Enterprise management data governance method and system based on big data analysis

Abstract

The present application relates to the technical field of data processing, in particular to an enterprise management data governance method and system based on big data analysis, comprising the following steps: based on the interaction record data of the existing customers and potential customers of an enterprise, a relationship network between customers is constructed, the interaction intensity within a customer group is calculated through the connection relationship between nodes, and a customer network structure diagram is generated; the present application extracts a closely interactive customer group by analyzing the relationship connection between customers, identifies behavior patterns and consumption tendencies, and especially the participation and economic influence in a multi-region market; according to the abnormal changes of customer behavior data and payment modes, the behavior deviation of abnormal customers and the influence on market fluctuations can be further analyzed; at the same time, the social influence of customers in the group is analyzed, the potential negative influence of customers is judged according to the interaction intensity and network status of customers, and the customers are classified and the resource allocation is optimized.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data processing, and in particular to an enterprise management data governance method and system based on big data analysis. BACKGROUND

[0002] The technical field of data processing involves a variety of operations such as collection, storage, conversion, analysis and visualization of data, with the purpose of extracting valuable information from large amounts of structured and unstructured data to support enterprise decision-making. The field includes data mining, data analysis, data cleaning, data integration and other technical means to ensure the accuracy, consistency and availability of data. With the popularity of big data, cloud computing and artificial intelligence, data processing technology has become an indispensable part of enterprise management, scientific research, finance, healthcare and other industries, helping enterprises optimize resource allocation.

[0003] Among them, the enterprise management data governance method refers to the strategies and measures taken by enterprises in handling and managing their data resources to ensure the quality, integrity, security and compliance of the data. It helps enterprises establish standardized data management processes to improve the reliability and availability of data, thereby supporting operational decision-making, innovation and business optimization.

[0004] Although the prior art can ensure the quality and compliance of data through basic data governance means, it mainly focuses on the standardization of data and optimization of processes in the depth analysis and personalized management of customer data, making it difficult to accurately identify potential high-value customers and abnormal behavior customers in the customer group. In addition, the prior art is difficult to deeply evaluate the interaction intensity and influence between customers when facing the complex social network structure of the customer group, resulting in neglecting customers with potential value or risk in resource allocation and customer management. This leads to the possibility of missing high-potential customer groups or reacting slowly to changes in customer behavior, thereby affecting market expansion and customer relationship maintenance. SUMMARY

[0005] The present application provides an enterprise management data governance method and system based on big data analysis to solve the technical problems in the prior art.

[0006] The technical solution of the present application to solve the above technical problems is as follows: an enterprise management data governance method based on big data analysis, comprising the following steps:

[0007] S1: Based on the interaction record data of existing customers and potential customers of the enterprise, a relationship network between customers is constructed, the interaction intensity within the customer group is calculated through the connection relationship between nodes, and a customer network structure diagram is generated;

[0008] S2: Based on the customer network structure diagram, the association strength of the customer is evaluated, the market participation degree of the customer group in the region and the economic influence are analyzed combined with the regional economic characteristics, the customers with commercial value and growth potential are identified and marked, and the marked potential customer group is generated;

[0009] S3: According to the behavior data of the marked potential customer group, the abnormal consumption behavior characteristics of the customer are extracted, the deviation degree from the regular consumption mode is calculated, the fluctuation influence of the customer on the regional market is evaluated combined with the regional economic situation, and the customer behavior abnormality identification result is generated;

[0010] S4: Based on the customer behavior abnormality identification result and the customer network structure diagram, the interaction strength of the abnormal customer and other customer groups in the customer relationship network is judged, the possibility of the abnormal customer to other customers is analyzed, and the customer group influence evaluation result is generated;

[0011] S5: Based on the customer group influence evaluation result, the customer group is classified, the customer needing risk monitoring, business adjustment optimization, or the priority of the customer in enterprise resource allocation is determined, and the customer management decision support information is generated;

[0012] S6: According to the customer management decision support information, the cooperation potential of the customer is predicted, the future cooperation mode and business opportunity are analyzed, the value level of the customer is displayed through visual data, and the enterprise customer information management record is generated.

[0013] The application improves that the customer network structure diagram includes a plurality of customer nodes, interaction connection lines between nodes and association strength identification of each node, the marked potential customer group includes customers with multiple purchase frequencies, stable order quantity fluctuations and optimized product combination information, the customer behavior abnormality identification result includes abnormal transaction amount change, irregular transaction frequency adjustment and payment method change, the customer group influence evaluation result includes social connection degree, influence range inside and outside the group and potential negative influence risk, the customer management decision support information includes risk monitoring customer list, customer segmentation needing to optimize business operation and high-priority customer group, and the enterprise customer information management record includes customer cooperation historical performance, future growth potential quantitative analysis and risk assessment data.

[0014] The application improves that based on the interaction record data of the existing customers and potential customers of the enterprise, the relationship network between the customers is constructed, the interaction strength inside the customer group is calculated through the connection relationship between the nodes, and the specific steps of generating the customer network structure diagram are as follows:

[0015] S101: Based on the interaction record data of the existing customers and potential customers of the enterprise, the interaction frequency, communication times and interaction intensity are extracted, data classification and grouping operations are performed, classification and labeling are performed according to the customer dimension, and interaction index data is generated;

[0016] S102: Based on the interaction index data, connection analysis between nodes is performed, weights are set through interaction frequency and communication times, initial connection structure between nodes is constructed, and an initial customer relationship network is generated;

[0017] S103: Based on the initial customer relationship network, the connection strength of each customer node is calculated, the hierarchical relationship is divided through the calculation of the interaction relationship in the group, the node position and hierarchical relationship are adjusted, and a customer network structure diagram is generated.

