Tobacco enterprise information system evaluation analysis method and system, electronic equipment and medium

Abstract

The embodiment of the invention provides a tobacco enterprise information system evaluation analysis method and system, electronic equipment and a storage medium, and the method comprises the steps: collecting evaluation indexes of a tobacco enterprise information system, calculating the subjective weight of the indexes, and carrying out the consistency check analysis to obtain the final subjective weight of the indexes; standardizing the subjective evaluation index and the objective evaluation index; calculating the standardized numerical data by using an entropy weight method to obtain an index objective weight, and combining the index objective weight with the final index subjective weight to obtain a combined weight; constructing a tobacco enterprise information system evaluation matrix based on a hierarchical clustering algorithm to obtain an index evaluation standard; determining a normal cloud model parameter for each evaluation index, constructing an improved normal cloud model, carrying out cross validation on the cloud model of each evaluation index, adjusting and optimizing the corresponding normal cloud model parameter, and obtaining an accurate membership value; and finally obtaining the overall evaluation grade of the tobacco enterprise information system by using weighted average operation.

Description

Technical Field

[0001] This document relates to the field of computer technology, and in particular to a tobacco enterprise information system evaluation and analysis method, system, electronic equipment and storage medium. Background Art

[0002] The construction of an information system is a comprehensive and systematic project, involving the coordination of various matters such as personnel, finance, and materials. The effectiveness of the information system also covers various aspects such as user satisfaction, economic benefits, and social benefits. In recent years, tobacco companies have actively carried out digital reforms and started the construction of multiple information systems with the development of the times. Due to the strong professionalism of the tobacco industry and the special users covered, tobacco companies often invest a lot of money, time, and manpower in the construction of information systems. In order to better carry out scientific and fair evaluations, conduct effective supervision and control, improve the management efficiency of information systems, and efficiently serve the various business operations of tobacco companies, an analysis method that can scientifically and comprehensively evaluate the information systems of tobacco companies is needed.

[0003] However, the current indicator system of information system evaluation and analysis methods lacks universality and is often only targeted at information systems in specific fields. There are too many indicators, which often weaken key indicators, and some indicators do not conform to the actual situation of tobacco. In addition, due to reasons such as information confidentiality, companies are unable to provide many cases, resulting in related research being mainly theoretical discussions and less empirical research, which makes it impossible to effectively verify the feasibility of the model or provide effective improvement strategies. Summary of the invention

[0004] The purpose of the present invention is to provide a tobacco enterprise information system evaluation and analysis method, system, electronic equipment and storage medium, aiming to solve the above-mentioned problems in the prior art.

[0005] The present invention provides a tobacco enterprise information system evaluation and analysis method, comprising: Step 1, collecting evaluation indicators of tobacco enterprise information systems, constructing an evaluation indicator matrix using a network analysis method, and calculating the subjective weights of indicators based on the evaluation indicator matrix, performing consistency test analysis on the subjective weights of indicators to obtain the final subjective weights of indicators, wherein the evaluation indicators include: subjective evaluation indicators and objective evaluation indicators; Step 2: standardize the subjective evaluation index and the objective evaluation index, calculate the proportion of the subjective evaluation index and the objective evaluation index, and correspond the subjective evaluation index and the objective evaluation index to a determined unified score according to the proportion to form standardized numerical data; Step 3, using the entropy weight method to calculate the standardized numerical data to obtain the objective weight of the indicator, and combining the objective weight of the indicator with the subjective weight of the final indicator to obtain the combined weight; Step 4, constructing a tobacco enterprise information system evaluation matrix based on a hierarchical clustering algorithm to obtain an indicator evaluation standard including an evaluation grade; determining normal cloud model parameters for each evaluation index, constructing an improved normal cloud model based on the normal cloud model parameters, generating a certain number of cloud droplets for each evaluation index according to the constructed improved normal cloud model, and for each cloud droplet, calculating its membership at each evaluation grade, that is, the cloud droplet value corresponds to the evaluation grade membership matrix of each evaluation grade, multiplying the cloud droplet of each evaluation index by the corresponding evaluation grade membership matrix to obtain the membership of each evaluation index at each evaluation grade, cross-validating the normal cloud model of each evaluation index, adjusting and optimizing the corresponding normal cloud model parameters, and finally obtaining the accurate membership value of each evaluation index at each evaluation grade; Step 5, based on the precise membership value and the combined weight, the evaluation results of each evaluation index are comprehensively processed using a weighted average operation, and the evaluation results after weighted summation are normalized to finally obtain the overall evaluation grade of the tobacco enterprise information system.

[0006] The present invention provides a tobacco enterprise information system evaluation and analysis system, comprising: The first weight calculation module is used to collect evaluation indicators of the tobacco enterprise information system, construct an evaluation indicator matrix using a network analysis method, calculate the subjective weight of the indicator based on the evaluation indicator matrix, and perform consistency test analysis on the subjective weight of the indicator to obtain the final subjective weight of the indicator, wherein the evaluation indicator includes: subjective evaluation indicators and objective evaluation indicators; The first processing module is used to standardize the subjective evaluation index and the objective evaluation index, calculate the proportion of the subjective evaluation index and the objective evaluation index, and correspond the subjective evaluation index and the objective evaluation index to a determined unified score according to the proportion to form standardized numerical data; A second weight calculation module is used to calculate the standardized numerical data using an entropy weight method to obtain an objective weight of the indicator, and combine the objective weight of the indicator with the final subjective weight of the indicator to obtain a combined weight; A normal cloud model module is used to construct an evaluation matrix of a tobacco enterprise information system based on a hierarchical clustering algorithm to obtain an indicator evaluation standard including an evaluation grade; determine normal cloud model parameters for each evaluation index, construct an improved normal cloud model based on the normal cloud model parameters, generate a certain number of cloud droplets for each evaluation index according to the constructed improved normal cloud model, calculate the membership of each cloud droplet at each evaluation grade, that is, the cloud droplet value corresponds to the evaluation grade membership matrix of each evaluation grade, multiply the cloud droplet of each evaluation index by the corresponding evaluation grade membership matrix, obtain the membership of each evaluation index at each evaluation grade, cross-validate the normal cloud model of each evaluation index, adjust and optimize the corresponding normal cloud model parameters, and finally obtain the accurate membership value of each evaluation index at each evaluation grade; The second processing module is used to comprehensively process the evaluation results of each evaluation index based on the precise membership value and the combined weight using weighted average operation, normalize the evaluation results after weighted summation, and finally obtain the overall evaluation level of the tobacco enterprise information system.

