Enterprise object evaluation method and device, electronic equipment and medium

Abstract

Embodiments of the invention provide an enterprise object evaluation method and apparatus, an electronic device and a medium. The method comprises the steps of constructing a basic data set of an enterprise object; wherein the basic data set comprises communication related data; determining index values of a plurality of index dimensions according to the basic data set; for each index dimension, clustering the index values of the plurality of enterprise objects to obtain a plurality of clusters, and determining the index weight of each index value according to the average contour coefficient of each cluster; and determining an evaluation score of each enterprise object in combination with the index values and the index weights. According to the embodiment of the invention, the method achieves the evaluation of the enterprise through the big data, combines the communication related data of the enterprise with a clustering algorithm, and improves the accuracy and comprehensiveness of the evaluation of the enterprise.

Description

Technical Field

[0001] The present invention relates to the field of big data technology, and in particular to an enterprise object evaluation method, device, electronic equipment and medium. Background Art

[0002] Enterprise growth assessment is an analysis and evaluation of a company's future development potential and expansion capabilities. This assessment is crucial for both the market economy and the company's own development. However, existing technologies often rely on manual evaluation, while others rely on algorithms to establish fixed strategies, resulting in inaccurate and incomplete assessments. Summary of the Invention

[0003] In view of the above problems, a method, device, electronic device and medium for evaluating an enterprise object are proposed to overcome or at least partially solve the above problems, including:

[0004] A method for evaluating an enterprise object, the method comprising:

[0005] Constructing a basic data set of enterprise objects; wherein the basic data set includes communication-related data;

[0006] Determining indicator values ​​of multiple indicator dimensions based on the basic data set;

[0007] For each indicator dimension, cluster the indicator values ​​of multiple enterprise objects to obtain multiple clusters, and determine the indicator weight of each indicator value based on the average silhouette coefficient of each cluster;

[0008] The evaluation score of each enterprise object is determined by combining the indicator value and the indicator weight.

[0009] Optionally, for each indicator dimension, the indicator values ​​of multiple enterprise objects are clustered to obtain multiple clusters, including:

[0010] For each indicator dimension, a target number of clusters is determined, and indicator values ​​of multiple enterprise objects are clustered according to the target number of clusters to obtain multiple clusters of the indicator dimension.

[0011] Optionally, determining the target number of clusters for each indicator dimension includes:

[0012] For each indicator dimension, determine the intra-cluster sum of squares under multiple candidate cluster numbers;

[0013] Determining a first-order difference sequence under the plurality of candidate clusters according to the intra-cluster square sum;

[0014] According to the first-order difference sequence, a target cluster number is determined from the multiple candidate cluster numbers.

[0015] Optionally, determining the target number of clusters from the multiple candidate numbers of clusters according to the first-order difference sequence includes:

[0016] Obtaining a threshold, and determining the number of candidate clusters whose first-order difference sequence is less than or equal to the threshold;

[0017] From the candidate cluster numbers whose first-order difference sequences are less than or equal to a preset threshold, the candidate cluster number with the smallest cluster number is determined as the target cluster number.

[0018] Optionally, before determining the indicator weight of each indicator value according to the average silhouette coefficient of each cluster, the method further includes:

[0019] For each cluster, determine the intra-cluster cohesion and inter-cluster separation;

[0020] An average silhouette coefficient is determined based on the intra-cluster cohesion and the inter-cluster separation.

[0021] Optionally, before clustering the indicator values ​​of multiple enterprise objects for each indicator dimension to obtain multiple clusters, the method further includes:

[0022] Discretize the indicator value of the enterprise object.

[0023] Optionally, combining the indicator value and the indicator weight to determine the evaluation score of each enterprise object includes:

[0024] For each indicator dimension, the indicator value is weighted using the indicator weight to obtain the indicator score;

[0025] According to the indicator scores of multiple indicator dimensions, the evaluation score of the enterprise object is obtained.

[0026] Optionally, the multiple indicator dimensions include indicator dimensions related to the communication-related data, including any one or more of the following:

[0027] Communication activity, communication service index, and market coverage.

[0028] Optionally, the basic data set further includes enterprise-related information, and the multiple indicator dimensions include indicators related to the enterprise-related information, including any one or more of the following:

[0029] Total operating income growth rate, total assets growth rate, total equity investment amount, return on net assets, net profit margin of total assets, cost and expense growth rate, debt-to-asset ratio, current ratio, quick ratio, net profit growth rate, total assets growth rate, and equity growth rate.

[0030] An evaluation device for an enterprise object, comprising:

[0031] A basic data set construction module is used to construct a basic data set of an enterprise object; wherein the basic data set includes communication-related data;

[0032] An indicator value determination module, configured to determine indicator values ​​of multiple indicator dimensions based on the basic data set;

[0033] An indicator weight determination module is used to cluster the indicator values ​​of multiple enterprise objects for each indicator dimension to obtain multiple clusters, and determine the indicator weight of each indicator value based on the average silhouette coefficient of each cluster;

[0034] The evaluation score determination module is used to determine the evaluation score of each enterprise object by combining the indicator value and the indicator weight.

[0035] An electronic device includes a processor, a memory, and a computer program stored in the memory and capable of running on the processor, wherein the computer program implements the method described above when executed by the processor.

[0036] A computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method described above is implemented.