[0018] The application improves that, based on the customer network structure diagram, the correlation strength of the customer is evaluated, the market participation degree of the customer group in the region and the economic influence are analyzed combined with the regional economic characteristics, the customers with commercial value and growth potential are identified and labeled, and the specific steps of generating the labeled potential customer group are as follows:

[0019] S201: Based on the customer network structure diagram, the correlation between customer nodes is analyzed according to the interaction times, the classification operation is performed by aggregating closely related customers, and the correlation strength evaluation result is generated;

[0020] S202: Based on the correlation strength evaluation result, the behavior mode of the customer group is extracted, the purchase cycle, order quantity fluctuation and product combination data are analyzed, the customer is segmented combined with the economic data of the current region, and the behavior mode analysis result is generated;

[0021] S203: Based on the behavior mode analysis result, the key customer group is screened according to the market participation degree, the economic influence is evaluated and data matching operation is performed, and the labeled potential customer group is generated.

[0022] The application improves that, according to the behavior data of the labeled potential customer group, the abnormal consumption behavior characteristics of the customer are extracted, the deviation degree from the regular consumption mode is calculated, the volatility influence of the customer on the regional market is evaluated combined with the economic situation of the region, and the specific steps of generating the customer behavior anomaly identification result are as follows:

[0023] S301: Based on the transaction record of the labeled potential customer group, the transaction amount and payment method of the customer are extracted, the transaction frequency is sorted and classified according to time sequence, the difference of transaction behavior is analyzed, and the transaction behavior data analysis result is generated;

[0024] S302: Based on the transaction behavior data analysis result, compare the regular consumption mode, calculate the deviation of transaction amount fluctuation and payment method change, mark the abnormal change of transaction frequency, integrate the data, and generate the customer deviation analysis result;

[0025] S303: Based on the customer deviation analysis result, combined with the economic data of the region, analyze the fluctuation influence of abnormal consumption behavior on the regional market, match the relevant market data, perform fluctuation evaluation, and generate the customer behavior anomaly identification result.

[0026] The specific steps of the present application for improving the customer group influence evaluation result based on the customer behavior anomaly identification result and the customer network structure diagram are as follows:

[0027] S401: Based on the customer behavior anomaly identification result and the customer network structure diagram, analyze the connection relationship of the abnormal customer in the network, extract the social connection degree of the customer node, integrate the interaction information with the group, and generate the abnormal customer social influence analysis result;

[0028] S402: Based on the abnormal customer social influence analysis result, compare the interaction intensity of other customer groups, judge the interaction frequency and connection tightness of the abnormal customer, mark the influence node, and generate the abnormal customer interaction intensity evaluation result;

[0029] S403: Based on the abnormal customer interaction intensity evaluation result, combined with the social connection degree and interaction evaluation data, analyze the influence of the abnormal customer, evaluate the possibility of negative influence on the group, and generate the customer group influence evaluation result.

[0030] The specific steps of the present application for improving the customer group influence evaluation result based on the customer group influence evaluation result are as follows:

[0031] S501: Based on the customer group influence evaluation result, extract the customer influence data, establish a priority list from high to low according to the importance of the customer group, and generate the customer influence sorting data;

[0032] S502: Based on the customer influence sorting data, combined with the overall business demand of the enterprise, perform customer screening, classify the customers into multiple risk categories through influence threshold setting, and generate the customer classification result;

[0033] S503: Based on the customer classification results, analyze customer resource needs, mark customers who need optimization and risk monitoring, allocate priority resources, determine the customer processing order, and generate customer management decision support information.

[0034] The present invention improves upon this invention by predicting the cooperation potential of customers based on the customer management decision support information, analyzing future cooperation models and business opportunities, and generating enterprise customer information management records by visualizing customer value levels through data visualization. The specific steps are as follows:

[0035] S601: Based on the customer management decision support information, extract the customer's historical cooperation records and business growth data, analyze the customer's historical transaction volume and risk change trends, integrate and organize the data, and generate customer historical data analysis results;

[0036] S602: Based on the analysis results of the customer's historical data, combined with the current business performance, assess the growth potential of future cooperation, predict future cooperation models and business opportunities, and generate customer cooperation potential analysis results;

[0037] S603: Based on the analysis results of customer cooperation potential, visualize the data, present the customer's value level, analyze and integrate the results and classify them, output customer information, and generate enterprise customer information management records.

[0038] The present invention is improved by using the following formula to predict future cooperation models and business opportunities:

[0039]

[0040] Calculate the probability distribution of future customer cooperation patterns and business opportunities. ;

[0041] in, In terms of current business performance The probability of potential future cooperation with customers under certain conditions. It is the probability of the current business data. This indicates continued cooperation with clients. The probability of current business performance under certain conditions. It is the historical growth fluctuation coefficient. It is the customer's historical risk coefficient.

[0042] A big data analytics-based enterprise management data governance system, the system comprising:

[0043] The customer relationship network building module extracts the frequency of interaction and number of communications between customers based on the interaction records of the company's existing and potential customers, builds customer nodes and sets the connection relationship of each node, calculates the interaction density of customer groups according to the connection strength, and generates a customer network structure diagram.

[0044] The customer value evaluation module evaluates the strength of the correlation between customers based on the customer network structure diagram, analyzes the market participation of customers in combination with regional market information, classifies the value of customers according to the influence of customers in different regions, and generates a marked potential customer group;

[0045] The customer behavior analysis module extracts the transaction amount, frequency and payment method of customers based on the marked potential customer group, analyzes the change of consumption mode, identifies the deviation degree from the regular behavior mode, and generates a customer behavior anomaly identification result in combination with regional economic data;

[0046] The customer influence evaluation module extracts the social connection degree and interaction frequency of abnormal customers based on the customer behavior anomaly identification result and the customer network structure diagram, analyzes the position of abnormal customers in the network in combination with the interaction records within the customer group, and generates a customer group influence evaluation result;

[0047] The customer management decision module filters the customer group according to the influence in combination with the business needs of the enterprise, determines the customers that need to be optimized or monitored, integrates data and outputs customer management information, and generates an enterprise customer information management record.