[0007] An embodiment of the present invention also provides an electronic device, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program implements the steps of the above-mentioned tobacco enterprise information system evaluation and analysis method when executed by the processor.

[0008] An embodiment of the present invention further provides a computer-readable storage medium, on which a program for implementing information transmission is stored, and when the program is executed by a processor, the steps of the above-mentioned tobacco enterprise information system evaluation and analysis method are implemented.

[0009] The embodiment of the present invention is adopted. Starting from the actual situation of tobacco enterprises, a quantitative and qualitative indicator system is formulated, which comprehensively considers expert opinions and objective data, forms an optimized and scientific indicator system, and improves the reliability of the indicator combination weight. Taking into account the multi-factor and fuzziness of the system comprehensive evaluation, the indicator weights are deeply analyzed using the network analysis method and the entropy weight method, and the interdependence between the indicators is fully considered to improve the accuracy of the weight calculation. Then, the improved normal cloud model algorithm is used in combination with the combined weights to carry out system rating, which improves the robustness and uncertainty processing ability of the evaluation results, and forms a scientific, effective and accurate system evaluation method. On the basis of the evaluation method, the actual system data is used for empirical research, and a comparative analysis is carried out with the existing evaluation mechanism of the enterprise to further optimize and improve the indicator system and evaluation method, and promote and implement it as needed, so as to truly provide rational and scientific suggestions for the construction of the information system of tobacco enterprises. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] In order to more clearly illustrate one or more embodiments of this specification 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 recorded in this specification. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0011] Figure 1 is a flow chart of a tobacco enterprise information system evaluation and analysis method according to an embodiment of the present invention; Figure 2 It is a processing schematic diagram of the process of the tobacco enterprise information system evaluation and analysis method according to an embodiment of the present invention; Figure 3 is a schematic diagram of normalized feature quantities according to an embodiment of the present invention; Figure 4 is a schematic diagram of a tobacco enterprise information system evaluation and analysis device according to an embodiment of the present invention; Figure 5 is a schematic diagram of an electronic device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0012] In order to enable those skilled in the art to better understand the technical solutions in one or more embodiments of this specification, the following will be combined with the drawings in one or more embodiments of this specification to clearly and completely describe the technical solutions in one or more embodiments of this specification. Obviously, the described embodiments are only part of the embodiments of this specification, not all of the embodiments. Based on one or more embodiments of this specification, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of this document.

[0013] Method Embodiment According to an embodiment of the present invention, a tobacco enterprise information system evaluation and analysis method is provided. Figure 1 Flow chart of the tobacco enterprise information system evaluation and analysis method according to an embodiment of the present invention. Figure 1 As shown, the tobacco enterprise information system evaluation and analysis method according to an embodiment of the present invention specifically includes: Step S101, collecting evaluation indicators of tobacco enterprise information systems, constructing an evaluation indicator matrix using a network analysis method, and calculating the subjective weights of indicators based on the evaluation indicator matrix, performing consistency test analysis on the subjective weights of indicators to obtain the final subjective weights of indicators, wherein the evaluation indicators include: subjective evaluation indicators and objective evaluation indicators; Wherein, step S101 specifically includes: Determine the evaluation indicators, where the evaluation indicators specifically include: the degree of coordinated construction, the degree of matching with strategic planning, the degree of matching with business, the degree of demand realization, the system flexibility, the degree of data quality, the system security, the system response speed, the degree of matching with the progress plan, the failure rate, the operating cost, the operation and maintenance response time, the upgrade frequency, the degree of system support, the user usage rate, the user satisfaction, the degree of improvement of organizational decision-making ability, the degree of cost reduction and efficiency improvement, and the degree of recognition outside the industry; The evaluation indicators are divided into two categories: subjective evaluation indicators and objective evaluation indicators. For subjective evaluation indicators, a questionnaire is prepared to clarify the survey objects; for objective evaluation indicators, the data source and calculation formula are clarified; The network analysis method is used to construct a pairwise comparison evaluation index matrix to form multiple comparison matrices. An unweighted supermatrix is ​​constructed based on each comparison matrix, and normalized to obtain a normalized matrix. The normalized matrix of each level is multiplied by the corresponding weight to obtain a weighted supermatrix. The geometric mean of the weighted supermatrix is ​​calculated and subjected to limiting processing to obtain a limiting supermatrix. The limiting supermatrix is ​​combined with the control layer weight to calculate the global priority of each evaluation indicator and obtain the subjective weight of the indicator after priority sorting. Based on the consistency test algorithm, the consistency ratio of the subjective weight of the indicator is calculated, and the comparison value of the comparison matrix is ​​corrected to ensure that the consistency CR of the comparison matrix is ​​less than 0.1, and the final subjective weight of the indicator is determined.

