph are drawn respectively according to the comprehensive evaluation index, the original data or the pre-processed data.

2. The method for calculating a comprehensive evaluation index and processing data according to claim 1, characterized in that: The method further comprises: Analyze the correlation between indicators based on raw data or pre-processed data; Draw the third graph based on the correlation analysis results.

3. The method for calculating the comprehensive evaluation index and processing data according to claim 2, characterized in that: The first graph includes a total score plot and a horizontal bar graph; The second graph includes one graph for each indicator and one graph for each indicator for each country; The third graph includes a correlation scatter plot and a correlation coefficient matrix heat value plot.

4. The method for calculating a comprehensive evaluation index and processing data according to claim 1, characterized in that: The preprocessing of the original data includes: firstly performing positive transformation on the index, and then performing Z-score processing or Min-Max processing on the original data.

5. The method for calculating the comprehensive evaluation index and processing data according to claim 4, characterized in that: The Z-score processing includes: converting the original data into standard normal distribution data with a mean of 0 and a standard deviation of 1; The Min-Max processing includes: scaling the value of each indicator in the original data to between 0 and 1.

6. The method for calculating a comprehensive evaluation index and processing data according to claim 1, characterized in that: The weight determination methods include: principal component analysis method, entropy weight method, equal weight method, subjective plus principal component analysis method, subjective plus entropy weight method, subjective plus equal weight method and expert subjective weighting method.

7. A system for calculating a comprehensive evaluation index and processing data, characterized in that: The system includes an identification module, a preprocessing module, a weight determination module, an index calculation module and a drawing module; The identification module is used to import the original data of the indicator system to be processed, and identify the name of each indicator, research object, and research time in the original data; The preprocessing module is used to preprocess the raw data, make all indicators positive, and normalize them to eliminate the dimension and scale differences between different indicators; The weight determination module is used to determine the weight of each indicator based on the preprocessed data; The calculation index module is used to calculate the comprehensive evaluation index through the weight and the preprocessed index value; The drawing module is used to draw a first graph and a second graph respectively according to the comprehensive evaluation index, the original data or the pre-processed data.

8. The system for calculating comprehensive evaluation index and processing data according to claim 7, characterized in that: The system also includes a correlation analysis module; The correlation analysis module is used to analyze the correlation between indicators based on original data or preprocessed data.

9. The system for calculating comprehensive evaluation index and processing data according to claim 8, characterized in that: The drawing module is used to draw a third graph according to the correlation analysis.

10. The system for calculating comprehensive evaluation index and processing data according to claim 9, characterized in that: The first graph includes a total score plot and a horizontal bar graph; The second graph includes one graph for each indicator and one graph for each indicator for each country; The third graph includes a correlation scatter plot and a correlation coefficient matrix heat value plot.

11. The system for calculating comprehensive evaluation index and processing data according to claim 7, characterized in that: The preprocessing module is used to preprocess the original data, including: The preprocessing module is used to first positively transform the index, and then perform Z-score processing or Min-Max processing on the original data.

12. The system for calculating comprehensive evaluation index and processing data according to claim 7, characterized in that: The weight determination methods include: principal component analysis method, entropy weight method, equal weight method, subjective plus principal component analysis method, subjective plus entropy weight method, subjective plus equal weight method and expert subjective weighting method.

Images