[0048] The beneficial effects of the present application are: by analyzing the relationship connection between customers, extracting the closely interactive customer group, identifying the behavior mode and consumption tendency, especially the participation and economic influence in the multi-regional market. According to the abnormal changes of customer behavior data and payment mode, the behavior deviation of abnormal customers and its influence on market fluctuations can be further analyzed. At the same time, the social influence of customers in the group is analyzed, the interaction strength and network status of customers are comprehensively analyzed, the potential negative influence is judged, the customer classification is helped, the resource allocation is optimized, and the support information for customer management decision is provided. In addition, in combination with the historical transaction data and growth trend of customers, the future cooperation potential is predicted, and the value level of customers is presented through visualization, which provides all-round support for the customer management of enterprises, improves the decision quality and customer management efficiency of enterprises, and enhances the market competitiveness and rationality of resource allocation of enterprises. BRIEF DESCRIPTION OF DRAWINGS

[0049] Figure 1 A flowchart of the enterprise management data governance method based on big data analysis is provided for the present application;

[0050] Figure 2 A detailed process schematic diagram for step S1 of the present application is provided;

[0051] Figure 3 A detailed process schematic diagram for step S2 of the present application is provided;

[0052] Figure 4 The detailed flowchart of step S3 of the present application is shown in the figure;

[0053] Figure 5 The detailed flowchart of step S4 of the present application is shown in the figure;

[0054] Figure 6 The detailed flowchart of step S5 of the present application is shown in the figure;

[0055] Figure 7 The detailed flowchart of step S6 of the present application is shown in the figure;

[0056] Figure 8 The module diagram of the enterprise management data governance system based on big data analysis is provided in the present application. DETAILED DESCRIPTION

[0057] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work fall within the scope of protection of the present application.

[0058] In the description of the present application, the terms "first", "second" are used only for the purpose of description, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined with "first", "second" can explicitly or implicitly include one or more of the features. In the description of the present application, the meaning of "multiple" is two or more, unless otherwise specifically limited.

[0059] In the description of the present application, the term "for example" is used to indicate "as an example, illustration or explanation". Any embodiment described as "for example" in the present application is not necessarily interpreted as more preferred or more advantageous than other embodiments. The following description is given in order to enable any person skilled in the art to implement and use the present application. In the following description, details are listed for the purpose of explanation. It should be understood that those skilled in the art can realize the present application without using these specific details. In other examples, well-known structures and processes will not be described in detail to avoid unnecessary details making the description of the present application obscure. Therefore, the present application is not intended to be limited to the shown embodiments, but is consistent with the broadest scope of principles and features disclosed in the present application.

[0060] Referring to Figure 1 The present application provides a technical solution: an enterprise management data governance method based on big data analysis, comprising the following steps:

[0061] S1: Based on the interaction record data of the existing customers and potential customers of the enterprise, the interaction frequency, communication times and interaction intensity of each customer are analyzed, the relationship network between customers is constructed, the interaction intensity within the customer group is calculated through the connection relationship between nodes, the hierarchical division of the customer group is formed, including the correlation intensity of each customer in the group, and a customer network structure diagram is generated;

[0062] S2: Based on the customer network structure diagram, the correlation intensity of the customers is evaluated, the customer group with multiple interactions and close correlation is extracted, the behavior patterns and consumption tendencies of these customer groups are analyzed, the behavior patterns include the procurement cycle, order volume fluctuation and product portfolio information, combined with the regional economic characteristics, the market participation of the customer group in multiple regions and its economic influence are analyzed, the customers with commercial value and growth potential are identified and marked, and a marked potential customer group is generated;

[0063] S3: According to the transaction record, transaction frequency and payment method of the marked potential customer group, the trend analysis of behavior change is carried out, the abnormal consumption behavior characteristics of the customer are extracted, the abnormal consumption behavior characteristics are abnormal change of transaction amount, abnormal increase or decrease of transaction frequency or change of payment method, the deviation degree from the regular consumption mode is calculated, combined with the regional economic situation, the volatility influence of the customer on the regional market is evaluated, and the customer behavior abnormality identification result is generated;

[0064] S4: Based on the customer behavior abnormality identification result and the customer network structure diagram, the status of the abnormal customer in the customer relationship network is further analyzed, the social connection degree and influence of the customer in the group are extracted, by judging the interaction intensity of the abnormal customer with other customer groups, whether the customer has the possibility to have a negative impact on other customers is analyzed, and a customer group influence evaluation result is generated;

[0065] S5: Based on the customer group influence evaluation result, according to the influence data of the customer and the overall business demand of the enterprise, according to the influence ranking, the customer group is classified in detail, the customers who need to be monitored, adjusted and optimized in business, or whose priority in enterprise resource allocation needs to be strengthened are determined, and customer management decision support information is generated;

[0066] S6: According to the customer management decision support information, the historical cooperation data, business growth trend and risk indicators of the customer are analyzed, combined with the current business performance of the customer, the cooperation potential of the customer is predicted, the future cooperation mode and business opportunity are analyzed, the value level of the customer is displayed through visual data, and enterprise customer information management records are generated.

[0067] The customer network structure diagram includes a plurality of customer nodes, interactive connections between the nodes, and an association strength identifier of each node, the marked potential customer group includes customers with multiple purchase frequencies, stable order quantity fluctuations, and optimized product portfolio information, the customer behavior anomaly identification result includes abnormal transaction amount changes, irregular transaction frequency adjustments, and payment method changes, the customer group influence evaluation result includes social connection degree, influence range within and outside the group, and potential negative influence risk, the customer management decision support information includes risk monitoring customer list, customer segmentation requiring optimization of business operations, and high-priority customer group, and the enterprise customer information management record includes customer cooperation history, future growth potential quantitative analysis, and risk assessment data.

[0068] Referring to Figure 2 , based on the interaction record data of the existing customers and potential customers of the enterprise, a relationship network between the customers is constructed, the interaction strength within the customer group is calculated through the connection relationship between the nodes, and the specific steps of generating the customer network structure diagram are as follows:

[0069] S101: Based on the interaction record data of the existing customers and potential customers of the enterprise, the interaction frequency, communication times and interaction strength are extracted, the data is classified and grouped, and the interaction index data is generated by classifying and marking according to the customer dimension;

[0070] Based on the interaction record data of the existing customers and potential customers of the enterprise, the interaction frequency, communication times and interaction strength information of each customer are extracted, the information is classified and grouped according to the customer dimension, the interaction data of different customer groups is analyzed by layers, and the interaction data of each customer is sorted and classified, so that the data can accurately reflect the interaction relationship between different customers, then, the historical records of the customers are matched and analyzed according to the time sequence and the interaction strength, so that the data integrity and consistency are ensured, finally, the interaction frequency and communication times of each customer are classified and marked to generate the interaction index data, which is used to support subsequent analysis and network construction.

