Customer portrait information data management system

Abstract

The invention discloses a client portrait information data management system, and the system comprises a demand response module which is used for responding to a portrait construction demand, and creating a new portrait construction task; the service data comprises service types, service ranges and service capabilities of suppliers; the data screening module is used for screening and obtaining a plurality of to-be-selected enterprises and an enterprise data set of each to-be-selected enterprise in the enterprise database according to the service data; the enterprise data set comprises enterprise names, enterprise contacts and enterprise operation data; and the portrait construction module is used for acquiring a customer portrait of each to-be-selected enterprise according to the enterprise data set of the to-be-selected enterprises within the task time limit and confirming a plurality of target enterprises according to the customer portraits. The technical problem of how to improve the utilization rate of the data and reduce the analysis cost of the data in the customer portrait construction process of the enterprise in the prior art is solved.

Description

Technical Field

[0001] The present invention relates to the technical field of customer portrait data management, and in particular to a customer portrait information data management system. Background Art

[0002] Customer portrait is a virtual image created by suppliers for customers. It is a customer model based on a series of real data. Its purpose is to better understand and serve customers, locate target customer groups more accurately, and formulate more effective service strategies.

[0003] In the prior art, a series of public data on the Internet platform is obtained to construct a customer profile of the enterprise, and then decisions are made based on the customer profile of the enterprise. However, in the process of constructing the customer profile of the enterprise, the data of the same enterprise is repeatedly collected multiple times, resulting in low data utilization; different suppliers require different profile construction methods due to different service contents, which makes the data analysis cost high.

[0004] From this, we can see that the technical problem existing in the existing technology is how to improve data utilization while reducing data analysis costs in the process of building a customer portrait for an enterprise. Summary of the invention

[0005] The purpose of the present invention is to provide a customer portrait information data management system, which solves the technical problem in the prior art of how to improve data utilization while reducing data analysis costs during the process of building customer portraits for an enterprise.

[0006] The present invention provides a customer portrait information data management system, the management system comprising:

[0007] A demand response module, which is used to respond to the demand for portrait building and create a new portrait building task; the portrait building task includes a task number, a task time limit, and service data; the service data includes the service type, service scope, and service capability of the supplier;

[0008] A data screening module, the data screening module is used to screen and obtain a number of candidate enterprises and a set of enterprise data of each candidate enterprise in the enterprise database according to the service data; the enterprise data set includes the enterprise name, enterprise contact person and enterprise operation data;

[0009] A portrait construction module is used to obtain the customer portrait of each candidate enterprise based on the enterprise data set of the candidate enterprise within the task time limit and confirm several target enterprises based on the customer portrait.

[0010] Furthermore, the data screening module is used to screen and obtain a number of candidate enterprises in the enterprise database according to the service data, including:

[0011] Based on the service type, obtain type filtering conditions; service types include raw material supply, raw material processing, software operation, transportation and distribution, and financial insurance;

[0012] Based on the service scope, obtain the scope filtering conditions; the service scope includes industrial field, agricultural field, financial field and life field;

[0013] Based on service capabilities, obtain capability screening conditions; service capabilities include processing and production capabilities, transportation and distribution capabilities, software operation capabilities, and financial risk-bearing capabilities;

[0014] Based on the type screening condition, the range screening condition and the capability screening condition, the enterprises in the enterprise database are screened to obtain a number of candidate enterprises.

[0015] Furthermore, the data screening module obtains the enterprise data set of each enterprise to be selected, including:

[0016] Get basic data items;

[0017] Based on the service type and service scope, identify the associated data items;

[0018] Based on the basic data items and the associated data items, an enterprise data set corresponding to each to-be-selected enterprise is obtained in the enterprise database.

[0019] Furthermore, the portrait construction module is used to obtain the customer portrait of each candidate enterprise according to the enterprise data set of the candidate enterprise within the task time limit, including:

[0020] Acquire a portrait relationship database, wherein the portrait relationship database includes basic data items, associated data items, customer portraits, and mapping relationships between the basic data items, associated data items, and customer portraits;

[0021] Obtain a customer portrait analysis model by training a neural network model through a portrait relational database;

[0022] The enterprise data set of the candidate enterprise is input into the customer portrait analysis model to obtain the customer portrait of the candidate enterprise; the customer portrait includes consumption capacity, financial risk index and business stability index.

[0023] Furthermore, the portrait building module identifies several target companies based on the customer portrait, including:

[0024] Construct a fitness function based on the customer profile of the candidate enterprise;

[0025] According to a number of candidate enterprises, a number of combination plans are randomly obtained and an initial population is formed;

[0026] Based on the fitness function, obtain the fitness corresponding to each combination scheme in the initial population;

[0027] Iterative optimization is performed based on fitness to obtain the target combination plan, and each candidate enterprise in the target combination plan is used as the target enterprise.