[0014] Step S102, standardizing the subjective evaluation index and the objective evaluation index, calculating the proportion of the subjective evaluation index and the objective evaluation index, and corresponding the subjective evaluation index and the objective evaluation index to a determined unified score according to the proportion to form standardized numerical data; specifically including: Select the same measurement method, determine the score range and standard scoring; analyze the collected subjective evaluation indicators and objective evaluation indicators, and calculate the cumulative proportion of the subjective evaluation indicators and the objective evaluation indicators; according to the cumulative proportion, formulate standardization rules, map the subjective evaluation indicators and objective evaluation indicators corresponding to the cumulative proportion to the determined standardized scores, and form standardized numerical data.

[0015] Step S103, using the entropy weight method to calculate the standardized numerical data to obtain the objective weight of the indicator, and combining the objective weight of the indicator with the final subjective weight of the indicator to obtain the combined weight; specifically including: The entropy value and its corresponding entropy weight of the standardized numerical data are calculated using the entropy weight method to obtain the objective weight of the indicator; the sensitivity coefficient is set, and the objective weight of the indicator is combined with the final subjective weight of the indicator to obtain the combined weight.

[0016] Step S104, constructing a tobacco enterprise information system evaluation matrix based on a hierarchical clustering algorithm to obtain an indicator evaluation standard including an evaluation grade; determining normal cloud model parameters for each evaluation index, constructing an improved normal cloud model based on the normal cloud model parameters, generating a certain number of cloud droplets for each evaluation index according to the constructed improved normal cloud model, calculating the membership of each cloud droplet at each evaluation grade, that is, the cloud droplet value corresponds to the evaluation grade membership matrix of each evaluation grade, multiplying the cloud droplet of each evaluation index by the corresponding evaluation grade membership matrix to obtain the membership of each evaluation index at each evaluation grade, cross-validating the cloud model of each evaluation index, adjusting and optimizing the corresponding normal cloud model parameters, and finally obtaining the accurate membership value of each evaluation index at each evaluation grade; specifically including: Collect the measurement data of the evaluation index, preset the number of clusters of each evaluation index based on the hierarchical clustering algorithm, and define the evaluation grade standard of each evaluation index; initialize the collected measurement data corresponding to each evaluation index into a cluster, take each cluster as a class, use the single-linkage clustering method, calculate the minimum distance between each other, merge the two classes with the smallest distance into a new class, and recalculate the distance between the new class and all classes until the number of clusters is reached, correspond the new class obtained by each evaluation index to different evaluation grades based on the evaluation grade standard, and construct the index evaluation standard; According to the complexity of the tobacco enterprise information system, the corresponding normal cloud parameters are selected, and the expected Ex, entropy En and super entropy He of each evaluation index are calculated according to formula 1, formula 2 and formula 3 respectively. A normal cloud model is constructed for each evaluation index based on the normal cloud parameters. A certain number of cloud droplets are generated for each index using the normal distribution function. For each cloud droplet, its membership at each evaluation level is calculated, that is, the cloud droplet value corresponds to the evaluation level membership matrix of each evaluation level. The cloud droplet of each index is multiplied by the corresponding evaluation level membership matrix to obtain the membership of each index at each evaluation level. The cloud model of each evaluation index is cross-validated, and the expected Ex, entropy En and super entropy He of the normal cloud parameters are adjusted and optimized. Finally, the accurate membership value of each evaluation index at each evaluation level is obtained: Formula 1; Formula 2; Formula 3; in, represents the i-th collected value in a single indicator data set, and n represents the total number of samples; It is the probability density function of the i-th value in the single index X set based on the Gaussian fuzzy factor, b represents the logarithmic base, usually b=2.

[0017] Step S105, based on the precise membership value and the combined weight, the evaluation results of each evaluation index are comprehensively processed by weighted average operation, and the evaluation results after weighted summation are normalized to finally obtain the overall evaluation level of the tobacco enterprise information system. Specifically including: Based on the precise membership value and the combined weight, the precise membership value of each evaluation index is multiplied by the combined weight using a weighted average operation to obtain a weighted membership, and the weighted memberships of all evaluation levels are added together to obtain an overall evaluation level of the tobacco enterprise information system.

[0018] It can be seen from the above technical scheme that the embodiment of the present invention is based on the internal vertical management model and external public service model of the tobacco industry, combined with the characteristics of tobacco government-enterprise integration, and comprehensively considers the importance, advancement, and efficiency of the information system's technical support for daily internal and external business of tobacco. It plans to build a comprehensive evaluation model from four aspects, including planning, construction, operation, and effectiveness, based on the hierarchical clustering information system evaluation matrix, using improved normal cloud combination weighting, and through empirical comparative analysis, explore the establishment of a scientific evaluation and analysis system for enterprise information systems, and provide reasonable suggestions for improving the operation of enterprise information systems.

[0019] The above technical solution of the embodiment of the present invention is described in detail below in conjunction with the accompanying drawings.

[0020] Figure 2 Schematic diagram of the processing flow of the tobacco enterprise information system evaluation and analysis method according to an embodiment of the present invention. Figure 2 As shown, the embodiment of the present invention discloses a tobacco enterprise information system evaluation and analysis research method based on the improved normal cloud model combined weighting, which aims to objectively and comprehensively evaluate the overall effectiveness and benefits of the tobacco enterprise's information system. This method combines qualitative and quantitative evaluation, and uses fuzzy mathematics theory to deal with the uncertainty and ambiguity in information system evaluation, providing a scientific basis for enterprise decision makers. The following are specific research data, methods, experiments and analysis, and conclusions.

[0021] 1. Research Data and Methods 1. Research data The embodiment of the present invention collects detailed data from multiple information systems, including the following aspects: System planning data: including the collection of users’ subjective evaluations of the degree of coordinated construction of information systems, the degree of matching with strategic planning, the degree of matching with business, and quantitative data on the degree of realization of system requirements through questionnaires and on-site surveys.