[0071] S102: Based on the interaction index data, the connection analysis between the nodes is performed, the weights of the interaction frequency and the communication times are set, the initial connection structure between the nodes is constructed, and the initial customer relationship network is generated;

[0072] First, each customer is regarded as a node, and the interaction frequency and communication times of the customers are regarded as the basic data of the connection between the nodes. Next, the interaction records between the customers are matched to ensure that the interaction relationship between the nodes can be accurately reflected in the initial customer relationship network. Specifically, according to the interaction index data, the connection relationship of the frequently interacting customers is preferentially constructed according to the high and low of the customer interaction frequency. A hierarchical connection strategy is adopted. The customer nodes exceeding the threshold value are preferentially connected by setting the threshold value of the interaction frequency and the communication times. Then, the nodes below the threshold value are sequentially added to the network, and the corresponding weight is set to reflect the interaction intensity. In this way, the difference of the connection strength between the nodes reflects the close degree of the interaction between the customers. Then, it is checked whether there is an isolated node or a connection deficiency between the nodes. The connection parameters are adjusted or the connection weight is re-divided to ensure that there is a reasonable interaction relationship between all customer nodes. Finally, all nodes and their connections are integrated and optimized to generate the initial customer relationship network. The network shows the basic interaction relationship between the customer groups to support subsequent analysis.

[0073] S103: Based on the initial customer relationship network, the connection strength of each customer node is calculated. The hierarchical relationship is adjusted by calculating the interaction relationship in the group to generate a customer network structure diagram.

[0074] For calculating the connection strength of each customer node, the formula is as follows:

[0075]

[0076] The connection strength of the customer node is calculated. In the formula: represents the connection strength of the customer , which is the total of all connections of the customer in the network. The higher the value, the more frequent the interaction between the customer and other customers, represents the index of the other customers having a connection relationship with the customer , and the value ranges from 1 to , wherein is the total number of customers associated with the customer , represents whether there is a connection between the customer and the customer . If there is a connection, the value is 1, otherwise it is 0, represents the interaction frequency between the customer and the customer . It is usually obtained through the number of times in the customer transaction record or communication record. The higher the value, the more frequent the interaction.

[0077] If there are 5 interactions between customer A and customer B, the interaction frequency is 5, that is, According to the interaction data of customer A and other customers, if customer A has interaction frequency of 4 and 2 times with other customers C and D respectively, and both have connection relationship, that is and , the data is brought into the formula:

[0078]

[0079] By comparing the historical value of customer A, assuming that the previous connection strength is 8, it can be seen that the current connection strength increases to 11, indicating that the interaction strength of customer A with other customers increases by 37.5%. This result helps to determine the rising trend of customer A's activity in the group, and provides support for subsequent customer stratification and resource allocation decisions. According to the connection strength of each customer, the customer group is divided into different levels, and the customers with higher connection strength are placed in the network center, representing the core customers, and the customers with less interaction are placed in the periphery, forming a customer relationship structure with different levels. Finally, adjust the relative position and level relationship of each level node to ensure that customers with high interaction strength are located in the network center and customers with low strength are distributed in the periphery, and generate a customer network structure diagram.

[0080] Please refer to Figure 3 , based on the customer network structure diagram, the association strength of the customer is evaluated, combined with the regional economic characteristics, the market participation and economic influence of the customer group in the region are analyzed, and the customers with commercial value and growth potential are identified and marked, and the specific steps of generating the marked potential customer group are as follows:

[0081] S201: Based on the customer network structure diagram, the association between customer nodes is analyzed according to the number of interactions, and the classification operation is performed by aggregating closely related customers, and the association strength evaluation result is generated;

[0082] First, extract the interaction frequency data of each customer node, convert the interaction frequency of customer nodes into the association strength between nodes through association analysis method, then use threshold-based clustering method to aggregate customer nodes with high interaction frequency together, set the threshold of interaction frequency to filter out the customer group with close interaction, and classify them into the same class, then analyze the association between different groups to ensure that the customers in the group interact frequently and avoid frequent interaction between customers across groups, finally, classify all closely related customer nodes, and sort and group them according to the association strength, to generate the association strength evaluation result, which supports the analysis of customer behavior patterns.

[0083] S202: Based on the association strength evaluation result, extract the behavior pattern of the customer group, analyze the procurement cycle, order volume fluctuation and product portfolio data, and combine the economic data of the current region to segment the customers, and generate the behavior pattern analysis result;

[0084] Using data mining methods, the purchasing cycle and order quantity fluctuation data of each customer group are obtained, then, according to the order history of the customer, the main product combination information of the customer is extracted, the pattern of the customer in the purchasing and consumption behavior is identified, then, combined with the economic data of the current region, the behavior pattern of the customer group is cross compared with the regional economic performance, the behavior characteristics of the customer in different economic backgrounds are further refined, and the customer group is classified by the segmentation rule, ensuring that the customer group is accurately matched according to the regional economic performance and the behavior pattern, finally, the behavior data of all customer groups are integrated to generate the behavior pattern analysis result, supporting the subsequent key customer screening.

[0085] S203: Based on the behavior pattern analysis result, the key customer group is screened according to the market participation degree, the economic influence is evaluated and the data is matched, and the marked potential customer group is generated;

[0086] First, the core economic indicators of each customer group are collected, including sales, market share, profit margin and annual order quantity data, which can be obtained from the enterprise's internal sales data platform or customer financial statements, then, the market position of the customer group in its industry is analyzed, especially the proportion of its sales relative to the same industry and market expansion ability, then, these economic indicators are combined with market participation data, the market share, order quantity proportion and profit growth rate of the customer in the industry are calculated, and the economic influence of each customer group is quantified, then, the market activity and economic influence of these customers are compared, and the customer group with high market participation and excellent economic indicators is screened out, finally, these customer groups are marked as customers with commercial value and growth potential, and the marked potential customer group is generated, which is used for subsequent enterprise decision and resource allocation.