[0037] The embodiments of the present invention have the following advantages:

[0038] In an embodiment of the present invention, a basic data set of enterprise objects is constructed, wherein the basic data set includes communication-related data. Based on the basic data set, indicator values ​​of multiple indicator dimensions are determined. For each indicator dimension, the indicator values ​​of multiple enterprise objects are clustered to obtain multiple cluster clusters. Based on the average silhouette coefficient of each cluster cluster, the indicator weight of each indicator value is determined. Combining the indicator value and the indicator weight, the evaluation score of each enterprise object is determined, thereby realizing the use of big data to evaluate the enterprise. Not only the communication-related data of the enterprise but also the clustering algorithm are combined, thereby improving the accuracy and comprehensiveness of the evaluation of the enterprise. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] In order to more clearly illustrate the technical solution of the present invention, the following briefly introduces the drawings required for use in the description of the present invention. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0040] Figure 1is a flowchart of steps of an enterprise object evaluation method provided by some embodiments of the present invention;

[0041] Figure 2 is a schematic diagram of a system architecture provided by some embodiments of the present invention;

[0042] Figure 3 is a flowchart of steps of another enterprise object evaluation method provided by some embodiments of the present invention;

[0043] Figure 4 This is a structural block diagram of an enterprise object evaluation device provided by some embodiments of the present invention. DETAILED DESCRIPTION

[0044] To make the above-mentioned objects, features, and advantages of the present invention more clearly understood, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the embodiments described are only a portion of the embodiments of the present invention, not all of them. All other embodiments derived by persons of ordinary skill in the art based on the embodiments of the present invention without inventive effort are also within the scope of protection of the present invention.

[0045] With the sustainable development of an open economy, the market plays a decisive role in resource allocation. How enterprises adapt to the market environment, optimize resource allocation, and clarify their development direction has become a crucial issue for their long-term sustainable development and self-reform. Enterprises conducting growth capability assessments based on their current status and the situation of existing companies in the industry can help them identify their strengths and weaknesses, understand their competitive position within the industry, and provide a basis for formulating long-term development strategies. Furthermore, the growth capability assessment process can also assist in analyzing market competitiveness. A positive growth capability assessment can enhance a company's brand image and credit rating, attract the confidence of investors and financial institutions, reduce financing costs, and increase market recognition. In summary, enterprise growth capability assessments are not only of far-reaching significance for the company's own development but also have significant implications for the macroeconomic development and microeconomic regulation of the economic market.

[0046] One related technology assesses enterprise growth capacity by leveraging data such as business registration changes, combined with housing price data around points of interest (POIs) in the area, to generate a growth score and assess the enterprise's growth coefficient. However, this approach utilizes limited data dimensions and cannot fully and effectively reflect the capabilities of key aspects of an enterprise, such as scale and technological innovation. Furthermore, due to the frequent relocation of enterprises, timely identifying a company's actual operating address through its business registration address presents significant challenges.

[0047] In another related technology, a supervised learning model is built to extract the importance of indicators, and the indicator values ​​and indicator weights are used to evaluate the company's growth capabilities. In addition, internal company data such as projects, R&D, and financial loans are added to the data dimensions. Although internal company data dimensions such as company projects and R&D investment have been added, the data in these dimensions involve objective factors such as the data privacy and business secrets of the evaluated company, and are difficult to obtain. In addition, using "total project amount" as a prediction label to construct a supervised learning model requires manual classification and sorting of the "total project amount". At the same time, using "total project amount" as a prediction target cannot comprehensively evaluate the company's growth capabilities based on multiple factors.

[0048] In another related technology, the hierarchical analysis method is used to construct evaluation indicators for the enterprise technology dimensions, and a technology transformation evaluation analysis table is used to achieve a quantitative assessment of the enterprise technology. However, it only focuses on evaluating the enterprise's technological growth capabilities from the perspective of enterprise scientific and technological research, and cannot comprehensively and effectively evaluate the enterprise's growth capabilities based on multiple factors. In addition, the hierarchical analysis method relies on business experts to construct a judgment matrix. Although consistency testing is required, it introduces human errors, which will ultimately affect the effectiveness of the enterprise growth evaluation model.

[0049] In another related technology, we focus on the analysis of cross-border e-commerce enterprises and industries. With the theory of enterprise growth as the theoretical basis, based on the financial data of listed companies and factor analysis, we construct a growth evaluation index system suitable for export cross-border e-commerce enterprises, and finally conduct enterprise cluster analysis based on the comprehensive growth ability indicators to explore the reasons why export cross-border e-commerce has high growth. We only focus on listed export cross-border e-commerce enterprises and mainly evaluate the growth ability of enterprises based on the financial information of statistical enterprises.

[0050] In the embodiments of the present invention, relevant data information can be provided for enterprise self-positioning and market insights, and data support can be provided for decision-making innovation, mainly including the following points:

[0051] 1. In terms of data sources, by increasing the communication behavior data of enterprise employees, taking the internal communication big data behavior of employees as an indicator, and checking the business address of the enterprise, it is possible to correct abnormal data such as the difference between the business address of the enterprise and the actual office address during the operation of the enterprise, and the failure to report to the government agency in time after the enterprise relocation behavior.

[0052] 2. Using the communication behavior and spatiotemporal big data information of operators, we analyze the communication service usage and business coverage during the communication process of enterprise employees, characterize the trajectories of enterprise personnel during work, construct a set of communication activity, communication service index and market coverage indicators, and form a model of the relationship between enterprise and industry layout, thereby achieving effective quantification of the business spatial distribution status and business scale in the daily business activities of enterprises.

[0053] 3. To systematically and comprehensively design a system for assessing enterprise growth capabilities, we employed the unsupervised learning k-means algorithm to discretize indicators, controlling the impact of indicator dimensions on scoring results. We also pioneered the use of the silhouette coefficient to extract key insights into indicator importance. Furthermore, we introduced an adaptive mechanism based on a first-order difference threshold to dynamically select the optimal number of clusters. This reduces the influence of subjective judgment on model results, improves the model's reliability and generalization capabilities, and enables it to provide effective insights and predictions across a wide range of application scenarios.