[0022] System construction data: including the collection of users' subjective evaluation of the flexibility (compatibility) of the information system through questionnaires and on-site surveys, as well as quantitative data such as the system data quality, system security, system response speed, and the degree of matching between progress and plan.

[0023] System operation data: including quantitative data such as operation failure rate, operation cost, operation and maintenance response time, upgrade frequency, and degree of institutional support after the implementation of the information system.

[0024] System effectiveness data: including subjective evaluations of users’ satisfaction with the information system, improvement in organizational decision-making capabilities, degree of cost reduction and efficiency improvement, and recognition outside the industry, as well as quantitative data on system user usage rates collected through questionnaires and on-site surveys.

[0025] 2. Research Methods The embodiment of the present invention adopts the improved normal cloud combination weighting to evaluate and analyze the enterprise information system, and the specific steps are as follows: Step 1: Determine the evaluation indicators of the information system. The embodiment of the present invention obtains evaluation indicators designed jointly by industry experts, scholars in the field of information systems, and key user representatives within the enterprise. The embodiment of the present invention establishes four first-level evaluation dimensions, which are independent and represent the four core aspects of information system evaluation, namely system planning, system construction, system operation, and system effectiveness. For each first-level indicator, it is further refined into a number of second-level indicators. These second-level indicators are a deepening and refinement of the first-level indicators, aiming to accurately describe the various performance characteristics of the information system. Furthermore, each second-level indicator is further divided into multiple third-level indicators, which serve as the most basic evaluation unit and directly correspond to quantifiable or perceptible system attributes. Finally, as shown in Table 1, the embodiment of the present invention constructs a multi-level, all-round evaluation system including nine second-level indicators and nineteen third-level indicators to ensure a comprehensive and detailed evaluation of the information system.

[0026] Step 2: Use the network analysis method (ANP) to construct the evaluation index matrix and determine the index weights. Use the network analysis method to construct a decision-making hierarchy with the first-level index as the top layer, the second-level and third-level indicators as the middle layer and the bottom layer respectively. Collect the relative importance of the indicators at each layer, construct the evaluation index matrix, and then accurately calculate the relative weights of the indicators at each level through hierarchical single sorting and consistency test.

[0027] Step 3: Data standardization. Based on expert scoring and preliminary survey data, a unified quantitative value range is selected, and the percentage of different types of subjective and objective data is analyzed. According to the percentage, the subjective and objective data are mapped to a determined unified score to form standardized numerical data.

[0028] Step 4: Calculate the combined weight. Based on the standardized data obtained in step 3, calculate the entropy value of each indicator and its corresponding entropy weight to reflect the contribution of objective data to the importance of the indicator. And by setting the sensitivity coefficient, combine the subjective weight obtained by the network analysis method in step 2 with the objective weight of the entropy weight method, so as to give each indicator a comprehensive weight that takes into account both expert experience and data rules, ensuring the scientificity and rationality of the weight distribution of the evaluation system.

[0029] Step 5: Construct an enterprise information system evaluation matrix based on a hierarchical clustering algorithm. Preset the number of clusters for each three-level indicator, and define the evaluation level standards of "excellent", "good", "average", "poor" and "bad" for each three-level indicator. According to the three-level indicators, initialize each collected measurement data sample corresponding to each indicator as a cluster, regard each cluster as a class, use the single-linkage clustering method, calculate the minimum distance between the two, repeat the previous step until the preset number of clusters is reached, correspond the new classes obtained from each three-level indicator to different evaluation levels, and construct the indicator evaluation standard.

[0030] Step 6: Normal cloud model evaluation and cross-validation: Through expert review and systematic data collection, the key parameters of the normal cloud model are determined: super entropy (He) reflects the level of uncertainty, expected value (Ex) reflects the ideal state, and entropy (En) measures the uncertainty of information. Using the normal distribution theory, a group of cloud droplets is generated for each indicator. Each group of cloud droplets represents a large number of samples of possible values ​​of the indicator under the normal cloud model, effectively simulating the inherent uncertainty in the evaluation. The cloud droplets of each indicator are multiplied by the corresponding evaluation level membership matrix to obtain the membership of each indicator at each evaluation level. The cloud model of each indicator is cross-validated to adjust and optimize the expected value Ex, entropy En, and super entropy He, and finally obtain the accurate membership value of each indicator at each evaluation level.

[0031] Step 7: Comprehensive system rating. The evaluation results of each evaluation index are comprehensively processed using weighted average operation, where the weight is the combined weight of each index determined in step 4. After normalizing the weighted summation result, the overall evaluation grade of the enterprise information system is finally obtained, that is, its membership value in the five grades of "excellent", "good", "average", "poor" and "bad".

[0032] 2. Experiment and Analysis In the experimental design stage of the embodiment of the present invention, in order to verify and demonstrate the application effect of the evaluation method based on the improved normal cloud model combined weighting in the evaluation of tobacco enterprise information systems, the embodiment of the present invention selected 9 typical enterprise information systems covering different types, scales and industry backgrounds as research samples. Through in-depth research and data analysis of these samples, the embodiment of the present invention will use the constructed comprehensive evaluation index system and normal cloud model to conduct specific evaluation and analysis.

[0033] Step 1: Determine the evaluation indicators of the information system. Invite industry experts, scholars in the field of information systems, and key user representatives within the enterprise to participate in the design of evaluation indicators and build a multi-level, comprehensive evaluation system that includes nine secondary indicators and nineteen tertiary indicators, as shown in Table 1.

[0034] Table 1 shows the evaluation indicators of information system

[0035] Step 2: Use the network analysis method (ANP) to construct the evaluation index matrix and determine the index weights. For each sample information system, the relative importance judgment matrix of indicators at each level is obtained through expert scoring and user surveys, and the relative weights of indicators at each level are accurately calculated through hierarchical single sorting and consistency tests.