[0087] Please refer to Figure 4 According to the behavior data of the marked potential customer group, the abnormal consumption behavior characteristics of the customer are extracted, the deviation degree of the customer from the regular consumption pattern is calculated, combined with the economic situation of the region, the volatility influence of the customer on the regional market is evaluated, and the specific steps of generating the customer behavior anomaly identification result are as follows:

[0088] S301: Based on the transaction records of the marked potential customer group, the transaction amount and payment method of the customer are extracted, the transaction frequency is sorted and classified in time sequence, the difference of transaction behavior is analyzed, and the transaction behavior data analysis result is generated;

[0089] First, the transaction records of each customer are sorted, and data such as transaction amount, transaction time and payment method are extracted. Time series analysis method is used to sort the transaction data of the customer by time, and the data is divided into time periods such as day, week and month. Then, the transaction frequency in each time period is classified and processed, and the fluctuation range of transaction amount and the change of payment method are particularly focused on. The transaction amount is divided into high, medium and low three intervals, and the change of payment method used by the customer in different time periods is analyzed to identify whether the customer frequently changes the payment method. Through correlation analysis method, the transaction amount and payment method data are compared to find the change point in the customer behavior. Finally, these data are integrated to generate transaction behavior data analysis results, which are used to identify the transaction behavior characteristics of the customer and provide support for subsequent deviation analysis.

[0090] S302: Based on the transaction behavior data analysis results, compare the regular consumption mode, calculate the deviation of transaction amount fluctuation and payment method change, mark the abnormal change of transaction frequency, and integrate the data to generate customer deviation analysis results;

[0091] For comparison with the regular consumption mode, according to the formula:

[0092]

[0093] The deviation of transaction amount fluctuation and payment method change of the customer is calculated.

[0094] In the formula: represents the deviation of transaction amount or payment method, which reflects the difference between the customer's transaction behavior and the regular mode. The larger the value, the more the customer's transaction behavior deviates from the regular mode, represents the actual amount or payment method data of the th transaction. This data is usually obtained from the customer's transaction records, and the system will extract these information from each transaction of the customer, represents the average transaction amount or payment method standard value under the regular consumption mode, which is usually calculated based on historical data through statistical method. The fluctuation range may be different for different customer groups.

[0095] If the transaction amount of the th transaction of a customer is 500 yuan, and the average transaction amount under the regular consumption mode is 400 yuan, then according to the above formula, the transaction amount deviation of the customer can be calculated as:

[0096]

[0097] The result indicates that the amount of the customer in this transaction is 25% higher than the regular mode. Similarly, assuming that the payment method of the customer is usually bank transfer, and in a transaction, an electronic payment method is used, the deviation degree can be calculated by comparing the data of the regular payment method with the current payment method. According to the calculation result, the abnormal change of the transaction frequency is marked. The abnormal standard of the transaction frequency is set by using the classification threshold method, for example, if the transaction frequency of the customer suddenly increases by more than 50%, it is marked as abnormal transaction behavior. All data with a deviation degree greater than the set threshold are screened, and the transaction records that meet the conditions are marked as abnormal behavior. The deviation degree and the abnormal change of the transaction frequency are integrated to ensure that the abnormal transaction behavior of each customer is recorded, and finally the customer deviation degree analysis result is generated for further regional market analysis and risk assessment.

[0098] S303: Based on the customer deviation degree analysis result, combined with the economic data of the region, analyze the influence of abnormal consumption behavior on the fluctuation of the regional market, match the relevant market data, conduct fluctuation evaluation, and generate customer behavior anomaly identification result;

[0099] First, combined with the economic data of the region where the customer is located, the key economic indicators in the region are obtained, such as consumption level, market demand in the region, economic growth rate, inflation rate, etc. The consumption behavior data of the customer is matched with the regional economic data, and through stepwise regression analysis, it is focused on identifying whether the abnormal consumption behavior of the customer corresponds to the regional economic fluctuation. Then, analyze the potential impact of these abnormal behaviors on the fluctuation of the regional market, especially the impact of the customer's large consumption, payment method change, etc. on the market consumption structure. Then, compare the abnormal consumption behavior of the customer with the consumption trend of other customers in the region to evaluate whether the abnormal behavior of the customer may trigger changes in the consumption trend in the region. Finally, all analysis results are integrated to generate customer behavior anomaly identification result for further monitoring and evaluation of the volatility risk of the regional market.

[0100] Please refer to Figure 5 Based on the customer behavior anomaly identification result and the customer network structure diagram, the interaction intensity between the abnormal customers and other customer groups in the customer relationship network is judged, the possibility of the abnormal customers having a negative impact on other customers is analyzed, and the specific steps of generating the customer group influence evaluation result are as follows:

[0101] S401: Based on the customer behavior anomaly identification result and the customer network structure diagram, analyze the connection relationship of the abnormal customers in the network, extract the social connection degree of the customer node, integrate the interaction information with the group, and generate the abnormal customer social influence analysis result;

[0102] Firstly, the connection relationship data of the abnormal customer and other customer nodes is obtained, the social connection degree of the customer node is extracted, the social connection degree is obtained through the interaction quantity and intensity of the customer and other nodes, then the connection relationship of the abnormal customer is integrated with the overall interaction information of the group, the interaction frequency of the abnormal customer and other customers is compared, the position and importance of the abnormal customer in the social network are analyzed, and whether the abnormal customer is in the core position of the social network is focused on, then the abnormal customers with high social connection degree are clustered, and the nodes that frequently interact with other customers are identified, finally, the customer nodes with high social connection degree and active interaction are taken as high-impact customer nodes, and the abnormal customer social influence analysis result is generated, which provides support for the subsequent interaction intensity evaluation.