[0054] Specifically, the main advantages of the embodiments of the present invention are as follows:

[0055] 1. Rich, high-quality data sources: This model leverages the unique communications big data of carriers, incorporating enterprise information such as business registration, financial information, and asset information, expanding data dimensions to enhance the accuracy and applicability of the model. Carrier communications data is massive, real-time, continuous, and multi-dimensional, encompassing comprehensive data on user communication behavior, spatiotemporal trajectories, network bandwidth, and other dimensions. Compared to other data sources, carrier data can be updated at a near-real-time frequency and maintains data continuity, thus demonstrating enterprise changes across spatiotemporal dimensions. This effectively reduces the issue of inaccurate enterprise growth capability assessments due to timeliness issues, and makes the construction of an enterprise growth capability assessment system more accurate.

[0056] 2. Stronger model generalization and applicability: In terms of model design, the unsupervised k-means algorithm is used to discretize continuous indicator variables, and the silhouette coefficient is used to extract the importance of key indicators. This method does not require pre-labeled data, thereby enhancing the model's adaptability to new situations.

[0057] In addition, an adaptive method based on the first-order difference threshold is introduced to dynamically select the optimal number of clusters. The adaptive mechanism significantly reduces the impact of human subjective judgment on the model results, ensuring the generalization of the model in different data sets and application scenarios.

[0058] 3. Big Data Analysis Advantages: By integrating multi-dimensional data indicators and fully considering the growth capabilities of enterprises in different scenarios, we can perform in-depth data mining and comprehensive data processing tasks on multi-dimensional data indicators of different types of enterprises. This greatly expands the application scope of the model, making it adaptable to diverse business needs and complex data analysis scenarios. Through efficient processing of big data, we not only improve insight into information, but also enhance flexibility and applicability in practical applications, ensuring the depth and breadth of analytical results.

[0059] The present invention is further described below with reference to the accompanying drawings:

[0060] Reference Figure 1 , which shows a flowchart of a method for evaluating an enterprise object provided by some embodiments of the present invention, which may specifically include the following steps:

[0061] Step 101: construct a basic data set of enterprise objects; wherein the basic data set includes communication-related data.

[0062] In some embodiments of the present invention, the basic data set also includes enterprise-related information.

[0063] In actual applications, enterprise-related information and communication-related data of enterprise objects can be collected, and then a basic data set of the enterprise objects can be constructed based on the enterprise-related information and communication-related data. This allows the operator's spatiotemporal data to be used as the basis for constructing enterprise capability indicators, and combines the enterprise's industrial and commercial information and enterprise management information to form a multi-source data indicator set for enterprise capability assessment.

[0064] By comprehensively collecting and integrating enterprise-related information such as corporate industrial and commercial information, annual reports, corporate structure, corporate legal and asset mortgage information, and corporate activity information disclosed on the Internet, and based on enterprise-related information and operator data (i.e., communication-related data), a basic data set for enterprise growth capability assessment is constructed.

[0065] Communication-related data, including operator data, such as user identification information, location information, and user behavior, has unique characteristics such as real-time, massive, continuous, and multi-dimensional. Through communication behavior data such as corporate call frequency, Internet access time, and short message sending and receiving, we can real-time explore changes in the company's business activities on different timelines.

[0066] Furthermore, by analyzing the statistics and changes to enterprise employees' communication devices, combined with spatial communication data such as the enterprise's communication base stations, business locations, and network broadband addresses, we can spatially explore the actual location of an enterprise's operations and production, as well as the scope of its business reach. Therefore, operator data can reveal the characteristics of an enterprise's actual business activities from both temporal and spatial dimensions.

[0067] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with the relevant laws, regulations and standards of relevant countries and regions, and provide corresponding operation entrances for users to choose to authorize or refuse.

[0068] For enterprise-related data, the enterprise industrial and commercial registration information in the enterprise industry and commerce analyzes whether the enterprise situation is abnormal from the dimensions of enterprise establishment time, enterprise status, enterprise violations, etc.; information such as enterprise annual reports, enterprise social security payment status and enterprise activities can fully explore the enterprise's business dynamics and the industry type to which the main business belongs; enterprise assets such as shareholder information, capital contribution ratio and capital contribution amount in the enterprise structure and asset mortgage can show the financial and macro-asset status of the enterprise.

[0069] Step 102: Determine the indicator values ​​of multiple indicator dimensions based on the basic data set.

[0070] After obtaining the basic data set of the enterprise object, the indicator values ​​of the corresponding indicator dimensions can be calculated according to the pre-set multiple indicator dimensions using pre-defined rules.

[0071] In some embodiments of the present invention, the multiple indicator dimensions include indicator dimensions related to the communication-related data, including any one or more of the following:

[0072] Communication activity, communication service index, and market coverage.

[0073] In some embodiments of the present invention, the multiple indicator dimensions include indicators related to the enterprise-related information, including any one or more of the following: total operating income growth rate, total assets growth rate, total equity investment amount, return on net assets, net interest rate of total assets, cost growth rate, debt-to-asset ratio, current ratio, quick ratio, net profit growth rate, total assets growth rate, and equity growth rate.

[0074] In practical applications, the basic data set for enterprise growth capability assessment is constructed, such as Figure 2 , from the five dimensions of enterprise operation ability, profitability, debt capacity, development ability and communication ability, the enterprise growth ability evaluation index is constructed, as shown in Table 1 below:

[0075]

[0076] Table 1

[0077] For enterprise-related information, indicators related to enterprise-related information are constructed as follows:

[0078] In some examples, the basic data set can be screened to eliminate corporate data with abnormal operations such as deregistration and serious violations, in order to construct an accurate basic enterprise data set.