[0028] Furthermore, the fitness function is calculated as follows:

[0029]

[0030] Among them, K is the fitness function; α i is the weight coefficient corresponding to the i-th selected enterprise; X i is the consumption capacity of the i-th candidate enterprise, C i is the financial risk index corresponding to the i-th candidate enterprise; W i is the operating stability index of the ith candidate enterprise; n is the number of candidates in the combination scheme; e, f and g are all fixed coefficients.

[0031] Furthermore, iterative optimization is performed based on fitness to obtain the target combination solution, including:

[0032] S10: Sort the combination schemes of the initial population in descending order according to the fitness function;

[0033] S11: Select several combinations with the highest fitness rankings for mutation and crossover to form a new generation of population;

[0034] S12: Obtain the fitness of several combination schemes of the new generation population, and sort them in descending order of fitness;

[0035] S13: Repeat steps S11 to S13 until the preset conditions are met, the iteration ends, and the combination scheme with the highest fitness in the contemporary population is selected as the target combination scheme.

[0036] Furthermore, the preset conditions include that the number of iterations reaches a preset number or there is a combination scheme in the population whose fitness is greater than the set fitness.

[0037] Furthermore, the preset condition also includes the expiration of the task deadline.

[0038] Furthermore, the management system also includes: a data collection module, which is used to collect several items of enterprise data of the enterprise according to a preset period, and store each item of enterprise data in the enterprise database after marking the enterprise mark.

[0039] Compared with the prior art, the present invention has the following beneficial effects:

[0040] In the present invention, by setting a demand response module to meet the portrait construction requirements proposed by the supplier, the supplier does not need to collect data from each enterprise to build a customer portrait by itself; by setting a data screening module, the selected enterprises and corresponding enterprise data sets that meet the portrait construction requirements are screened in the enterprise database, so that the acquired enterprises and enterprise data are more accurate, the work efficiency is improved, and the analysis cost is reduced; by setting a portrait construction module, the customer portrait of the enterprise is constructed according to the enterprise data set of the selected enterprise, so as to further obtain the target enterprise. The efficiency and effect of building customer portraits are improved. The technical problem of how to improve the utilization rate of data while reducing the analysis cost of data in the process of building customer portraits of enterprises in the prior art is solved. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 This is a principle block diagram of a customer portrait information data management system of the present invention. DETAILED DESCRIPTION

[0042] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0043] like Figure 1 As shown, a customer portrait information data management system, the management system includes:

[0044] A demand response module, which is used to respond to the demand for portrait building and create a new portrait building task; the portrait building task includes a task number, a task time limit, and service data; the service data includes the service type, service scope, and service capability of the supplier;

[0045] A data screening module, the data screening module is used to screen and obtain a number of candidate enterprises and a set of enterprise data of each candidate enterprise in the enterprise database according to the service data; the enterprise data set includes the enterprise name, enterprise contact person and enterprise operation data;

[0046] A portrait construction module is used to obtain the customer portrait of each candidate enterprise based on the enterprise data set of the candidate enterprise within the task time limit and confirm several target enterprises based on the customer portrait.

[0047] The specific implementation process of this embodiment includes:

[0048] In this embodiment, by setting up a demand response module to meet the portrait construction requirements proposed by the supplier, the supplier does not need to collect data from each enterprise to build a customer portrait by itself; by setting up a data screening module, the selected enterprises and corresponding enterprise data sets that meet the portrait construction requirements are screened in the enterprise database, so that the acquired enterprises and enterprise data are more accurate, which improves work efficiency and reduces analysis costs; by setting up a portrait construction module, the customer portrait of the enterprise is constructed according to the enterprise data set of the selected enterprise, so as to further obtain the target enterprise. The efficiency and effect of building customer portraits are improved. The technical problem of how to improve the utilization rate of data while reducing the analysis cost of data in the process of building customer portraits of enterprises in the prior art is solved.

[0049] In this embodiment, the management system further includes: a data collection module, which is used to collect several items of enterprise data of the enterprise according to a preset period, and store each item of enterprise data in the enterprise database after marking the enterprise tag. The data collection module collects enterprise data of the enterprises in the enterprise database according to the preset period and updates it on time. This makes the customer portrait more accurate in the subsequent customer portrait construction process.