[0103] S402: Based on the abnormal customer social influence analysis result, the interaction intensity of other customer groups is compared, the interaction frequency and connection tightness of the abnormal customer are judged, the impact node is marked, and the abnormal customer interaction intensity evaluation result is generated;

[0104] Firstly, the interaction intensity data of the abnormal customer and other customers is obtained, the interaction times and interaction frequency between nodes are used as the judgment standard, the interaction behavior of each abnormal customer is compared in detail through the association analysis method, then the interaction frequency of the abnormal customer is compared with the interaction frequency of other customers in the group, it is judged which customer's interaction frequency is significantly higher than the average level of the group, then the customer nodes with high-frequency interaction are marked, and the customers who maintain frequent interaction in the group are identified, the clustering method is used to classify the customer nodes with frequent interaction and mark them as impact nodes, finally, all customer nodes with high interaction frequency are integrated to generate the abnormal customer interaction intensity evaluation result, which provides the basis for further customer influence evaluation.

[0105] S403: Based on the abnormal customer interaction intensity evaluation result, the influence of the abnormal customer is analyzed by combining the social connection degree and interaction evaluation data, the possibility of negative influence of the abnormal customer on the group is evaluated, and the customer group influence evaluation result is generated;

[0106] Firstly, the influence parameter of each abnormal customer is extracted by combining the social connection degree and interaction frequency data of the abnormal customer, the influence parameter is quantified by the social connection number, interaction intensity and influence range of the customer on other customers, then the interaction mode of the abnormal customer is compared with other customers, whether these abnormal customers have high negative influence potential is analyzed, and those abnormal customers with high influence are focused on, especially those customers who frequently participate in interaction and have wide influence range, then whether the influence of these customers can produce negative effects on the decision and behavior of other customers is evaluated, finally, the customers with high influence and negative influence potential are marked out, and the customer group influence evaluation result is generated, which provides the basis for subsequent risk management.

[0107] Referring to Figure 6 Based on the customer group influence evaluation result, the customer group is classified in detail to determine the customers that need risk monitoring, business adjustment and optimization, or to strengthen their priority in enterprise resource allocation, and the specific steps of generating customer management decision support information are as follows:

[0108] S501: Based on the customer group influence evaluation result, extract customer influence data, establish a priority list from high to low according to the importance of the customer group, and generate customer influence ranking data;

[0109] First, extract the influence data of each customer, including the customer's social connection degree, interaction frequency with other customers, market share, sales, and profit contribution, etc. Then, according to the relative importance of the customer group, these data are sorted, and each influence index is scored using the weighted scoring method. Different weights are assigned to social connection degree, interaction frequency, market contribution, etc. to ensure that the total influence score of each customer can objectively reflect its position in the group. Subsequently, according to the weighted score result, a priority list is established from high to low according to the influence, and the customers are divided into high, medium and low influence categories. Finally, all the sorted customers are integrated into a list to generate customer influence ranking data for subsequent customer screening and classification operations.

[0110] S502: Based on the customer influence ranking data, combine the overall business needs of the enterprise to screen customers, set classification standards through influence threshold, and divide customers into multiple risk categories to generate customer classification results;

[0111] First, combine the current business needs of the enterprise to determine the customer group that needs to be prioritized. Then, set classification standards through influence threshold to divide customers into multiple risk categories. The specific operation is as follows: set the division standards of high, medium and low risk categories according to the influence score of the customer, set a certain influence threshold, select high-influence customers as high-risk category and prioritize them in the list of key customers. Then, classify medium-influence customers as medium-risk category and allocate corresponding resources according to the business needs of the enterprise. Finally, monitor and manage low-influence customers as low-risk category to generate customer classification results for subsequent risk management and resource allocation operations.

[0112] S503: Based on the customer classification results, analyze customer resource needs, mark customers that need optimization and risk monitoring, allocate priority resources, determine customer processing order, and generate customer management decision support information;

[0113] First, extract each customer's resource demand data, get the customer's historical resource usage and future resource demand prediction, then, through the customer's risk classification, judge which customers need to optimize resource allocation or conduct risk monitoring, combine the customer's importance and risk category, prioritize allocating more resources to high-risk and high-influence customers, while adjusting resources for customers who need optimization, specific operation is: analyze the customer's order history, current inventory, service request and other information, compare these data with the customer's resource demand, mark out the customers who need more support or monitoring, finally, determine the resource allocation and processing order of each customer, generate customer management decision support information, ensure that the enterprise makes the best decision on customer resource allocation and management.

[0114] Please refer to Figure 7 , according to the customer management decision support information, predict the cooperation potential of customers, analyze the future cooperation mode and business opportunities, and visualize the data to show the value level of customers, and the specific steps to generate enterprise customer information management records are as follows:

[0115] S601: Based on the customer management decision support information, extract the customer's historical cooperation records and business growth data, analyze the customer's historical transaction volume and risk change trend, integrate and arrange the data, and generate customer historical data analysis results;

[0116] First, extract cooperation details from the customer's historical cooperation records, including contract content, order execution, project completion time, and customer cooperation cycle with the enterprise, then get the customer's business growth data, such as annual sales, customer's contribution to the enterprise's total revenue, and order volume fluctuation trend, classify and compare these data according to year, then analyze the customer's historical transaction volume change through trend analysis method, especially identify the trend of increasing or decreasing customer order volume year by year, then analyze the customer's risk data, get the historical default, contract interruption or delayed payment information, correlate the risk data with the customer's business growth, identify the customer's risk change trend, finally integrate all cooperation records, transaction volume data and risk information together, generate customer historical data analysis results, provide a basis for subsequent cooperation potential assessment.

