A method and system for generating enterprise management index relationship

Abstract

The present invention provides a method and system for generating enterprise management indicator relationships, including obtaining basic attribute data, historical business data, historical warning data and user click habit data of the enterprise to be evaluated; analyzing each indicator in the basic attribute data, the historical business data, the historical warning data and the user click habit data through a preset indicator evaluation model to obtain multiple correlation coefficients corresponding to each indicator; grouping the multiple correlation coefficients according to a preset correlation coefficient grouping rule, judging whether the corresponding indicators are correlated with each other according to each group of correlation coefficients, and forming the correlated indicators into a corresponding indicator collection; screening the indicator collection according to a preset indicator identification standard to obtain a final indicator relationship set, and outputting it as an enterprise management indicator relationship result. The present invention overcomes the difficulties caused by complex indicator relationships in analyzing the reasons for the quality of indicator data, and improves the accuracy of indicator data prediction.

Description

Technical Field

[0001] The present invention relates to the technical field of enterprise management indicator relationship generation, and in particular to an enterprise management indicator relationship generation method and system. Background Art

[0002] The current status of enterprise operation monitoring systems is: there are many indicators, many indicator categories, diversified indicator monitoring rules, and complex relationships between indicators. The intricate indicator relationship brings certain difficulties to the analysis of the reasons for the quality of indicator data, resulting in the inability of enterprise decision makers to quickly and accurately find the reasons for the quality of indicator data, which in turn makes it impossible to make accurate decisions quickly. In addition, due to the lack of an accurate and scientific indicator knowledge graph, the prediction of indicator data has also brought great troubles. In view of this situation, the invention of constructing an enterprise operation indicator knowledge graph model based on the fitting and analysis mining algorithm of indicator basic attribute data, historical business data, and user indicator usage habit data is of great importance. Summary of the invention

[0003] The purpose of the present invention is to propose a method and system for generating business operation indicator relationships, so as to overcome the difficulties caused by the complex indicator relationships in analyzing the reasons for the quality of indicator data, and to improve the accuracy of indicator data prediction.

[0004] On the one hand, a method for generating a relationship between business indicators of an enterprise is provided, comprising:

[0005] Obtain basic attribute data, historical business data, historical warning data and user click habit data of the enterprise to be evaluated;

[0006] Analyze each indicator in the basic attribute data, the historical business data, the historical warning data and the user click habit data through a preset indicator evaluation model to obtain multiple correlation coefficients corresponding to each indicator;

[0007] The plurality of correlation coefficients are grouped according to a preset correlation coefficient grouping rule, and whether corresponding indicators are correlated with each other is determined according to each group of correlation coefficients, and the correlated indicators are grouped into a corresponding indicator set;

[0008] The indicator collection is screened according to the preset indicator identification criteria to obtain a final indicator relationship set, which is output as an enterprise management indicator relationship result.

[0009] Preferably, obtaining a plurality of correlation coefficients corresponding to each indicator specifically includes:

[0010] Analyze the basic attribute data and the historical business data through a preset evaluation model to obtain corresponding first correlation coefficient, second correlation coefficient and third correlation coefficient;

[0011] Analyze the historical warning data through a preset evaluation model to obtain corresponding fourth correlation coefficients and fifth correlation coefficients;

[0012] The user click habit data is analyzed by a preset evaluation model to obtain the corresponding sixth correlation coefficient and seventh correlation coefficient.

[0013] Preferably, the analyzing the basic attribute data and the historical business data by using a preset evaluation model specifically includes:

[0014] The first correlation coefficient is calculated according to the following formula:

[0015]

[0016]

[0017]

[0018] Among them, ρ XY It represents the first correlation coefficient between indicator X and indicator Y, namely the Pearson correlation coefficient, DX represents the Euclidean distance between indicator X and the average value, Cov(X,Y) represents the change trend between indicator X and indicator Y, n represents the total number of indicators, i represents the indicator sequence number, EX represents the average value of indicator X, and EY represents the average value of indicator Y;

[0019] The second correlation coefficient is calculated according to the following formula:

[0020]

[0021] Among them, r s represents the second correlation coefficient between indicator X and indicator Y, d i Represents X i and Y i The level difference between them is the Pearson correlation coefficient between the levels, n represents the total number of indicators, and i represents the indicator sequence number;

[0022] The third correlation coefficient is calculated according to the following formula:

[0023]

[0024] Wherein, R represents the third correlation coefficient between indicator X and indicator Y, P represents the number of statistical objects with the same arrangement size relationship as indicator X and indicator Y, and n represents the total number of indicators.