[0036] Specific process: The information system evaluation R is used as the target layer, and the evaluation indicators and their mutually reachable influence relationships are used as the network layer. Evaluation indicator layer The lower-level evaluation index in (i=1,2,...N) between, Evaluation index layer and There is a network relationship of mutual dependence and influence between the evaluation indicator layers (j=1,2,...N), and the indicator correlation is shown in Table 2.

[0037] Table 2 is the correlation table of information system evaluation indicators

[0038] Among them, the top element is the affected factor, and the left column is the factors that may affect the top element.

[0039] Construct an unweighted initial supermatrix, compare the pairwise dominance of the evaluation indicators, and obtain an initial judgment matrix. The present invention uses the indirect dominance comparison method to construct the initial judgment matrix. Based on the indirect dominance comparison, with the help of expert group discussion and judgment, each evaluation level is Evaluation indicators under (i=1,2,...N; j=1,2,...N) are compared pairwise, and the 1-9 scaling method is used to establish the pairwise comparison judgment matrix of the evaluation indicators as shown in Table 3.

[0040] Table 3 is an example of the relative importance judgment matrix of indicators

[0041] The scale value can be 1-9, 1 means that the two indicators are equally important, 3 means that indicator A is slightly more important than indicator B, 5 means that indicator A is obviously more important than indicator B, 7 means that indicator A is strongly more important than indicator B, 9 means that indicator A is extremely important than indicator B, and 2, 4, 6, and 8 are the medians of adjacent judgments.

[0042] like Figure 3 As shown, the maximum eigenvalue is calculated, the eigenvector is normalized, and the maximum eigenvalue of the judgment matrix is ​​introduced. The ratio of the difference between n and n (n is the order of the judgment matrix) and (n-1) is used as an indicator to measure the deviation of the judgment matrix from consistency: , the ratio of the consistency index CI to the average random consistency index RI of the order is recorded as CR, <0.1, which means the consistency check has passed.

[0043] The sorting vector is calculated by the characteristic root method, ( , ,... )remember :

[0044] The column vector is the evaluation index Evaluation indicators The influence ranking vector of right No impact, then =0 (i=1,2,...N; j=1,2,...N). The unweighted supermatrix for evaluating mutual influence is shown in the formula: ; Calculate the weighted super matrix and evaluate the index layer Evaluation index layer for sub-criteria Compare the importance, construct the judgment matrix, calculate the eigenvector, and perform consistency test. The ranking vector of irrelevant evaluation index groups is 0, and the weighted matrix A is obtained: ; Weight the factors in the supermatrix W to obtain the weighted supermatrix: ; Calculate the limit supermatrix to determine the factor weights: ; Evaluation indicators In their respective evaluation levels Normalize the weight vector in: .

[0045] The specific ANP index weight results calculated based on the network analysis method are shown in Table 4.

[0046] Table 4 shows the weight results of ANP indicators in network analysis method

[0047] Step 3: Data standardization. Based on the expert scores and the preliminary survey data, a unified quantitative value range for each indicator was selected, and the percentage of different types of subjective and objective data was statistically analyzed. According to the percentage, the subjective and objective data were mapped to a determined unified score to form standardized numerical data, as shown in Table 5.

[0048] Table 5 shows the evaluation values ​​of information system 1 after standardization

[0049] Step 4: Calculate the combined weight. Based on the standardized data obtained in step 3, calculate the entropy value of each indicator and its corresponding entropy weight, and by setting the sensitivity coefficient, combine the subjective weight obtained by the network analysis method in step 2 with the objective weight of the entropy weight method to obtain the combined weight of each indicator.

[0050] Specific process: According to the expert scores, the evaluation results of each sample system are obtained.

[0051] Construct an evaluation matrix and standardize the data: Positive indicators ; Negative indicators ; Among them, X max is the maximum value of the i-th index, X min is the minimum value of the i-th indicator.

[0052] Normalization process obtains the standard matrix: , where V ij Represents the data matrix after normalization of positive and negative indicators.

[0053] Calculate the information entropy of the indicator: , , where k is the calculation coefficient, m represents the number of data, P ij Represents the normalized data matrix.

[0054] Calculate the index difference coefficient: , where e j Represents information entropy.

[0055] Calculate indicator weights: , after normalization, the entropy weight of each indicator is obtained, see Table 6.

[0056] Table 6 shows the weight results of each indicator of the entropy weight method

[0057] The subjective weight obtained by the network analysis method and the objective weight obtained by the entropy weight method are weighted and integrated to calculate the combined weight.

[0058]

[0059] Where a is the sensitivity coefficient, set to 0.5, W ANP represents the indicator weight matrix obtained by network analysis method, W 熵权法 Represents the indicator weight matrix obtained by the entropy weight method; the combined weight results are shown in Table 7.

[0060] Table 7 shows the combined weight results of each indicator

[0061] Step 5: Construct an enterprise information system evaluation matrix based on a hierarchical clustering algorithm. Preset the number of clusters for each three-level indicator, and define the evaluation level standards of "excellent", "good", "average", "poor" and "bad" for each three-level indicator. Use single-linkage clustering to correspond the new classes obtained from each three-level indicator to different evaluation levels and construct the indicator evaluation standard.

[0062] Specific process: Initialization: According to the three-level indicators, each measurement data sample of each indicator is initialized into a cluster; Calculate distance: Calculate the shortest distance between all clusters; Merge: Select the two closest clusters and merge them into one cluster; Repeat the steps until the number of clusters reaches the preset number of clusters.

[0063] Step 6: Normal cloud model evaluation and cross-validation.