[0079] 1. Construct operational capacity indicators based on corporate business data

[0080] The operating capacity of an enterprise refers to its ability to effectively use its resources and assets to generate income and cash flow when conducting daily business activities. This ability reflects the efficiency and effectiveness of the enterprise in managing and using its assets. By selecting the growth rate of total operating income, the growth rate of total assets, and the total amount of equity investment as operating capacity indicators, the details are as follows:

[0081] (1) Total operating revenue growth rate: The total operating revenue growth rate is used to measure the growth rate of an enterprise's total operating revenue over a certain period of time, reflecting changes in the enterprise's sales capacity and market expansion capabilities.

[0082] (2) Total asset growth rate: The total asset growth rate can measure the growth rate of a company's asset scale over a certain period of time and reflect the expansion of the company's total assets. A higher total asset growth rate can indicate that the company has greater growth potential and market opportunities.

[0083] (3) Total equity investment: The total equity investment refers to the cumulative amount of equity investment made by an enterprise in a certain period of time. It reflects the scale of an enterprise's overseas investment and its expansion capacity. The investment structure of an enterprise shows the degree of diversification of its investment and the composition of its investment portfolio.

[0084] 2. Constructing profitability indicators based on corporate business data

[0085] Profitability of an enterprise refers to its ability to generate profits through its operating and investment activities. It is a key indicator for measuring the financial health, efficiency, and competitiveness of an enterprise. Return on equity, net profit margin on total assets, and cost-expense growth rate are selected as profitability indicators, as follows:

[0086] (1) Return on equity: Return on equity reflects the net income obtained by the company from the use of shareholders' equity, and also indicates the efficiency of the company's management in using the capital invested by shareholders.

[0087] (2) Net profit margin: Net profit margin can be used to quantify the ability of a company's total assets to generate net profits. It reflects the profitability of a company's assets, and its changing trend can be used as an indicator to assess the company's financial health.

[0088] (3) Cost and expense growth rate: The cost and expense growth rate is a measure of the percentage increase in an enterprise's costs and expenses over a certain period of time. It reflects the effectiveness of the enterprise's cost control and the growth of the enterprise's costs in the process of expanding production or business.

[0089] 3. Constructing a creditor capacity indicator based on corporate industrial and commercial data

[0090] The creditor capacity of an enterprise refers to its ability to fulfill its debt obligations, that is, its ability to repay short-term and long-term debts. It is directly related to the enterprise's credit status, financial stability and ability to continue operations. The debt-to-asset ratio, current ratio and quick ratio are selected as creditor capacity indicators, as follows:

[0091] (1) Debt-to-asset ratio: The debt-to-asset ratio can measure the financial leverage and debt repayment ability of an enterprise. It represents the proportion of the enterprise's total liabilities to its total assets and reflects the capital structure of the enterprise, that is, the degree to which the enterprise relies on debt financing and equity financing.

[0092] (2) Current Ratio: The current ratio is an important financial indicator that measures a company's short-term debt repayment ability. It indicates the company's ability to repay its current liabilities with its current assets. The current ratio can be used to assess a company's creditworthiness.

[0093] (3) Quick ratio: The quick ratio is another financial indicator that measures a company's short-term debt repayment ability. Compared with the current ratio, the quick ratio excludes current assets such as inventory and only considers assets that are most easily converted into cash.

[0094] 4. Constructing development capability indicators based on enterprise business data

[0095] An enterprise's development capability refers to its ability to achieve sustained growth and expansion. It is an important reflection of an enterprise's long-term competitiveness and market position. By selecting net profit growth rate, total asset growth rate, and equity growth rate as development capability indicators, the details are as follows:

[0096] (1) Net profit growth rate: Net profit growth rate can quantify the growth rate of a company's net profit over a certain period of time, reflecting the company's profitability and growth potential. It is also a key indicator for judging a company's profit prospects and asset value.

[0097] (2) Total asset growth rate: The total asset growth rate is a quantitative indicator that quantifies the growth rate of a company's asset size over a certain period of time. It shows the company's expansion capabilities and its competitive advantage in the market. The total asset growth rate is an important indicator for judging a company's growth potential and investment value.

[0098] (3) Equity growth rate: Equity growth rate refers to the rate of growth of a company's shareholders' equity, reflecting the company's ability to increase shareholder value through profit accumulation, capital injection, or other means. The shareholder growth rate is an important indicator of a company's growth and financial health.

[0099] For communication-related data, indicators related to communication-related data are constructed as follows:

[0100] Operator communication big data covers a wealth of communication behavior and spatiotemporal data. This data dimension can be further expanded to construct communication behavior indicators to assess enterprise growth capabilities from the perspective of communication capabilities. Communication activity, communication service index, and market coverage are selected as communication capability indicators, as follows:

[0101] (1) Communication activity: Communication activity is constructed through analysis of communication frequency, communication market, and communication peak. It can reflect the frequency of corporate communication activities and communication usage within a specific period of time, and further evaluate the market activity and business expansion capabilities of the enterprise.

[0102] (2) Communication service index: The communication service index is a data indicator constructed based on the use of operator digital services and enterprise communication peak analysis. Operator digital services include digital products such as industry digital services, cloud services, and mobile communication services handled by enterprises. By quantitatively counting the number of services and the value of services provided by the operator digital services used by enterprises, the business operation volume and enterprise service diversity of the enterprise can be analyzed. Communication peak analysis refers to analyzing the communication peak time period and frequency of the enterprise to be evaluated by comparing it with other enterprises in the same industry, and then normalizing and constructing the communication peak index. The communication service index is obtained by linearly summing the number of services, service value, and enterprise communication peak index of the enterprise. The communication service index can reflect the service diversity and market share of the enterprise to a certain extent, and can also reflect the competitiveness of the enterprise in the industry.

[0103] (3) Market coverage: Market coverage is the analysis of the geographic scope and statistical coverage of a company's business activities, extracted from its actual business address, broadband communication behavior, and base station interaction records, to analyze the breadth and depth of the company's market coverage within the region. Market coverage can reflect the company's service accessibility and market penetration, as well as the company's distribution and expansion within a certain area.