[0050] It should be noted that, in this embodiment, when the data module collects enterprise data, the data collection module is connected to the data interfaces of multiple data sources to collect several items of enterprise data of the enterprise, and mutually verify the same enterprise data of multiple data sources; when the same enterprise data is inconsistent in multiple data sources, the enterprise data in the data source with the highest confidence is taken and entered into the enterprise database, or is entered into the enterprise database after manual verification.

[0051] In this embodiment, the data screening module is used to screen and obtain a number of candidate enterprises in the enterprise database according to the service data, including:

[0052] Based on the service type, obtain type filtering conditions; service types include raw material supply, raw material processing, software operation, transportation and distribution, and financial insurance;

[0053] Based on the service scope, obtain the scope filtering conditions; the service scope includes industrial field, agricultural field, financial field and life field;

[0054] Based on service capabilities, obtain capability screening conditions; service capabilities include processing and production capabilities, transportation and distribution capabilities, software operation capabilities, and financial risk-bearing capabilities;

[0055] Based on the type screening condition, the range screening condition and the capability screening condition, the enterprises in the enterprise database are screened to obtain a number of candidate enterprises.

[0056] In this embodiment, each supplier has its corresponding service type, service scope and service capability; type screening conditions are constructed according to the service type, scope screening conditions are constructed according to the service scope, and capability screening conditions are constructed according to the service capability. The selected enterprises are obtained in the enterprise database through the type screening conditions, scope screening conditions and capability screening conditions.

[0057] In this embodiment, the data screening module obtains the enterprise data set of each enterprise to be selected, including:

[0058] Get basic data items;

[0059] Based on the service type and service scope, identify the associated data items;

[0060] Based on the basic data items and the associated data items, an enterprise data set corresponding to each to-be-selected enterprise is obtained in the enterprise database.

[0061] In this embodiment, each time the enterprise data set is obtained, basic data items are included, and the basic data items include the number of social security personnel, registered capital, etc.

[0062] Each service type and service scope corresponds to different associated data items. For example, a supplier's service type is raw material supply, and its service scope is steel product processing in the industrial field; its corresponding associated data items include the company's processing output, processing type, product type, raw material type, and operating conditions.

[0063] In this embodiment, the portrait construction module is used to obtain the customer portrait of each candidate enterprise according to the enterprise data set of the candidate enterprise within the task time limit, including:

[0064] Acquire a portrait relationship database, wherein the portrait relationship database includes basic data items, associated data items, customer portraits, and mapping relationships between the basic data items, associated data items, and customer portraits;

[0065] Obtain a customer portrait analysis model by training a neural network model through a portrait relational database;

[0066] The enterprise data set of the candidate enterprise is input into the customer portrait analysis model to obtain the customer portrait of the candidate enterprise; the customer portrait includes consumption capacity, financial risk index and business stability index.

[0067] In this embodiment, the portrait relationship database is obtained through big data and expert experience. The neural network model includes an input layer, a hidden layer, and an output layer; the input layer includes a plurality of input neurons, the number of which is the same as the number of basic data items and associated data items; the hidden layer includes a plurality of hidden neurons; the input layer includes a plurality of output neurons, the number of which is the same as the number of data items in the customer portrait, and in this embodiment, the number of output neurons is 3.

[0068] In this embodiment, the training process of the customer portrait analysis model includes:

[0069] Step 1: Divide the portrait relation database into a training set and a verification set;

[0070] Step 2: Initialize the weight parameters of hidden neurons;

[0071] Step 3: Forward propagate the data in the training set through the neural network to obtain the customer prediction portrait;

[0072] Step 4: Calculate the difference between the predicted result and the actual customer profile through the loss function;

[0073] Step 5: Update the weight parameters and offsets of hidden neurons through the back propagation algorithm according to the gradient of the loss function;

[0074] Step 6. Repeat steps 3 to 5 until the preset number of times is reached or the difference value is less than the set difference value to obtain the customer portrait analysis model.

[0075] In this embodiment, the calculation formula of the difference value is as follows:

[0076]

[0077] Among them, S is the difference value, m is the number of samples in the training set; X j is the true value of the consumption capacity of the jth sample; XY j is the predicted result of the consumption capacity of the jth sample; C j is the true value of the financial risk index of the jth sample; CY j is the prediction result of the financial risk index of the jth sample; W j is the true value of the business stability index of the jth sample; WY j is the predicted result of the operating stability index of the jth sample.

[0078] After the training is completed, the customer portrait analysis model is verified through the verification set, the F1 score of the customer portrait analysis model is obtained, the ROC curve is drawn, and the area under the ROC curve AUC value is calculated; in this embodiment, when the F1 score is greater than the preset score and the AUC value is greater than the set threshold, the customer portrait analysis model training is judged to be completed.