[0064] Determine the parameters of the normal cloud model. For each evaluation indicator, according to expert evaluation and historical data, determine the parameters of the normal cloud: expected value (Ex), entropy (En), and super entropy (He)

[0065]

[0066]

[0067] Construct a normal cloud model, use (Ex, En, He) to construct a normal cloud model, use the normal distribution function, and generate a certain number of cloud droplets for each indicator according to the constructed normal cloud model. The cloud droplets represent the random sampling of the indicator value in the normal distribution to simulate the uncertainty of the indicator. For each cloud droplet, according to the five-level evaluation scale, calculate its membership at each evaluation level, that is, the cloud droplet value corresponds to the membership function value of each level. Multiply the cloud droplet of each indicator with the corresponding evaluation level membership matrix to obtain the membership of each indicator at each evaluation level. Cross-validate the cloud model of each indicator and adjust the optimization expectation Ex, entropy En, and super entropy He.

[0068] Specific process: Data preprocessing: Ensure that the data of all indicators have been preprocessed, including missing value processing, outlier detection and processing, standardization, normalization, etc.; Initial parameter setting: Based on historical data and expert knowledge, the normal cloud model parameters of each indicator are preliminarily set: expected value Ex (reflecting the ideal or average state of the indicator), entropy En (measuring the degree of discreteness of the data), and super entropy He (increasing the model's ability to describe uncertainty); Cloud droplet generation and membership calculation: Using the normal distribution function, cloud droplets are generated for each indicator according to the current parameters, i.e., random sampling values, and the membership of each cloud droplet on the evaluation level is calculated. The membership function reflects the degree of compliance of the indicators at different levels. Cross-validation setting: The data set is divided into a training set and a validation set. The cross-validation method is used to divide the data set into k subsets. The model is trained with data from k-1 subsets each time, and the remaining subsets are used to verify the model performance. Parameter adjustment and optimization: For the training set, the gradient descent optimization algorithm is used to adjust the Ex, En, and He parameters, and cloud droplets are regenerated multiple times until the prediction error of the model on the validation set is minimized. After multiple cross-validations, the parameter combination with the best average performance on all validation subsets is selected.

[0069] Step 7: Comprehensive evaluation calculation. The evaluation results of each evaluation index are comprehensively processed using weighted average operation, where the weight is the combined weight of each index determined in step 4. After normalizing the weighted summation result, the overall evaluation grade of the tobacco enterprise information system is finally obtained, that is, its membership value in the five grades of "excellent", "good", "average", "poor" and "bad". Specific process: Multiply the membership of each indicator by the comprehensive weight in step 4 to obtain the weighted membership, calculate the sum of the weighted membership of all evaluation levels, and obtain the comprehensive evaluation value of the system, that is, the membership values ​​at the five levels. , is the comprehensive indicator weight, is the index membership, and the results are shown in Table 8.

[0070] Table 8 shows the comprehensive evaluation results of the sample system

[0071] Conclusion In the embodiment of the present invention, the evaluation method based on the combined weighting of the improved normal cloud model was successfully used to conduct a comprehensive and in-depth evaluation and analysis of the enterprise information system. By studying 9 representative enterprise information system samples, the embodiment of the present invention first constructed a comprehensive evaluation index system including system planning, system construction, system operation and system effectiveness, and used the combination of network analysis method (ANP) and entropy weight method to scientifically and reasonably determine the weights of indicators at all levels.

[0072] During the experiment, the embodiment of the present invention designed a corresponding membership function for each third-level indicator, combined with expert opinions and historical data, to ensure that the evaluation results can accurately reflect the actual performance of the information system in each evaluation dimension. By combining subjective judgment with objective data, the embodiment of the present invention established an evaluation matrix for each sample and performed a normal cloud model comprehensive evaluation operation.

[0073] The experimental results show that the combined weighting method based on the improved normal cloud model can effectively deal with the uncertainty and ambiguity in information system evaluation. Compared with the traditional evaluation method, the evaluation results are more objective and comprehensive. Compared with the traditional single evaluation method, the embodiment of the present invention can evaluate the overall effectiveness of the enterprise information system more finely and accurately, and is not affected by extreme data, reflecting good robustness and being more in line with reality.

[0074] In summary, the present invention demonstrates strong practicality and effectiveness by constructing a complete comprehensive evaluation index system and combining the normal cloud model combined weighting method to evaluate and analyze the tobacco enterprise information system. This innovative method not only enriches the theoretical system of information system evaluation, but also provides a powerful tool support for the informatization construction and operation management of tobacco enterprises. Therefore, the technical solution of the embodiment of the present invention has important scientific research value and broad market application prospects.

[0075] Device Example 1 According to an embodiment of the present invention, a tobacco enterprise information system evaluation and analysis system is provided. Figure 4 Schematic diagram of the tobacco enterprise information system evaluation and analysis system according to an embodiment of the present invention. Figure 4 As shown, the tobacco enterprise information system evaluation and analysis system according to the embodiment of the present invention specifically includes: The first weight calculation module 40 is used to collect evaluation indicators of the tobacco enterprise information system, construct an evaluation indicator matrix using a network analysis method, calculate the subjective weight of the indicator based on the evaluation indicator matrix, and perform consistency test analysis on the subjective weight of the indicator to obtain the final subjective weight of the indicator, wherein the evaluation indicator includes: subjective evaluation indicators and objective evaluation indicators; The first processing module 41 is used to standardize the subjective evaluation index and the objective evaluation index, calculate the proportion of the subjective evaluation index and the objective evaluation index, and correspond the subjective evaluation index and the objective evaluation index to a determined unified score according to the proportion to form standardized numerical data; A second weight calculation module 42 is used to calculate the standardized numerical data using an entropy weight method to obtain an objective weight of the indicator, and combine the objective weight of the indicator with the final subjective weight of the indicator to obtain a combined weight; The normal cloud model module 43 is used to construct an evaluation matrix of the tobacco enterprise information system based on a hierarchical clustering algorithm to obtain an indicator evaluation standard including an evaluation grade; determine normal cloud model parameters for each evaluation index, construct an improved normal cloud model based on the normal cloud model parameters, generate a certain number of cloud droplets for each evaluation index according to the constructed improved normal cloud model, calculate the membership of each cloud droplet at each evaluation grade, that is, the cloud droplet value corresponds to the evaluation grade membership matrix of each evaluation grade, multiply the cloud droplet of each evaluation index by the corresponding evaluation grade membership matrix, obtain the membership of each evaluation index at each evaluation grade, cross-validate the cloud model of each evaluation index, adjust and optimize the corresponding normal cloud model parameters, and finally obtain the accurate membership value of each evaluation index at each evaluation grade; The second processing module 44 is used to perform comprehensive processing on the evaluation results of each evaluation index based on the precise membership value and the combined weight using weighted average operation, normalize the evaluation results after weighted summation, and finally obtain the overall evaluation level of the tobacco enterprise information system.