[0104] Step 103 : clustering the indicator values ​​of multiple enterprise objects for each indicator dimension to obtain multiple clusters, and determining the indicator weight of each indicator value according to the average silhouette coefficient of each cluster.

[0105] After obtaining the indicator values ​​of multiple indicator dimensions, the indicator values ​​can be clustered for each indicator dimension to obtain multiple clusters of the indicator dimensions. That is, the enterprise object has multiple indicator dimensions, each indicator dimension belongs to a certain cluster, different indicator dimensions belong to different clusters, and the same cluster includes the indicator values ​​of the same indicator dimension of different enterprise objects.

[0106] After obtaining the clusters, the average silhouette coefficient of each cluster can be determined through analysis, and then the average silhouette coefficient can be used to determine the indicator weight of the indicator value in the cluster.

[0107] As an example, a k-means clustering algorithm may be used for clustering.

[0108] In practical applications, the indicator values ​​can be standardized first, and then an adaptive k-means clustering algorithm is designed to analyze and quantify the enterprise growth capability score and extract the indicator importance.

[0109] The main goal of the k-means algorithm is to effectively divide n sample data into k clusters, where each sample is assigned to the cluster represented by its nearest mean (i.e., cluster center). This process aims to minimize the variance within the cluster to ensure that the samples within the cluster have high similarity.

[0110] In some examples, the k-means clustering method uses Euclidean distance as a metric for sample similarity. Features of different dimensions can seriously affect the final distance calculation results. Therefore, z-score normalization can be used to unify the dimensions of the features in the evaluation indicator set.

[0111] For example, in symbolic form x i ,i=1,2,...,15 represents all features in the evaluation index set, and the mean of the corresponding features is u i ,i=1,2,…,15, standard deviation is σ i ,i=1,2,…,15, then the feature normalization formula is as follows:

[0112]

[0113] In some embodiments of the present invention, before clustering the indicator values ​​of multiple enterprise objects for each indicator dimension to obtain multiple clusters, the method further includes:

[0114] Discretize the indicator value of the enterprise object.

[0115] In the embodiment of the present invention, the sample data mainly refers to a series of evaluation indicators, most of which are continuous variables. Although the indicator values ​​of different enterprises may fluctuate to a certain extent, they actually need to reflect similar corporate growth potential. In order to ensure that the evaluation indicators can fairly reflect this similarity between enterprises, the embodiment of the present invention adopts k-means for discretization processing.

[0116] For example, 60 points is used as the dividing line between excellent and poor, but in fact, 60 points and 59.5 points may be almost the same. Through discretization processing, these data can be discretized more scientifically to make the distinction between classes greater, instead of simply dividing 59.5 points and 60 points into two different clusters.

[0117] In the embodiment of the present invention, the discretization process not only balances the impact of different numerical dimensions, but also enhances the consistency and comparability of the evaluation indicators in expressing the growth capabilities of enterprises.

[0118] In some embodiments of the present invention, clustering the indicator values ​​of multiple enterprise objects for each indicator dimension to obtain multiple clusters includes:

[0119] For each indicator dimension, a target number of clusters is determined, and indicator values ​​of multiple enterprise objects are clustered according to the target number of clusters to obtain multiple clusters of the indicator dimension.

[0120] In practical applications, the optimal number of clusters (i.e., the number of target clusters) can be dynamically selected by introducing an adaptive method based on the first-order difference threshold. The adaptive mechanism significantly reduces the impact of human subjective judgment on the model results and ensures the generalization of the model in different data sets and application scenarios.

[0121] In some embodiments of the present invention, determining the target number of clusters for each indicator dimension includes: determining, for each indicator dimension, the intra-cluster sum of squares under multiple candidate numbers of clusters; determining, based on the intra-cluster sum of squares, a first-order difference sequence under the multiple candidate numbers of clusters; and determining, based on the first-order difference sequence, the target number of clusters from the multiple candidate numbers of clusters.

[0122] In practical applications, for each indicator dimension, we can try to perform clustering under the number of candidate clusters, and then determine the intra-cluster sum of squares under the number of candidate clusters based on the clustering results.

[0123] After obtaining the intra-cluster sum of squares, the first-order difference sequence under the number of candidate clusters can be calculated based on the intra-cluster sum of squares, and then the target number of clusters can be determined based on the first-order difference sequence.

[0124] In some embodiments of the present invention, determining the target number of clusters from the multiple candidate cluster numbers based on the first-order difference sequence includes: obtaining a threshold and determining the number of candidate clusters for which the first-order difference sequence is less than or equal to the threshold; and determining the candidate cluster number with the smallest number of clusters as the target number of clusters from the candidate cluster numbers for which the first-order difference sequence is less than or equal to a preset threshold.

[0125] In practical applications, the k-means algorithm itself does not provide a direct method to determine the optimal number of clusters (i.e., the k value). It is determined manually by calculating the within-cluster sum of squares (WCSS) and using the elbow rule.

[0126] However, this method relies on plotting the WCSS as the value of k increases, searching for the point where the rate of WCSS decreases significantly. In some cases, when the change in silhouette coefficient with k is not obvious, the elbow rule may be difficult to apply accurately, making the choice of k value susceptible to subjective judgment.

[0127] In order to overcome this limitation, an adaptive k-means algorithm is proposed, which uses the first-order difference threshold to adaptively determine the k value, thereby assisting in determining the optimal number of clusters. It can not only dynamically discretize the evaluation indicators, but also improve the accuracy and reliability of the k value determination.

[0128] In practical applications, an appropriate threshold Th is set based on data characteristics. For example, if the difference of the first-order difference value is expected to be 80%, the threshold is set to 80%.