[0117] S602: Based on the customer historical data analysis results, combined with the current business performance, assess the growth potential of future cooperation, predict future cooperation mode and business opportunities, and generate customer cooperation potential analysis results;

[0118] For predicting future cooperation mode and business opportunities, use the formula:

[0119]

[0120] Computing the probability distribution of future cooperation patterns and business opportunities for customers ;

[0121] where, is the potential probability of future cooperation of the customer under the current business performance , representing the likelihood of the customer maintaining cooperation based on the existing business performance, which can be estimated through historical business performance statistics or prediction models, and the calculation method may be based on key business indicators such as historical order execution rate and sales growth rate, is the probability of current business data, indicating the success rate or business performance of the current business situation, which can be calculated by the current annual sales, market share, order completion rate, etc. data, usually using the current annual business data divided by the historical average data to estimate, represents the probability of current business performance under the condition of the customer's continuous cooperation , indicating the likelihood of current business performance assuming the customer continues to cooperate. This value can be estimated by analyzing the business performance in historical cooperation, estimating the success rate of business when the customer cooperates, is the historical growth fluctuation coefficient, reflecting the growth volatility of the customer's historical transaction volume. This coefficient can be measured by calculating the standard deviation of the customer's historical order volume or sales, and then deducing the size of the coefficient. Formula: , where, represents the standard deviation of the customer's historical growth rate, represents the average value of the customer's historical growth rate, is the historical risk coefficient of the customer, measuring the proportion of historical defaults or delayed payments of the customer. This value can be calculated by the ratio of the number of defaults to the total number of cooperations in the past cooperation of the customer, for example, when the default rate is 10%, .

[0122] Sales data is directly obtained from the customer's financial records or order management system for the past few years of annual sales. Order volume data can be obtained from the enterprise's order management system to obtain the customer's order execution, and the order volume fluctuation is calculated annually. The default rate data is calculated by the historical cooperation records of the customer to obtain the default information, the number of delayed payments, etc.

[0123] If a certain enterprise in the analysis process, the current annual sales of the customer is 50 million yuan, the sales of the previous two years are 45 million yuan and 48 million yuan respectively, the growth rates are (5000 - 4800) / 4800 = 4.17%, the growth rate of the previous year is (4800 - 4500) / 4500 = 6.67%, then the average growth rate of the three years is (4.17% + 6.67%) / 2 = 5.42%. If the standard deviation of historical growth fluctuation is 0.12, There is a 10% probability of default in the customer's history, so Recalculate the customer's future cooperation potential:

[0124]

[0125] The results show that the probability of the customer's potential in future cooperation is 80.52%. By comparing with the benchmark value of 80%, it can be found that the cooperation potential of the current customer has slightly improved, which indicates that the customer has a high potential for future cooperation, although the historical risk coefficient is high, but the business growth fluctuation is relatively stable, so it is still an important customer for future cooperation of the enterprise. This provides favorable information for the enterprise's business planning in the next few years, and the enterprise can prioritize resource allocation and cooperation deepening for such customers.

[0126] S603: Based on the analysis results of customer cooperation potential, visual data display is performed to present the value hierarchy of customers, the analysis and integration results are classified, and customer information is output to generate enterprise customer information management records;

[0127] First, the value of the customer is classified, and through the analysis of the business growth potential of the customer, the historical transaction records and the future cooperation opportunities, the customers are divided into high potential, medium potential and low potential customer groups. Then, through the data visualization tool, the value hierarchy of the customers is displayed in the form of column chart, pie chart, etc. to present the importance of each customer in the enterprise strategy, especially for the high potential customers, which are marked to ensure that these customers are given priority in resource allocation and business planning of the enterprise. Subsequently, the comprehensive data of the customers is further classified and arranged, including order quantity, market share, cooperation time length and other factors, to form different customer levels. Finally, the classified customer information is output to generate enterprise customer information management records to ensure that these information can be fully utilized in enterprise management decision-making.

[0128] Please refer to Figure 8 , the enterprise management data governance system based on big data analysis, the system comprises:

[0129] The customer relationship network construction module extracts the interaction frequency and communication times between customers based on the interaction record data of the existing customers and potential customers of the enterprise, constructs customer nodes and sets the connection relationship of each node, calculates the interaction density of the customer group according to the connection strength, and generates a customer network structure diagram;

[0130] The customer value evaluation module evaluates the correlation strength between customers based on the customer network structure diagram, analyzes the market participation of customers in combination with regional market information, classifies the value of customers according to their influence in different regions, and generates a marked potential customer group;

[0131] The customer behavior analysis module extracts customer transaction amount, frequency and payment method based on the marked potential customer group, analyzes the change of consumption mode, identifies the deviation degree from the regular behavior mode, generates customer behavior anomaly identification results in combination with regional economic data;

[0132] The customer influence evaluation module extracts the social connection degree and interaction frequency of the abnormal customer based on the customer behavior anomaly identification results and the customer network structure diagram, analyzes the position of the abnormal customer in the network in combination with the interaction records within the customer group, and generates customer group influence evaluation results.

[0133] The customer management decision module filters the customer group according to the influence in combination with the business needs of the enterprise, determines the customers to be optimized or monitored, integrates data and outputs customer management information, and generates enterprise customer information management records.

[0134] Although the preferred embodiments of the present application have been described, those skilled in the art who, once aware of the basic inventive concept, can make further changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications falling within the scope of the present application.

[0135] Obviously, those skilled in the art can make various modifications and variations to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalent technologies, the present application also intends to include these modifications and variations.