[0076] The embodiment of the present invention is a system embodiment corresponding to the above-mentioned method embodiment. The specific operations of each module can be understood by referring to the description of the method embodiment, which will not be repeated here.

[0077] Device Example 2 An embodiment of the present invention provides an electronic device, such as Figure 5 As shown, it includes: a memory 50, a processor 52, and a computer program stored in the memory 50 and executable on the processor 52. When the computer program is executed by the processor 52, the steps described in the method embodiment are implemented.

[0078] Device Example 3 An embodiment of the present invention provides a computer-readable storage medium, on which a program for implementing information transmission is stored. When the program is executed by the processor 52, the steps described in the method embodiment are implemented.

[0079] The computer-readable storage medium in this embodiment includes, but is not limited to, ROM, RAM, magnetic disk or optical disk, etc.

[0080] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A tobacco enterprise information system evaluation and analysis method, characterized in that: include: Step 1, collecting evaluation indicators of tobacco enterprise information systems, constructing an evaluation indicator matrix using a network analysis method, and calculating the subjective weights of indicators based on the evaluation indicator matrix, performing consistency test analysis on the subjective weights of indicators to obtain the final subjective weights of indicators, wherein the evaluation indicators include: subjective evaluation indicators and objective evaluation indicators; Step 2: standardize the subjective evaluation index and the objective evaluation index, calculate the proportion of the subjective evaluation index and the objective evaluation index, and correspond the subjective evaluation index and the objective evaluation index to a determined unified score according to the proportion to form standardized numerical data; Step 3, using the entropy weight method to calculate the standardized numerical data to obtain the objective weight of the indicator, and combining the objective weight of the indicator with the subjective weight of the final indicator to obtain the combined weight; Step 4, constructing a tobacco enterprise information system evaluation matrix based on a hierarchical clustering algorithm to obtain an indicator evaluation standard including an evaluation grade; determining normal cloud model parameters for each evaluation index, constructing an improved normal cloud model based on the normal cloud model parameters, generating a certain number of cloud droplets for each evaluation index according to the constructed improved normal cloud model, and for each cloud droplet, calculating its membership at each evaluation grade, that is, the cloud droplet value corresponds to the evaluation grade membership matrix of each evaluation grade, multiplying the cloud droplet of each evaluation index by the corresponding evaluation grade membership matrix to obtain the membership of each evaluation index at each evaluation grade, cross-validating the normal cloud model of each evaluation index, adjusting and optimizing the corresponding normal cloud model parameters, and finally obtaining the accurate membership value of each evaluation index at each evaluation grade; Step 5, based on the precise membership value and the combined weight, the evaluation results of each evaluation index are comprehensively processed using a weighted average operation, and the evaluation results after weighted summation are normalized to finally obtain the overall evaluation grade of the tobacco enterprise information system.

2. The method according to claim 1, characterized in that The evaluation indicators for collecting tobacco enterprise information systems include: Determine the evaluation indicators, where the evaluation indicators specifically include: the degree of coordinated construction, the degree of matching with strategic planning, the degree of matching with business, the degree of demand realization, the system flexibility, the degree of data quality, the system security, the system response speed, the degree of matching with the progress plan, the failure rate, the operating cost, the operation and maintenance response time, the upgrade frequency, the degree of system support, the user usage rate, the user satisfaction, the degree of improvement of organizational decision-making ability, the degree of cost reduction and efficiency improvement, and the degree of recognition outside the industry; The evaluation indicators are divided into two categories: subjective evaluation indicators and objective evaluation indicators. For subjective evaluation indicators, a questionnaire is prepared to clarify the survey objects; for objective evaluation indicators, the data source and calculation formula are clarified.

3. The method according to claim 1, characterized in that: The evaluation index matrix is ​​constructed by network analysis method, and the subjective weight of the index is calculated based on the evaluation index matrix. The subjective weight of the index is subjected to consistency test analysis to obtain the final subjective weight of the index, which specifically includes: The network analysis method is used to construct a pairwise comparison evaluation index matrix to form multiple comparison matrices. An unweighted supermatrix is ​​constructed based on each comparison matrix, and normalized to obtain a normalized matrix. The normalized matrix of each level is multiplied by the corresponding weight to obtain a weighted supermatrix. The geometric mean of the weighted supermatrix is ​​calculated and subjected to limiting processing to obtain a limiting supermatrix. The limiting supermatrix is ​​combined with the control layer weight to calculate the global priority of each evaluation indicator and obtain the subjective weight of the indicator after priority sorting. Based on the consistency test algorithm, the consistency ratio of the subjective weight of the indicator is calculated, and the comparison value of the comparison matrix is ​​corrected to ensure that the consistency CR of the comparison matrix is ​​less than 0.1, and the final subjective weight of the indicator is determined.