[0079] In this embodiment, consumption capacity refers to the total price of the supplier's services that the enterprise is expected to consume; the financial risk index refers to the loss that the enterprise expects to suffer from delayed payment; the longer the payment delay, the greater the loss and the higher the financial risk index; the business stability index refers to the stability of the amount of services that the enterprise expects to consume in subsequent production cycles; in this embodiment, the larger the business stability index, the worse the stability of the amount of services consumed by the enterprise in subsequent production cycles. Production cycles include days, weeks, months, quarters and years.

[0080] According to another embodiment of the present invention, the profile building module identifies several target enterprises according to the customer profile, including:

[0081] S1: Construct a fitness function based on the customer profile of the selected enterprise;

[0082] S2: Based on a number of candidate companies, randomly obtain a number of combination plans and form an initial population;

[0083] S3: Based on the fitness function, obtain the fitness corresponding to each combination scheme in the initial population;

[0084] S4: Perform iterative optimization based on fitness, obtain the target combination plan, and use each candidate enterprise in the target combination plan as the target enterprise.

[0085] In this embodiment, the fitness function is calculated as follows:

[0086]

[0087] Among them, K is the fitness function; α i is the weight coefficient corresponding to the i-th selected enterprise; X i is the consumption capacity of the i-th candidate enterprise, C i is the financial risk index corresponding to the i-th candidate enterprise; W i is the operating stability index of the ith candidate enterprise; n is the number of candidates in the combination scheme; e, f and g are all fixed coefficients.

[0088] It should be noted that, in this embodiment, where α i is the weight coefficient corresponding to the i-th candidate enterprise. Some combination schemes have the same candidate enterprises, but the weight coefficients corresponding to each candidate enterprise are different.

[0089] In this embodiment, the weight coefficient corresponding to each candidate enterprise is accurate to one or two decimal places.

[0090] By setting the weight coefficient, more combination schemes can be provided, and the best combination of candidate enterprises can be obtained, which is convenient for suppliers to make decisions. Therefore, in this embodiment, the combination scheme is recorded as (F, n) = [(QY1, QY2, QY3, ..., QY n ),(α1,α2,α3,…,α n )], where QY n For the nth enterprise.

[0091] It should be noted that, in the embodiment of the present application, the number n of enterprises to be selected in the combination scheme can be preset, or the best combination scheme under the corresponding number can be obtained according to different numbers n of enterprises to be selected for selection by suppliers.

[0092] In some embodiments, the number n of enterprises to be selected includes 3, 4 and 5; the best combination schemes when the number of enterprises to be selected is 3, 4 and 5 are obtained respectively; they are respectively recorded as combination scheme (F, 3), combination scheme (F, 4) and combination scheme (F, 5); suppliers make decisions based on combination scheme (F, 3), combination scheme (F, 4) and combination scheme (F, 5).

[0093] In another embodiment, the number n of enterprises to be selected includes multiple ones, and the best combination schemes are obtained when the number of enterprises to be selected is 3, 4, and 5, respectively; they are respectively recorded as combination scheme (F, 3), combination scheme (F, 4) and combination scheme (F, 5); then, the best scheme is obtained among combination scheme (F, 3), combination scheme (F, 4) and combination scheme (F, 5) by calculating the fitness method and feeding back to the supplier.

[0094] In this embodiment, iterative optimization is performed based on fitness to obtain a target combination solution, including:

[0095] S10: Sort the combination schemes of the initial population in descending order according to the fitness function;

[0096] S11: Select several combinations with the highest fitness rankings for mutation and crossover to form a new generation of population;

[0097] S12: Obtain the fitness of several combination schemes of the new generation population, and sort them in descending order of fitness;

[0098] S13: Repeat steps S11 to S13 until the preset conditions are met, the iteration ends, and the combination scheme with the highest fitness in the contemporary population is selected as the target combination scheme.

[0099] In this embodiment, the preset conditions include that the number of iterations reaches a preset number or there is a combination scheme in the population whose fitness is greater than the set fitness.

[0100] In some other embodiments, the preset condition also includes the expiration of the task deadline.

[0101] In this embodiment, the supplier can pre-set the demand time in the portrait construction demand according to the current service situation, and the demand time corresponds to the task time limit; in this embodiment, the task time limit is earlier than the demand time. For example, if the contract of a certain enterprise expires after a period of time, it will lead to excess service capacity if the target enterprise is not found in advance. Before this, the supplier can set multiple time nodes as the demand time, so as to obtain the target combination plan under the current situation at each demand time, which is convenient for the supplier to make decisions.