[0025] Preferably, the analyzing the historical warning data by using a preset evaluation model specifically includes:

[0026] The fourth correlation coefficient is calculated according to the following formula:

[0027]

[0028] Wherein, J(A, B) represents the fourth correlation coefficient between index X and index Y, |A∩B| represents the number of intersections of index A and index B, and |A∪B| represents the number of unions of index A and index B;

[0029] The fifth correlation coefficient is calculated according to the following formula:

[0030]

[0031] Among them, T(A, B) represents the fifth correlation coefficient between indicator A and indicator B, n represents the total number of indicators, and i represents the indicator sequence number.

[0032] Preferably, the analyzing the user's click habit data by using a preset evaluation model specifically includes:

[0033] The sixth correlation coefficient is calculated according to the following formula:

[0034]

[0035] Among them, Confidence(A|B) represents the sixth correlation coefficient, that is, the probability of visiting indicator B when indicator A has occurred, P(A∩B) represents the total probability of visiting indicators A and indicator B, and P(B) represents the probability of indicator A occurring in indicator B;

[0036] The seventh correlation coefficient is calculated according to the following formula:

[0037]

[0038] Among them, P(A∩B) represents the seventh correlation coefficient, that is, the total probability of the visit index A and the index B appearing, and P(I) represents the probability of the index A and the index B occurring simultaneously in the total number I.

[0039] Preferably, the correlation coefficient grouping rule specifically includes:

[0040] dividing the first correlation coefficient, the second correlation coefficient and the third correlation coefficient into a first group;

[0041] dividing the fourth correlation coefficient and the fifth correlation coefficient into the second group;

[0042] The sixth correlation coefficient and the seventh correlation coefficient are divided into the third group.

[0043] Preferably, judging whether corresponding indicators are correlated with each other according to each group of correlation coefficients specifically includes:

[0044] Determine whether the first correlation coefficient, the second correlation coefficient and the third correlation coefficient are the same or similar; if they are the same or similar, determine that the first correlation coefficient, the second correlation coefficient and the third correlation coefficient are related to each other; if they are not the same or similar, determine that the first correlation coefficient, the second correlation coefficient and the third correlation coefficient are not related to each other;

[0045] Determine whether the fourth correlation coefficient and the fifth correlation coefficient are the same or similar; if they are the same or similar, determine that the fourth correlation coefficient and the fifth correlation coefficient are related to each other; if they are not the same or similar, determine that the fourth correlation coefficient and the fifth correlation coefficient are not related to each other;

[0046] Determine whether the sixth correlation coefficient and the seventh correlation coefficient are the same or similar; if they are the same or similar, determine that the sixth correlation coefficient and the seventh correlation coefficient are correlated with each other; if they are not the same or similar, determine that the sixth correlation coefficient and the seventh correlation coefficient are not correlated with each other.

[0047] Preferably, the forming of the mutually related indicators into a corresponding indicator collection specifically includes:

[0048] The first correlation coefficient, the second correlation coefficient and the third correlation coefficient that are related to each other form a first indicator set;

[0049] The fourth correlation coefficient and the fifth correlation coefficient that are related to each other form a second indicator set;

[0050] The sixth correlation coefficient and the seventh correlation coefficient that are related to each other form a third indicator set;

[0051] The first indicator collection, the second indicator collection and the third indicator collection are combined to obtain a fourth indicator collection.