Claims

1. A method for enterprise management data governance based on big data analysis, characterized in that, The method comprises the following steps: Based on the interaction record data of the existing customers and potential customers of the enterprise, a relationship network between the customers is constructed, the interaction intensity within the customer group is calculated through the connection relationship between the nodes, and a customer network structure diagram is generated; Based on the customer network structure diagram, the correlation between the customer nodes is analyzed according to the number of interactions, the customers with strong correlation are aggregated, a classification operation is performed, and a correlation strength evaluation result is generated; Based on the correlation strength evaluation result, the behavior mode of the customer group is extracted, the procurement cycle, order quantity fluctuation and product combination data are analyzed, the customer is segmented in combination with the economic data of the current region, and a behavior mode analysis result is generated; Based on the behavior mode analysis result, the key customer group is screened according to the market participation degree, an economic influence evaluation and data matching operation are performed, and a marked potential customer group is generated; Based on the transaction record of the marked potential customer group, the transaction amount and payment method of the customer are extracted, the transaction frequency is sorted and classified in time sequence, the difference of transaction behavior is analyzed, and a transaction behavior data analysis result is generated; Based on the transaction behavior data analysis result, the deviation degree of transaction amount fluctuation and payment method change is calculated by comparing the regular consumption mode, the abnormal change of transaction frequency is marked, and the data is integrated to generate a customer deviation degree analysis result; Based on the customer deviation degree analysis result, the fluctuation influence of abnormal consumption behavior on the regional market is analyzed in combination with the economic data of the region, the relevant market data is matched, the fluctuation is evaluated, and a customer behavior anomaly identification result is generated; Based on the customer behavior anomaly identification result and the customer network structure diagram, the connection relationship of the abnormal customer in the network is analyzed, the social connection degree of the customer node is extracted, the interaction information with the group is integrated, and an abnormal customer social influence analysis result is generated; Based on the abnormal customer social influence analysis result, the interaction frequency and connection tightness of the abnormal customer are judged by comparing the interaction intensity of other customer groups, the influence node is marked, and an abnormal customer interaction strength evaluation result is generated; Based on the abnormal customer interaction strength evaluation result, the influence of the abnormal customer is analyzed in combination with the social connection degree and interaction evaluation data, the possibility of negative influence of the abnormal customer on the group is evaluated, and a customer group influence evaluation result is generated; Based on the customer group influence evaluation result, the customer group is classified in detail, the customers who need to be monitored, the business adjusted and optimized, and the enterprise resource allocation priority enhanced are determined, and customer management decision support information is generated; Based on the customer management decision support information, the customer historical cooperation record and business growth data are extracted, the historical transaction volume and risk change trend of the customer are analyzed, the data is integrated and sorted, and a customer historical data analysis result is generated; Based on the customer historical data analysis result, the growth potential of future cooperation is evaluated in combination with the current business performance, the future cooperation mode and business opportunity are predicted, and a customer cooperation potential analysis result is generated; Based on the customer cooperation potential analysis result, visual data display is performed, the value level of the customer is presented, the integrated result is classified, the customer information is output, and an enterprise customer information management record is generated.

2. The big data analytics based enterprise management data governance method as claimed in claim 1, wherein: The customer network structure diagram includes a plurality of customer nodes, interactive connections between nodes, and an association strength identifier of each node, the marked potential customer group includes customers with multiple purchase frequencies, stable order quantity fluctuations, and optimized product portfolio information, the customer behavior anomaly identification result includes abnormal transaction amount changes, irregular transaction frequency adjustments, and payment method changes, the customer group influence assessment result includes social connection degree, influence range within and outside the group, and potential negative impact risk, the customer management decision support information includes a risk monitoring customer list, a customer segment that needs to optimize business operations, and a high-priority customer group, and the enterprise customer information management record includes customer cooperation history, future growth potential quantitative analysis, and risk assessment data.

3. The big data analytics based enterprise management data governance method as claimed in claim 1, wherein: Based on the interaction record data of the existing customers and potential customers of the enterprise, a relationship network between the customers is constructed, the interaction strength within the customer group is calculated through the connection relationship between the nodes, and the specific steps of generating the customer network structure diagram are as follows: Based on the interaction record data of the existing customers and potential customers of the enterprise, the interaction frequency, communication times and interaction strength are extracted, the data is classified and grouped, and the interaction index data is generated according to the customer dimension. Based on the interaction index data, the connection analysis between nodes is performed, the weights of the interaction frequency and communication times are set, the initial connection structure between nodes is constructed, and the initial customer relationship network is generated. Based on the initial customer relationship network, the connection strength of each customer node is calculated, the hierarchical relationship within the group is calculated, the node position and hierarchical relationship are adjusted, and the customer network structure diagram is generated.

4. The big data analytics based enterprise management data governance method as claimed in claim 1, wherein: Based on the customer group influence assessment result, the customer group is classified in detail, the customers that need to be monitored for risk, adjusted for business, or have a higher priority in the allocation of enterprise resources are determined, and the specific steps of generating the customer management decision support information are as follows: Based on the customer group influence assessment result, the customer influence data is extracted, the priority order list is established from high to low according to the importance of the customer group, and the customer influence sorting data is generated. Based on the customer influence sorting data, the customers are screened in combination with the overall business demand of the enterprise, the classification standard is set through the influence threshold, the customers are divided into multiple risk categories, and the customer classification result is generated. Based on the customer classification result, the customer resource demand is analyzed, the customers that need to be optimized and monitored for risk are marked, the priority resources are allocated, the customer processing order is determined, and the customer management decision support information is generated.

5. The enterprise management data governance method based on big data analysis according to claim 1, characterized in that: For predicting future cooperation modes and business opportunities, the formula is used: Computing probability distributions of future collaboration patterns and business opportunities for a client ; in, In terms of current business performance The probability of potential future cooperation with customers under certain conditions. It is the probability of the current business data. This indicates continued cooperation with clients. The probability of current business performance under certain conditions. It is the historical growth fluctuation coefficient. It is the customer's historical risk coefficient.

6. A big data analysis based enterprise management data governance system characterized in that, The enterprise management data governance method based on big data analysis according to any one of claims 1-5 is executed, and the system comprises: The customer relationship network construction module extracts the interaction frequency and communication times between customers based on the interaction record data of the existing customers and potential customers of the enterprise, constructs customer nodes and sets the connection relationship of each node, calculates the interaction density of the customer group according to the connection strength, and generates a customer network structure diagram; The customer value evaluation module evaluates the correlation strength between customers based on the customer network structure diagram, analyzes the market participation degree of the customers in combination with the regional market information, classifies the customers according to the influence of the customers in different regions, and generates a marked potential customer group; The customer behavior analysis module extracts the transaction amount, frequency and payment method of the customers based on the marked potential customer group, analyzes the change of the consumption mode, identifies the deviation degree from the conventional behavior mode, and generates a customer behavior anomaly recognition result in combination with the regional economic data; The customer influence evaluation module extracts the social connection degree and interaction frequency of the abnormal customers based on the customer behavior anomaly recognition result and the customer network structure diagram, analyzes the position of the abnormal customers in the network in combination with the interaction record within the customer group, and generates a customer group influence evaluation result; The customer management decision module filters the customer group according to the influence in combination with the business demand of the enterprise, determines the customers that need to be optimized or monitored, integrates data and outputs customer management information, and generates an enterprise customer information management record.

Images