[0129] By trying multiple different numbers of candidate clusters, and calculating the index x i WCSS and first-order difference sequence under different numbers of candidate clusters and identify the sequence The minimum number of candidate clusters that does not exceed the threshold Th is used as the index x i The clustering k value (i.e. the number of target clusters) is calculated as follows:

[0130]

[0131] k is the number of clusters, C i represents all samples in the i-th cluster, c i represents the centroid of the i-th cluster.

[0132]

[0133] in, represents a first-order difference sequence.

[0134]

[0135] Among them, Th represents the set threshold, and K represents the number of final clusters, that is, the k value.

[0136] In some embodiments of the present invention, before determining the indicator weight of each indicator value based on the average silhouette coefficient of each cluster, the method further includes: determining the intra-cluster cohesion and inter-cluster separation for each cluster; and determining the average silhouette coefficient based on the intra-cluster cohesion and the inter-cluster separation.

[0137] In an embodiment of the present invention, an unsupervised k-means algorithm is used to discretize continuous indicator variables, and the silhouette coefficient is used to extract the importance of key indicators. This method does not require pre-labeled data, thereby enhancing the adaptability of the model to new situations.

[0138] In practical applications, the indicator x is used i The average silhouette coefficient S of the corresponding cluster i i As the indicator weight, the enterprise growth ability is evaluated. Among them, the silhouette coefficient is an indicator for evaluating the quality of sample clustering, which can comprehensively represent the indicator x i The discretized indicator weights.

[0139] The silhouette coefficient consists of two parts: intra-cluster cohesion a(i) and inter-cluster separation b(i), which refer to the distance between a sample and other samples in its cluster and the distance between samples in the nearest cluster, respectively. It can effectively quantify the similarities and differences in growth capabilities between companies. The calculation formula is as follows:

[0140]

[0141]

[0142] Among them, s(i) is the silhouette coefficient of sample point i, a(i) represents the cohesion within the cluster, b(i) represents the separation between clusters, S i represents the average silhouette coefficient of cluster i, and N represents the number of sample points in cluster i.

[0143] Step 104: Determine the evaluation score of each enterprise object by combining the indicator value and the indicator weight.

[0144] In some examples, the assessment score is an assessment score of the enterprise's growth capability.

[0145] After obtaining the indicator value and the average silhouette coefficient of the cluster where the indicator value belongs, the average silhouette coefficient can be used as the indicator weight to weight the indicator value to obtain the evaluation score of each enterprise object.

[0146] In some embodiments of the present invention, combining the indicator value and the indicator weight to determine the evaluation score of each enterprise object includes:

[0147] For each indicator dimension, the indicator value is weighted using the indicator weight to obtain an indicator score; based on the indicator scores of multiple indicator dimensions, an evaluation score of the enterprise object is obtained.

[0148] In practical applications, we can first calculate the indicator weights, set the threshold Th based on the data characteristics, and then evaluate the evaluation indicators x i , i=1,2,…,15, use the adaptive k-means algorithm to perform discretization and obtain the evaluation index x i The number of clusters K and the evaluation index x of each enterprise i The corresponding cluster class j is Denotes the evaluation index x i The actual indicator value corresponding to cluster j.

[0149] Then, the evaluation index x i , i=1,2,…,15, the silhouette coefficient s(i) of the N companies in the j-th cluster, and then calculate the evaluation index x corresponding to the j-th cluster i The average silhouette coefficient That is, the evaluation index x i The weight after discretization (cluster j).

[0150] After obtaining the indicator weights, the indicator values ​​and indicator weights are used to evaluate the enterprise's growth capabilities.

[0151] In some examples, to unify the scoring mechanism and avoid the impact of abnormal values ​​on the scoring results, the sigmoid operator is used to normalize the scores and build a percentage scoring system for enterprise growth capability assessment. For example, if the enterprise growth capability assessment score is denoted as SE, the calculation formula is as follows:

[0152]

[0153] in, Represents the index x i The corresponding evaluation index weight of cluster j (i.e., index weight), is the actual value of the discretized evaluation index (i.e. the index value after discretization).

[0154] In an embodiment of the present invention, the growth capability of each enterprise is quantitatively calculated to obtain the evaluation score (Score of Enterprise, SE) of the quantitative evaluation of all enterprises. The larger the SE value of an enterprise, the better the growth capability and development prospect of the related enterprise, and the enterprise can be used as a key marketing development target.

[0155] In an embodiment of the present invention, a basic data set of enterprise objects is constructed, wherein the basic data set includes communication-related data. Based on the basic data set, indicator values ​​of multiple indicator dimensions are determined. For each indicator dimension, the indicator values ​​of multiple enterprise objects are clustered to obtain multiple cluster clusters. Based on the average silhouette coefficient of each cluster cluster, the indicator weight of each indicator value is determined. Combining the indicator value and the indicator weight, the evaluation score of each enterprise object is determined, thereby realizing the use of big data to evaluate the enterprise. Not only the communication-related data of the enterprise but also the clustering algorithm are combined, thereby improving the accuracy and comprehensiveness of the evaluation of the enterprise.

[0156] Specifically, a basic data set was constructed based on basic enterprise information and operator communication data. A comprehensive evaluation index system was constructed based on five key dimensions: operational capability, profitability, debt capacity, development capability, and communication capability. A threshold was set by in-depth analysis of data characteristics. This threshold was then used to discretize each evaluation indicator in the evaluation index system using an adaptive k-means algorithm, effectively quantifying and capturing the similarities and differences in enterprise growth capabilities. The evaluation indicator weights were obtained by calculating the average silhouette coefficient. Finally, the evaluation indicators and corresponding weights were combined to obtain the enterprise growth capability score (i.e., the evaluation score) through a comprehensive scoring formula.