[0052] Preferably, it also includes:

[0053] Identify each indicator in the fourth indicator set according to a preset indicator identification standard, determine the indicators that meet the preset indicator identification standard as indicators with a real correlation relationship, and determine the indicators that do not meet the preset indicator identification standard as indicators without a real correlation relationship;

[0054] All indicators with real correlation relationships are combined into the final indicator relationship set.

[0055] On the other hand, a system for generating a relationship between business operating indicators is also provided, which is used to implement the method for generating a relationship between business operating indicators, and comprises:

[0056] Data acquisition module, used to obtain basic attribute data, historical business data, historical warning data and user click habit data of the enterprise to be evaluated;

[0057] An indicator evaluation module is used to analyze each indicator in the basic attribute data, the historical business data, the historical warning data and the user click habit data through a preset indicator evaluation model to obtain multiple correlation coefficients corresponding to each indicator; and group the multiple correlation coefficients according to a preset correlation coefficient grouping rule, determine whether the corresponding indicators are correlated with each other according to each group of correlation coefficients, and form the correlated indicators into a corresponding indicator collection;

[0058] The indicator relationship generation module is used to screen the indicator collection according to the preset indicator identification standard to obtain the final indicator relationship set and output it as the enterprise operation indicator relationship result.

[0059] In summary, the implementation of the embodiments of the present invention has the following beneficial effects:

[0060] The enterprise operation indicator relationship generation method and system provided by the present invention uses a method covering seven algorithms from three dimensions to make the indicator knowledge graph more scientific and accurate, thereby improving the accuracy of the enterprise's monitoring of operation indicators, and scientifically and accurately generating the indicator knowledge graph; accurately and scientifically determining the composition / correlation relationship between indicators, and using visualization display tools to form an indicator graph, which can save the cost of manually classifying the composition / correlation relationship of indicators; save the cost of analyzing the pros and cons of indicator operations; management efficiency. The scientific and accurate indicator knowledge graph provides a scientific basis for the analysis of the pros and cons of indicator operations. Managers can make more scientific and cost-effective decisions based on the analysis results. Thereby, the value output of the enterprise operation monitoring system can be greatly improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, without paying creative labor, other drawings obtained based on these drawings still belong to the scope of the present invention.

[0062] Figure 1 The main flow diagram of a method for generating a relationship between business operating indicators of an embodiment of the present invention is shown in FIG.

[0063] Figure 2 The figure is a logical diagram of a method for generating a relationship between business operating indicators of an embodiment of the present invention.

[0064] Figure 3Schematic diagram of a system for generating business operating indicator relationships in an embodiment of the present invention. DETAILED DESCRIPTION

[0065] In order to make the objectives, technical solutions and advantages of the present invention more clear, the present invention will be further described in detail below with reference to the accompanying drawings.

[0066] like Figure 1 and Figure 2 FIG. 1 is a schematic diagram of an embodiment of a method for generating a relationship between business indicators of an enterprise provided by the present invention. In this embodiment, the method comprises the following steps:

[0067] Obtain the basic attribute data, historical business data, historical warning data and user click habit data of the enterprise to be evaluated; that is, obtain the basic attribute data Data1 of the indicator from the enterprise operation monitoring system database; obtain the historical business data Data2 of the indicator from the enterprise operation monitoring system database; obtain the historical warning data Data3 of the indicator from the enterprise operation monitoring system database; obtain the user click habit data Data4 of the indicator from the enterprise operation monitoring system database.

[0068] Furthermore, the indicators in the basic attribute data, the historical business data, the historical warning data and the user click habit data are analyzed by a preset indicator evaluation model to obtain multiple correlation coefficients corresponding to each indicator; that is, in the dimension of "numerical correlation", the Person algorithm correlation coefficient A, the Speaman rank algorithm correlation coefficient B and the Kendall rank algorithm correlation coefficient C of the indicators are determined by analyzing and mining the data Data1 and Data2; indicators with the same or similar coefficients A, B and C are called indicators with related or constitutive relationships; the union of the sets of indicators with related and constitutive relationships, GroupA, is taken. In the dimension of "alarm correlation", the Jaccard correlation coefficient D and Tanimoto correlation coefficient E of the indicators are determined by analyzing and mining the data Data3; indicators with the same or similar coefficients D and E are called indicators with related or constitutive relationships; the union of the sets of indicators with related and constitutive relationships, GroupB, is taken. In the dimension of "user habits", the confidence F and lift G of the indicator are determined by analyzing and mining the data Data4; indicators with the same or similar confidence F and lift G are called related or constituent relationship indicators; the union GroupC of the sets of related and constituent relationship indicators is taken.