4. The method according to claim 1, characterized in that The subjective evaluation index and the objective evaluation index are standardized, the proportion of the subjective evaluation index and the objective evaluation index is counted, and the subjective evaluation index and the objective evaluation index are corresponded to a determined unified score according to the proportion to form standardized numerical data, which specifically includes: Select the same measurement method, determine the score range and standard scoring; analyze the collected subjective evaluation indicators and objective evaluation indicators, and calculate the cumulative proportion of the subjective evaluation indicators and the objective evaluation indicators; according to the cumulative proportion, formulate standardization rules, map the subjective evaluation indicators and objective evaluation indicators corresponding to the cumulative proportion to the determined standardized scores, and form standardized numerical data.

5. The method according to claim 1, characterized in that The entropy weight method is used to calculate the standardized numerical data to obtain the objective weight of the indicator, and the objective weight of the indicator is combined with the subjective weight of the final indicator to obtain the combined weight, which specifically includes: The entropy value and its corresponding entropy weight of the standardized numerical data are calculated using the entropy weight method to obtain the objective weight of the indicator; the sensitivity coefficient is set, and the objective weight of the indicator is combined with the final subjective weight of the indicator to obtain the combined weight.

6. The method according to claim 1, characterized in that Step 4 specifically includes: Collect the measurement data of the evaluation index, preset the number of clusters of each evaluation index based on the hierarchical clustering algorithm, and define the evaluation grade standard of each evaluation index; initialize the collected measurement data corresponding to each evaluation index into a cluster, take each cluster as a class, use the single-linkage clustering method, calculate the minimum distance between each other, merge the two classes with the smallest distance into a new class, and recalculate the distance between the new class and all classes until the number of clusters is reached, correspond the new class obtained by each evaluation index to different evaluation grades based on the evaluation grade standard, and construct the index evaluation standard; According to the complexity of the tobacco enterprise information system, the corresponding normal cloud parameters are selected, and the expected Ex, entropy En and super entropy He of each evaluation index are calculated according to formula 1, formula 2 and formula 3 respectively. A normal cloud model is constructed for each evaluation index based on the normal cloud parameters. A certain number of cloud droplets are generated for each index using the normal distribution function. For each cloud droplet, its membership at each evaluation level is calculated, that is, the cloud droplet value corresponds to the evaluation level membership matrix of each evaluation level. The cloud droplet of each index is multiplied by the corresponding evaluation level membership matrix to obtain the membership of each index at each evaluation level. The cloud model of each evaluation index is cross-validated, and the expected Ex, entropy En and super entropy He of the normal cloud parameters are adjusted and optimized. Finally, the accurate membership value of each evaluation index at each evaluation level is obtained: Formula 1; Formula 2: Formula 3: in, represents the i-th collected value in a single indicator data set, and n represents the total number of samples; It is the probability density function of the i-th value in the single index X set based on the Gaussian fuzzy factor, b represents the logarithmic base, usually b=2.

7. The method according to claim 1, characterized in that The step 5 specifically includes: Based on the precise membership value and the combined weight, the precise membership value of each evaluation index is multiplied by the combined weight using a weighted average operation to obtain a weighted membership, and the weighted memberships of all evaluation levels are added together to obtain an overall evaluation level of the tobacco enterprise information system.

8. A tobacco enterprise information system evaluation and analysis system, characterized in that: Specifically include: The first weight calculation module is used to collect evaluation indicators of the tobacco enterprise information system, construct an evaluation indicator matrix using a network analysis method, calculate the subjective weight of the indicator based on the evaluation indicator matrix, and perform consistency test analysis on the subjective weight of the indicator to obtain the final subjective weight of the indicator, wherein the evaluation indicator includes: subjective evaluation indicators and objective evaluation indicators; The first processing module is used to standardize the subjective evaluation index and the objective evaluation index, calculate the proportion of the subjective evaluation index and the objective evaluation index, and correspond the subjective evaluation index and the objective evaluation index to a determined unified score according to the proportion to form standardized numerical data; A second weight calculation module is used to calculate the standardized numerical data using an entropy weight method to obtain an objective weight of the indicator, and combine the objective weight of the indicator with the final subjective weight of the indicator to obtain a combined weight; A normal cloud model module is used to construct an evaluation matrix of a tobacco enterprise information system based on a hierarchical clustering algorithm to obtain an indicator evaluation standard including an evaluation grade; determine normal cloud model parameters for each evaluation index, construct an improved normal cloud model based on the normal cloud model parameters, generate a certain number of cloud droplets for each evaluation index according to the constructed improved normal cloud model, calculate the membership of each cloud droplet at each evaluation grade, that is, the cloud droplet value corresponds to the evaluation grade membership matrix of each evaluation grade, multiply the cloud droplet of each evaluation index by the corresponding evaluation grade membership matrix, obtain the membership of each evaluation index at each evaluation grade, cross-validate the normal cloud model of each evaluation index, adjust and optimize the corresponding normal cloud model parameters, and finally obtain the accurate membership value of each evaluation index at each evaluation grade; The second processing module is used to comprehensively process the evaluation results of each evaluation index based on the precise membership value and the combined weight using weighted average operation, normalize the evaluation results after weighted summation, and finally obtain the overall evaluation level of the tobacco enterprise information system.

9. An electronic device, characterized in that: include: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program, when executed by the processor, implements the steps of the tobacco enterprise information system evaluation and analysis method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores an implementation program for information transmission, and when the program is executed by a processor, the steps of the tobacco enterprise information system evaluation and analysis method according to any one of claims 1 to 7 are implemented.

Images