[0157] The Enterprise Growth Capability Score not only quantifies a company's growth potential but also, through the calculated specific scores, allows for objective comparison and grouping of different companies' growth capabilities. Based on the above evaluation indicator system and algorithm design, an effective and objective enterprise growth capability assessment method based on communications big data is provided to evaluate and compare companies' growth capabilities and competitiveness in the market.

[0158] Reference Figure 3 , which shows a flowchart of another enterprise object evaluation method provided by some embodiments of the present invention, which may specifically include the following steps:

[0159] Step 301: construct a basic data set of enterprise objects; wherein the basic data set includes communication-related data.

[0160] Step 302: Determine the indicator values ​​of multiple indicator dimensions based on the basic data set.

[0161] Step 303: discretize the indicator value of the enterprise object.

[0162] Step 304 : For each indicator dimension, determine the target number of clusters, and cluster the indicator values ​​of multiple enterprise objects according to the target number of clusters to obtain multiple clusters of the indicator dimension.

[0163] Step 305: Determine the index weight of each index value according to the average silhouette coefficient of each cluster.

[0164] Step 306: Determine the evaluation score of each enterprise object by combining the indicator value and the indicator weight.

[0165] It should be noted that for the sake of simplicity, the method embodiments are described as a series of actions. However, those skilled in the art should be aware that the embodiments of the present invention are not limited by the order of the actions described, because according to the embodiments of the present invention, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in this specification are all preferred embodiments, and the actions involved are not necessarily required by the embodiments of the present invention.

[0166] Reference Figure 4 , which shows a schematic diagram of the structure of an enterprise object evaluation device provided by some embodiments of the present invention, which may specifically include the following modules:

[0167] A basic data set construction module 401 is used to construct a basic data set of an enterprise object; wherein the basic data set includes communication-related data;

[0168] An indicator value determination module 402 is configured to determine indicator values ​​of multiple indicator dimensions based on the basic data set;

[0169] The indicator weight determination module 403 is used to cluster the indicator values ​​of multiple enterprise objects for each indicator dimension to obtain multiple clusters, and determine the indicator weight of each indicator value based on the average silhouette coefficient of each cluster;

[0170] The evaluation score determination module 404 is configured to determine the evaluation score of each enterprise object by combining the indicator value and the indicator weight.

[0171] In some embodiments of the present invention, clustering the indicator values ​​of multiple enterprise objects for each indicator dimension to obtain multiple clusters includes:

[0172] For each indicator dimension, a target number of clusters is determined, and indicator values ​​of multiple enterprise objects are clustered according to the target number of clusters to obtain multiple clusters of the indicator dimension.

[0173] In some embodiments of the present invention, determining the target number of clusters for each indicator dimension includes:

[0174] For each indicator dimension, determine the intra-cluster sum of squares under multiple candidate cluster numbers;

[0175] Determining a first-order difference sequence under the plurality of candidate clusters according to the intra-cluster square sum;

[0176] According to the first-order difference sequence, a target cluster number is determined from the multiple candidate cluster numbers.

[0177] In some embodiments of the present invention, determining the target number of clusters from the plurality of candidate numbers of clusters according to the first-order difference sequence includes:

[0178] Obtaining a threshold, and determining the number of candidate clusters whose first-order difference sequence is less than or equal to the threshold;

[0179] From the candidate cluster numbers whose first-order difference sequences are less than or equal to a preset threshold, the candidate cluster number with the smallest cluster number is determined as the target cluster number.

[0180] In some embodiments of the present invention, before determining the indicator weight of each indicator value according to the average silhouette coefficient of each cluster, the method further includes:

[0181] For each cluster, determine the intra-cluster cohesion and inter-cluster separation;

[0182] An average silhouette coefficient is determined based on the intra-cluster cohesion and the inter-cluster separation.

[0183] In some embodiments of the present invention, before clustering the indicator values ​​of multiple enterprise objects for each indicator dimension to obtain multiple clusters, the method further includes:

[0184] Discretize the indicator value of the enterprise object.

[0185] In some embodiments of the present invention, combining the indicator value and the indicator weight to determine the evaluation score of each enterprise object includes:

[0186] For each indicator dimension, the indicator value is weighted using the indicator weight to obtain the indicator score;

[0187] According to the indicator scores of multiple indicator dimensions, the evaluation score of the enterprise object is obtained.

[0188] In some embodiments of the present invention, the multiple indicator dimensions include indicator dimensions related to the communication-related data, including any one or more of the following:

[0189] Communication activity, communication service index, and market coverage.

[0190] In some embodiments of the present invention, the basic data set further includes enterprise-related information, and the multiple indicator dimensions include indicators related to the enterprise-related information, including any one or more of the following:

[0191] Total operating income growth rate, total assets growth rate, total equity investment amount, return on net assets, net profit margin of total assets, cost and expense growth rate, debt-to-asset ratio, current ratio, quick ratio, net profit growth rate, total assets growth rate, and equity growth rate.

[0192] In an embodiment of the present invention, a basic data set of enterprise objects is constructed, wherein the basic data set includes communication-related data. Based on the basic data set, indicator values ​​of multiple indicator dimensions are determined. For each indicator dimension, the indicator values ​​of multiple enterprise objects are clustered to obtain multiple cluster clusters. Based on the average silhouette coefficient of each cluster cluster, the indicator weight of each indicator value is determined. Combining the indicator value and the indicator weight, the evaluation score of each enterprise object is determined, thereby realizing the use of big data to evaluate the enterprise. Not only the communication-related data of the enterprise but also the clustering algorithm are combined, thereby improving the accuracy and comprehensiveness of the evaluation of the enterprise.

[0193] Some embodiments of the present invention further provide an electronic device, comprising a processor, a memory, and a computer program stored in the memory and capable of running on the processor, wherein the computer program implements the above method when executed by the processor.