[0069] In a specific embodiment, the basic attribute data and the historical business data are analyzed by a preset evaluation model to obtain the corresponding first correlation coefficient, second correlation coefficient and third correlation coefficient; wherein the first correlation coefficient is calculated according to the following formula:

[0070]

[0071]

[0072]

[0073] Among them, ρ XY It represents the first correlation coefficient between indicator X and indicator Y, namely the Pearson correlation coefficient, DX represents the Euclidean distance of indicator X from the average value, Cov(X, Y) represents the changing trend between indicator X and indicator Y, n represents the total number of indicators, i represents the indicator serial number, EX represents the average value of indicator X, and EY represents the average value of indicator Y; it can be understood that the Pearson coefficient of the analysis and mining algorithm, the covariance of two variables divided by their standard deviation, can be understood as standardizing and eliminating the covariance affected by the dimension. Because the standard deviation reflects the degree of fluctuation of the data, due to different dimensions, the degree of fluctuation varies. After dividing by the standard deviation, it is equivalent to eliminating the degree of fluctuation of each, and more purely reflects the changing relationship between the variables. The Pearson correlation coefficient represents the linear correlation between the two indicators. The correlation coefficient is between [-1, 1], and its value quantitatively describes the degree of correlation between indicator X and indicator Y, that is, |ρ XY | is larger, the greater the correlation between the two. Generally speaking, if |ρ XY |≥0.8, called strong correlation; 0.5<|ρ XY |<0.8, called moderate correlation; 0.2<|ρ XY |≤0.5 is called weak correlation; others are uncorrelated. Among them, Cov(X, Y) measures the trend of change between two variables. If the change trends of the two variables are consistent, that is, they get bigger when they get bigger or smaller when they get smaller, even if one is greater than the mean and the other is less than the mean at some times, the product may be negative, but in general, the covariance should be positive. On the contrary, if the change trends of the two variables are opposite, the covariance should be negative. If the changes of the two variables are unrelated, they will be sometimes positive and sometimes negative, and in the end it is likely that the positive and negative will offset each other, making the covariance tend to 0. DX measures the Euclidean distance of the variable from the mean.

[0074] The second correlation coefficient is calculated according to the following formula:

[0075]

[0076] Among them, r s represents the second correlation coefficient between indicator X and indicator Y, d i Represents X i and Y iThe difference in levels between them, that is, the Pearson correlation coefficient between levels, n represents the total number of indicators, i represents the indicator sequence number; it can be understood that the Spearman coefficient of the analysis and mining algorithm; the Pearson correlation coefficient between levels, the levels can be divided by quantiles. The same is true for the Pearson correlation coefficient. The Pearson correlation coefficient is used for two indicators with continuous data and normal distribution, and the Spearman correlation coefficient is used for the rest.

[0077] The third correlation coefficient is calculated according to the following formula:

[0078]

[0079] Among them, R represents the third correlation coefficient between indicator X and indicator Y, P represents the number of statistical objects with the same size relationship as indicator X and indicator Y, and n represents the total number of indicators. It can be understood that the Kendall rank coefficient of the analysis and mining algorithm; the two indicators are two sides of the same coin at the same time point, and the values ​​of indicator A are arranged without loss of generality. Assume that the arrangement of the values ​​of indicator B at this time is disordered. Let P be the number of statistical objects with the same size relationship as the two indicator values. The indicators of ordered variables use the Kendall rank correlation coefficient.