[0102] It should be noted that, in this article, relational terms such as "first" and "second" are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.

[0103] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. A customer portrait information data management system, characterized by: The management system comprises: A demand response module, which is used to respond to the demand for portrait building and create a new portrait building task; the portrait building task includes a task number, a task time limit, and service data; the service data includes the service type, service scope, and service capability of the supplier; A data screening module, the data screening module is used to screen and obtain a number of candidate enterprises and a set of enterprise data of each candidate enterprise in the enterprise database according to the service data; the enterprise data set includes the enterprise name, enterprise contact person and enterprise operation data; A portrait construction module is used to obtain the customer portrait of each candidate enterprise based on the enterprise data set of the candidate enterprise within the task time limit and confirm several target enterprises based on the customer portrait.

2. A customer portrait information data management system as claimed in claim 1, characterized in that: The data screening module is used to screen and obtain several candidate enterprises in the enterprise database according to the service data, including: Based on the service type, obtain type filtering conditions; service types include raw material supply, raw material processing, software operation, transportation and distribution, and financial insurance; Based on the service scope, obtain the scope filtering conditions; the service scope includes industrial field, agricultural field, financial field and life field; Based on service capabilities, obtain capability screening conditions; service capabilities include processing and production capabilities, transportation and distribution capabilities, software operation capabilities, and financial risk-bearing capabilities; Based on the type screening condition, the range screening condition and the capability screening condition, the enterprises in the enterprise database are screened to obtain a number of candidate enterprises.

3. A customer portrait information data management system as claimed in claim 2, characterized in that: The data screening module obtains the enterprise data set of each enterprise to be selected, including: Get basic data items; Based on the service type and service scope, identify the associated data items; Based on the basic data items and the associated data items, an enterprise data set corresponding to each to-be-selected enterprise is obtained in the enterprise database.

4. A customer portrait information data management system as claimed in claim 3, characterized in that: The portrait construction module is used to obtain the customer portrait of each candidate enterprise based on the enterprise data set of the candidate enterprise within the task time limit, including: Acquire a portrait relationship database, wherein the portrait relationship database includes basic data items, associated data items, customer portraits, and mapping relationships between the basic data items, associated data items, and customer portraits; Obtain a customer portrait analysis model by training a neural network model through a portrait relational database; The enterprise data set of the candidate enterprise is input into the customer portrait analysis model to obtain the customer portrait of the candidate enterprise; the customer portrait includes consumption capacity, financial risk index and business stability index.

5. A customer portrait information data management system as claimed in claim 4, characterized in that: The portrait building module identifies several target companies based on the customer portrait, including: Construct a fitness function based on the customer profile of the candidate enterprise; According to a number of candidate enterprises, a number of combination plans are randomly obtained and an initial population is formed; Based on the fitness function, obtain the fitness corresponding to each combination scheme in the initial population; Iterative optimization is performed based on fitness to obtain the target combination plan, and each candidate enterprise in the target combination plan is used as the target enterprise.

6. A customer portrait information data management system as claimed in claim 5, characterized in that: The fitness function is calculated as follows: Among them, K is the fitness function; α i is the weight coefficient corresponding to the i-th selected enterprise; X i is the consumption capacity of the i-th candidate enterprise, C i is the financial risk index corresponding to the i-th candidate enterprise; W i is the operating stability index of the ith candidate enterprise; n is the number of candidates in the combination scheme; e, f and g are all fixed coefficients.

7. A customer portrait information data management system as claimed in claim 5, characterized in that: Perform iterative optimization based on fitness to obtain the target combination solution, including: S10: Sort the combination schemes of the initial population in descending order according to the fitness function; S11: Select several combinations with the highest fitness rankings for mutation and crossover to form a new generation of population; S12: Obtain the fitness of several combination schemes of the new generation population, and sort them in descending order of fitness; S13: Repeat steps S11 to S13 until the preset conditions are met, the iteration ends, and the combination scheme with the highest fitness in the contemporary population is selected as the target combination scheme.

8. A customer portrait information data management system as claimed in claim 7, characterized in that: The preset conditions include that the number of iterations reaches a preset number or there is a combination scheme in the population whose fitness is greater than the set fitness.

9. A customer portrait information data management system as claimed in claim 8, characterized in that: The preset conditions also include the expiration of the task deadline.

10. A customer portrait information data management system according to claim 1, characterized in that: The management system also includes: a data collection module, which is used to collect several items of enterprise data of the enterprise according to a preset period, and mark each item of enterprise data with an enterprise mark and store it in the enterprise database.

Images