[0194] Some embodiments of the present invention further provide a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the above method is implemented.

[0195] Some embodiments of the present invention further provide a computer program product, including a computer program, which implements the above method when executed by a processor.

[0196] As for the device embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0197] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with the relevant laws, regulations and standards of relevant countries and regions, and provide corresponding operation entrances for users to choose to authorize or refuse.

[0198] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0199] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, apparatus, or computer program products. Thus, embodiments of the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, embodiments of the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0200] The embodiments of the present invention are described with reference to the flowcharts and / or block diagrams of the methods, terminal devices (systems), and computer program products according to the embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of the processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing terminal device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing terminal device generate instructions for implementing the process in the flowchart and / or block diagram. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0201] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing terminal device to operate in a specific manner, so that the instructions stored in the computer readable memory produce a manufactured product including an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0202] These computer program instructions can also be loaded onto a computer or other programmable data processing terminal device so that a series of operating steps are executed on the computer or other programmable terminal device to produce a computer-implemented process, thereby providing instructions for executing on the computer or other programmable terminal device to implement the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0203] Although the preferred embodiments of the present invention have been described, those skilled in the art may make additional changes and modifications to these embodiments once they become aware of the basic creative concepts. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the embodiments of the present invention.

[0204] Finally, it should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprises," "includes," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or terminal device that includes a series of elements includes not only those elements, but also other elements not explicitly listed, or elements inherent to such process, method, article, or terminal device. In the absence of further limitations, an element defined by the phrase "comprises a ..." does not exclude the presence of additional identical elements in the process, method, article, or terminal device that includes the above elements.

[0205] The above is a detailed introduction to the enterprise object evaluation method, device, electronic equipment and medium provided. Specific examples are used herein to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core idea. At the same time, for those skilled in the art, according to the ideas of the present invention, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting the present invention.

Claims

1. A method for evaluating an enterprise object, characterized in that: The method comprises: Constructing a basic data set of enterprise objects; wherein the basic data set includes communication-related data; Determining indicator values ​​of multiple indicator dimensions according to the basic data set; For each indicator dimension, the indicator values ​​of multiple enterprise objects are clustered to obtain multiple clusters, and the indicator weight of each indicator value is determined according to the average silhouette coefficient of each cluster; The evaluation score of each enterprise object is determined by combining the indicator value and the indicator weight.

2. The method according to claim 1, characterized in that For each indicator dimension, the indicator values ​​of multiple enterprise objects are clustered to obtain multiple clusters, including: For each indicator dimension, the target number of clusters is determined, and the indicator values ​​of multiple enterprise objects are clustered according to the target number of clusters to obtain multiple clustering clusters of the indicator dimension.

3. The method according to claim 2, characterized in that Determining the number of target clusters for each indicator dimension includes: For each indicator dimension, determine the intra-cluster sum of squares under multiple candidate cluster numbers; Determining a first-order difference sequence under the number of the plurality of candidate clusters according to the intra-cluster sum of squares; According to the first-order difference sequence, a target cluster number is determined from the multiple candidate cluster numbers.

4. The method according to claim 3, characterized in that The step of determining the target number of clusters from the plurality of candidate numbers of clusters according to the first-order difference sequence includes: Obtaining a threshold, and determining the number of candidate clusters whose first-order difference sequence is less than or equal to the threshold; From the candidate cluster numbers whose first-order difference sequences are less than or equal to a preset threshold, determine the candidate cluster number with the smallest cluster number as the target cluster number.

5. The method according to any one of claims 1 to 4, characterized in that: Before determining the indicator weight of each indicator value according to the average silhouette coefficient of each cluster, the method further includes: For each cluster, determine the intra-cluster cohesion and inter-cluster separation; The average silhouette coefficient is determined according to the intra-cluster cohesion and the inter-cluster separation.

6. The method according to any one of claims 1 to 4, characterized in that: Before clustering the indicator values ​​of multiple enterprise objects for each indicator dimension to obtain multiple clusters, the method further includes: The indicator value of the enterprise object is discretized.

7. The method according to any one of claims 1 to 4, characterized in that: The step of combining the indicator value and the indicator weight to determine the evaluation score of each enterprise object includes: For each indicator dimension, the indicator value is weighted using the indicator weight to obtain the indicator score; According to the indicator scores of multiple indicator dimensions, an evaluation score of the enterprise object is obtained.

8. The method according to any one of claims 1 to 4, characterized in that: The multiple indicator dimensions include indicator dimensions related to the communication-related data, including any one or more of the following: Communication activity, communication service index, and market coverage.

9. The method according to any one of claims 1 to 4, characterized in that: The basic data set also includes enterprise-related information, and the multiple indicator dimensions include indicators related to the enterprise-related information, including any one or more of the following: Total operating income growth rate, total assets growth rate, total equity investment amount, return on net assets, net profit margin of total assets, cost and expense growth rate, debt-to-asset ratio, current ratio, quick ratio, net profit growth rate, total assets growth rate, and equity growth rate.

10. An evaluation device for an enterprise object, characterized in that: The device comprises: A basic data set construction module, used to construct a basic data set of an enterprise object; wherein the basic data set includes communication-related data; An indicator value determination module, used to determine the indicator values ​​of multiple indicator dimensions according to the basic data set; An indicator weight determination module is used to cluster the indicator values ​​of multiple enterprise objects for each indicator dimension to obtain multiple clusters, and determine the indicator weight of each indicator value according to the average silhouette coefficient of each cluster; The evaluation score determination module is used to determine the evaluation score of each enterprise object by combining the indicator value and the indicator weight.

11. An electronic device, characterized in that: The method comprises a processor, a memory, and a computer program stored in the memory and capable of running on the processor, wherein the computer program implements the method according to any one of claims 1 to 9 when executed by the processor.

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

Images