[0080] Specifically, the historical warning data is analyzed by a preset evaluation model to obtain the corresponding fourth correlation coefficient and fifth correlation coefficient; wherein the fourth correlation coefficient is calculated according to the following formula:

[0081]

[0082] Among them, J(A, B) represents the fourth correlation coefficient between indicator X and indicator Y, |A∩B| represents the number of intersections of indicator A and indicator B, and |A∪B| represents the number of unions of indicator A and indicator B. It can be understood that the Jaccard correlation coefficient of the analysis mining algorithm is the ratio of the number of intersections of indicator A and indicator B to the number of unions of indicator A and indicator B.

[0083] The fifth correlation coefficient is calculated according to the following formula:

[0084]

[0085] Among them, T(A, B) represents the fifth correlation coefficient between indicator A and indicator B, n represents the total number of indicators, and i represents the indicator sequence number. It can be understood that the Tanimoto correlation coefficient of the analysis mining algorithm is considered as a vector and vector operations are performed.

[0086] More specifically, the user click habit data is analyzed by a preset evaluation model to obtain the corresponding sixth correlation coefficient and seventh correlation coefficient. The sixth correlation coefficient is calculated according to the following formula:

[0087]

[0088] Among them, Confidence(A|B) represents the sixth correlation coefficient, that is, the probability of accessing indicator B when indicator A has occurred, P(A∩B) represents the total probability of accessing indicator A and indicator B, and P(B) represents the probability of indicator A occurring in indicator B; it can be understood that the support of the analysis mining algorithm; support refers to the probability of simultaneous access of indicator A and indicator B. The total probability of the occurrence of user access indicators A and indicator B. It represents the probability of indicator A and indicator B occurring simultaneously in the total number I.

[0089] The seventh correlation coefficient is calculated according to the following formula:

[0090]

[0091] Among them, P(A∩B) represents the seventh correlation coefficient, that is, the total probability of the occurrence of access indicators A and B, and P(I) represents the probability of both indicators A and B occurring in the total number I. It can be understood that the confidence of the analysis mining algorithm; confidence refers to the probability of the occurrence of access indicator B when access indicator A has occurred.

[0092] Furthermore, the multiple correlation coefficients are grouped according to the preset correlation coefficient grouping rules, and whether the corresponding indicators are correlated with each other is determined based on each group of correlation coefficients, and the correlated indicators are grouped into corresponding indicator sets; that is, indicators with the same or similar coefficients A, B and C are called indicators with correlation or constituent relationship, and the union of the sets of indicators with correlation and constituent relationship is GroupA. Indicators with the same or similar coefficients D and E are called indicators with correlation or constituent relationship, and the union of the sets of indicators with correlation and constituent relationship is GroupB. Indicators with the same or similar confidence F and lift G are called indicators with correlation or constituent relationship, and the union of the sets of indicators with correlation and constituent relationship is GroupC.

[0093] In a specific embodiment, the first correlation coefficient, the second correlation coefficient and the third correlation coefficient are divided into a first group; the fourth correlation coefficient and the fifth correlation coefficient are divided into a second group; and the sixth correlation coefficient and the seventh correlation coefficient are divided into a third group.

[0094] Specifically, it is determined whether the first correlation coefficient, the second correlation coefficient and the third correlation coefficient are the same or similar; if they are the same or similar, it is determined that the first correlation coefficient, the second correlation coefficient and the third correlation coefficient are related to each other; if they are not the same or similar, it is determined that the first correlation coefficient, the second correlation coefficient and the third correlation coefficient are not related to each other;

[0095] Determine whether the fourth correlation coefficient and the fifth correlation coefficient are the same or similar; if they are the same or similar, determine that the fourth correlation coefficient and the fifth correlation coefficient are related to each other; if they are not the same or similar, determine that the fourth correlation coefficient and the fifth correlation coefficient are not related to each other;

[0096] Determine whether the sixth correlation coefficient and the seventh correlation coefficient are the same or similar; if they are the same or similar, determine that the sixth correlation coefficient and the seventh correlation coefficient are correlated with each other; if they are not the same or similar, determine that the sixth correlation coefficient and the seventh correlation coefficient are not correlated with each other.

[0097] More specifically, the first correlation coefficient, the second correlation coefficient and the third correlation coefficient that are related to each other form a first indicator set;

[0098] The fourth correlation coefficient and the fifth correlation coefficient that are related to each other form a second indicator set;

[0099] The sixth correlation coefficient and the seventh correlation coefficient that are related to each other form a third indicator set;

[0100] The first indicator collection, the second indicator collection and the third indicator collection are combined to obtain a fourth indicator collection.

[0101] Furthermore, the indicator collection is screened according to the preset indicator identification criteria to obtain the final indicator relationship set, which is output as the enterprise operation indicator relationship result. That is, the relationship forms the indicator relationship set GroupE, removes the indicator set without real structure / association relationship; and constructs the indicator knowledge graph through the indicator relationship set GroupE.

[0102] In a specific embodiment, each indicator in the fourth indicator set is identified according to a preset indicator identification standard, and the indicators that meet the preset indicator identification standard are determined as indicators with real correlation relationships, and the indicators that do not meet the preset indicator identification standard are determined as indicators without real correlation relationships; all indicators with real correlation relationships are formed into a final indicator relationship set.

[0103] like Figure 3 FIG. 1 is a schematic diagram of an embodiment of a system for generating a relationship between enterprise management indicators provided by the present invention. In this embodiment, the system includes:

[0104] Data acquisition module, used to obtain basic attribute data, historical business data, historical warning data and user click habit data of the enterprise to be evaluated;

[0105] An indicator evaluation module is used to analyze each indicator in the basic attribute data, the historical business data, the historical warning data and the user click habit data through a preset indicator evaluation model to obtain multiple correlation coefficients corresponding to each indicator; and group the multiple correlation coefficients according to a preset correlation coefficient grouping rule, determine whether the corresponding indicators are correlated with each other according to each group of correlation coefficients, and form the correlated indicators into a corresponding indicator collection;

[0106] The indicator relationship generation module is used to screen the indicator collection according to the preset indicator identification standard to obtain the final indicator relationship set and output it as the enterprise operation indicator relationship result.

[0107] It should be noted that the system described in the above embodiment corresponds to the method described in the above embodiment. Therefore, the undetailed parts of the system described in the above embodiment can be obtained by referring to the contents of the method described in the above embodiment, and will not be repeated here.

[0108] In summary, the implementation of the embodiments of the present invention has the following beneficial effects:

[0109] The enterprise operation indicator relationship generation method and system provided by the present invention uses a method covering seven algorithms from three dimensions to make the indicator knowledge graph more scientific and accurate, thereby improving the accuracy of the enterprise's monitoring of operation indicators, and scientifically and accurately generating the indicator knowledge graph; accurately and scientifically determining the composition / correlation relationship between indicators, and using visualization display tools to form an indicator graph, which can save the cost of manually classifying the composition / correlation relationship of indicators; save the cost of analyzing the pros and cons of indicator operations; management efficiency. The scientific and accurate indicator knowledge graph provides a scientific basis for the analysis of the pros and cons of indicator operations. Managers can make more scientific and cost-effective decisions based on the analysis results. Thereby, the value output of the enterprise operation monitoring system can be greatly improved.

[0110] The above disclosure is only the preferred embodiment of the present invention, which certainly cannot be used to limit the scope of the present invention. Therefore, equivalent changes made according to the claims of the present invention are still within the scope of the present invention.

Claims

1. A method for generating a relationship between business indicators of an enterprise, characterized in that: include: Obtain basic attribute data, historical business data, historical warning data and user click habit data of the enterprise to be evaluated; Analyze each indicator in the basic attribute data, the historical business data, the historical warning data and the user click habit data through a preset indicator evaluation model to obtain multiple correlation coefficients corresponding to each indicator; The plurality of correlation coefficients are grouped according to a preset correlation coefficient grouping rule, and whether corresponding indicators are correlated with each other is determined according to each group of correlation coefficients, and the correlated indicators are grouped into a corresponding indicator set; The indicator set is screened according to the preset indicator identification criteria to obtain a final indicator relationship set, which is output as an enterprise operation indicator relationship result; The obtaining of the multiple correlation coefficients corresponding to the indicators specifically includes: Analyze the basic attribute data and the historical business data through a preset evaluation model to obtain corresponding first correlation coefficient, second correlation coefficient and third correlation coefficient; Analyze the historical warning data through a preset evaluation model to obtain corresponding fourth correlation coefficients and fifth correlation coefficients; Analyze the user's click habit data through a preset evaluation model to obtain a corresponding sixth correlation coefficient and a seventh correlation coefficient; The correlation coefficient grouping rules specifically include: dividing the first correlation coefficient, the second correlation coefficient and the third correlation coefficient into a first group; dividing the fourth correlation coefficient and the fifth correlation coefficient into the second group; Divide the sixth correlation coefficient and the seventh correlation coefficient into the third group; The step of judging whether corresponding indicators are correlated with each other according to each group of correlation coefficients specifically includes: Determine whether the first correlation coefficient, the second correlation coefficient and the third correlation coefficient are the same or similar; if they are the same or similar, determine that the first correlation coefficient, the second correlation coefficient and the third correlation coefficient are related to each other; if they are not the same or similar, determine that the first correlation coefficient, the second correlation coefficient and the third correlation coefficient are not related to each other; Determine whether the fourth correlation coefficient and the fifth correlation coefficient are the same or similar; if they are the same or similar, determine that the fourth correlation coefficient and the fifth correlation coefficient are related to each other; if they are not the same or similar, determine that the fourth correlation coefficient and the fifth correlation coefficient are not related to each other; Determine whether the sixth correlation coefficient and the seventh correlation coefficient are the same or similar; if they are the same or similar, determine that the sixth correlation coefficient and the seventh correlation coefficient are related to each other; if they are not the same or similar, determine that the sixth correlation coefficient and the seventh correlation coefficient are not related to each other; The step of forming the mutually related indicators into a corresponding indicator collection specifically includes: The first correlation coefficient, the second correlation coefficient and the third correlation coefficient that are related to each other form a first indicator set; The fourth correlation coefficient and the fifth correlation coefficient that are related to each other form a second indicator set; The sixth correlation coefficient and the seventh correlation coefficient that are related to each other form a third indicator set; The first indicator collection, the second indicator collection and the third indicator collection are combined to obtain a fourth indicator collection.

2. The method according to claim 1, characterized in that The analyzing the basic attribute data and the historical business data by using a preset evaluation model specifically includes: The first correlation coefficient is calculated according to the following formula: Among them, ρ XY It represents the first correlation coefficient between indicator X and indicator Y, namely the Pearson correlation coefficient, DX represents the Euclidean distance between indicator X and the average value, Cov(X,Y) represents the change trend between indicator X and indicator Y, n represents the total number of indicators, i represents the indicator sequence number, EX represents the average value of indicator X, and EY represents the average value of indicator Y; The second correlation coefficient is calculated according to the following formula: Among them, r s represents the second correlation coefficient between indicator X and indicator Y, d i Represents X i and Y i The level difference between them is the Pearson correlation coefficient between the levels, n represents the total number of indicators, and i represents the indicator sequence number; The third correlation coefficient is calculated according to the following formula: Wherein, R represents the third correlation coefficient between indicator X and indicator Y, P represents the number of statistical objects with the same arrangement size relationship as indicator X and indicator Y, and n represents the total number of indicators.

3. The method according to claim 2, characterized in that The analyzing of the historical warning data by the preset evaluation model specifically includes: The fourth correlation coefficient is calculated according to the following formula: Wherein, J(A,B) represents the fourth correlation coefficient between index X and index Y, |A∩B| represents the number of intersections of index A and index B, and |A∪B| represents the number of unions of index A and index B; The fifth correlation coefficient is calculated according to the following formula: Among them, T(A,B) represents the fifth correlation coefficient between indicator A and indicator B, n represents the total number of indicators, and i represents the indicator sequence number.

4. The method according to claim 2, characterized in that The analyzing the user's click habit data by using a preset evaluation model specifically includes: The sixth correlation coefficient is calculated according to the following formula: Among them, Confidence(A|B) represents the sixth correlation coefficient, that is, the probability of visiting indicator B when indicator A has occurred, P(A∩B) represents the total probability of visiting indicators A and indicator B, and P(B) represents the probability of indicator A occurring in indicator B; The seventh correlation coefficient is calculated according to the following formula: Among them, P(A∩B) represents the seventh correlation coefficient, that is, the total probability of the visit index A and the index B appearing, and P(I) represents the probability of the index A and the index B occurring simultaneously in the total number I.

5. The method according to claim 1, characterized in that Also includes: Identify each indicator in the fourth indicator set according to a preset indicator identification standard, determine the indicators that meet the preset indicator identification standard as indicators with a real correlation relationship, and determine the indicators that do not meet the preset indicator identification standard as indicators without a real correlation relationship; All indicators with real correlation relationships are combined into the final indicator relationship set.

6. A system for generating business operating indicator relationships, used to implement the method according to any one of claims 1 to 5, characterized in that: include: Data acquisition module, used to obtain basic attribute data, historical business data, historical warning data and user click habit data of the enterprise to be evaluated; An indicator evaluation module is used to analyze each indicator in the basic attribute data, the historical business data, the historical warning data and the user click habit data through a preset indicator evaluation model to obtain multiple correlation coefficients corresponding to each indicator; and group the multiple correlation coefficients according to a preset correlation coefficient grouping rule, determine whether the corresponding indicators are correlated with each other according to each group of correlation coefficients, and form the correlated indicators into a corresponding indicator collection; An indicator relationship generation module is used to screen the indicator set according to a preset indicator identification standard to obtain a final indicator relationship set and output it as an enterprise operation indicator relationship result; The obtaining of the multiple correlation coefficients corresponding to the indicators specifically includes: Analyze the basic attribute data and the historical business data through a preset evaluation model to obtain corresponding first correlation coefficient, second correlation coefficient and third correlation coefficient; Analyze the historical warning data through a preset evaluation model to obtain corresponding fourth correlation coefficients and fifth correlation coefficients; Analyze the user's click habit data through a preset evaluation model to obtain a corresponding sixth correlation coefficient and a seventh correlation coefficient; The correlation coefficient grouping rules specifically include: dividing the first correlation coefficient, the second correlation coefficient and the third correlation coefficient into a first group; dividing the fourth correlation coefficient and the fifth correlation coefficient into the second group; Divide the sixth correlation coefficient and the seventh correlation coefficient into the third group; The step of judging whether corresponding indicators are correlated with each other according to each group of correlation coefficients specifically includes: Determine whether the first correlation coefficient, the second correlation coefficient and the third correlation coefficient are the same or similar; if they are the same or similar, determine that the first correlation coefficient, the second correlation coefficient and the third correlation coefficient are related to each other; if they are not the same or similar, determine that the first correlation coefficient, the second correlation coefficient and the third correlation coefficient are not related to each other; Determine whether the fourth correlation coefficient and the fifth correlation coefficient are the same or similar; if they are the same or similar, determine that the fourth correlation coefficient and the fifth correlation coefficient are related to each other; if they are not the same or similar, determine that the fourth correlation coefficient and the fifth correlation coefficient are not related to each other; Determine whether the sixth correlation coefficient and the seventh correlation coefficient are the same or similar; if they are the same or similar, determine that the sixth correlation coefficient and the seventh correlation coefficient are related to each other; if they are not the same or similar, determine that the sixth correlation coefficient and the seventh correlation coefficient are not related to each other; The step of forming the mutually related indicators into a corresponding indicator collection specifically includes: The first correlation coefficient, the second correlation coefficient and the third correlation coefficient that are related to each other form a first indicator set; The fourth correlation coefficient and the fifth correlation coefficient that are related to each other form a second indicator set; The sixth correlation coefficient and the seventh correlation coefficient that are related to each other form a third indicator set; The first indicator collection, the second indicator collection and the third indicator collection are combined to obtain a fourth indicator collection.

Images