Marketing strategy processing method and device, equipment, storage medium and program product

Abstract

The invention provides a marketing strategy processing method and device, equipment, a storage medium and a program product, and relates to the field of big data. The method comprises the steps of obtaining supply chain data information of a supply chain where a first customer is located; the supply chain data information comprises financial data, financial interaction data and supply chain quality data, and the supply chain quality data is used for representing the quality of a supply chain; performing data preprocessing on the supply chain data information; a plurality of prediction models are used to acquire a plurality of initial alternative supply chains according to the preprocessed supply chain data information, and an alternative supply chain is acquired from the plurality of initial alternative supply chains; and determining a target supply chain based on the quality data of the alternative supply chain so as to complete precise marketing of customers on the target supply chain. The supply chain data information of the supply chain where the first customer is located is obtained, the high-value supply chain is screened out, the marketing strategy is generated for the customer in the target supply chain, accurate putting is achieved, and efficiency is high.

Description

Technical Field

[0001] The present application relates to the field of big data, and in particular to a marketing strategy processing method, device, equipment, storage medium and program product. Background Art

[0002] With the continuous development of globalization and informatization, the supply chain relationship between enterprises has become increasingly complex. Supply chain finance refers to a model in which banks provide comprehensive financial services through overall analysis of core enterprises and their upstream and downstream enterprises in the supply chain. Compared with traditional single enterprise financing, supply chain finance can more comprehensively assess the credit risk and capital needs of enterprises.

[0003] Traditional analysis and marketing for a single enterprise is highly efficient but not very accurate. Therefore, how to leverage the supply chain relationships between enterprises to tap into the core value of the supply chain, develop supply chain finance, develop appropriate marketing strategies, and accurately deliver appropriate financial products to the required nodes in the supply chain are opportunities and challenges currently facing commercial banks. Summary of the invention

[0004] The present application provides a marketing strategy processing method, device, equipment, storage medium and program product to solve the technical problem of how to tap the value of the supply chain.

[0005] In a first aspect, the present application provides a marketing strategy processing method, the method comprising:

[0006] Obtaining supply chain data information of the supply chain where the first customer is located; the supply chain data information includes financial data, financial interaction data and supply chain quality data, and the supply chain quality data is used to characterize the quality of the supply chain;

[0007] Pre-process the supply chain data information;

[0008] Using multiple prediction models, respectively, based on the preprocessed supply chain data information, multiple initial candidate supply chains are obtained, and alternative supply chains are obtained from the multiple initial candidate supply chains;

[0009] Determine the target supply chain based on the quality data of the alternative supply chains;

[0010] Generate marketing strategies for customers along the target supply chain.

[0011] In one embodiment, obtaining supply chain data information of the supply chain where the first customer is located specifically includes:

[0012] Building a supply chain graph data model of the first customer based on the transaction data of the first customer;

[0013] The supply chain of the first customer is obtained according to the supply chain graph data model; a node on the supply chain of the first customer is a customer, wherein the attribute data of the node is financial data, and the edges between adjacent nodes are used to represent financial interaction data;

[0014] Supply chain quality data is obtained based on attribute data of nodes in the supply chain of the first customer and edges between adjacent nodes.

[0015] In one embodiment, the attribute data of the node includes first financial data related to credit and second financial data related to financial capability; and the supply chain quality data is obtained according to the attribute data of the node in the supply chain of the first customer and the edges between adjacent nodes, specifically including:

[0016] If the first financial data related to credit indicates that there is a credit problem in the supply chain, the quality data is determined according to a quotient obtained by dividing the sum of the second financial data by the product of the first financial data;

[0017] If the first financial data related to credit indicates that there is no credit problem in the supply chain, the quality data is determined based on the sum of the second financial data.

[0018] In one embodiment, the supply chain data information further includes public data of customers in the supply chain, and the method further includes:

[0019] Obtain web page data related to customers in the supply chain within a preset historical period;

[0020] The content of web page data is analyzed through large-scale language analysis models or language processing technology to obtain public data of customers in the supply chain.

[0021] In one embodiment, determining a target supply chain based on quality data of candidate supply chains specifically includes:

[0022] Query quality data of alternative supply chains;

[0023] Sort the quality data of the alternative supply chains from large to small, and select the alternative supply chains that meet the requirements as the target supply chains.

[0024] In one embodiment, before generating a marketing strategy for a customer in a target supply chain, the method further includes:

[0025] If there is a first customer belonging to the bank system in the target supply chain, the bank's private profile data and public data will be supplemented for the first customer;

[0026] If there is a second customer in the target supply chain that is not part of the banking system, supplement the public data for the second customer;

[0027] Based on the supplemented target supply chain, generate marketing strategies for customers on the target supply chain.

[0028] In one embodiment, a marketing strategy is generated for a customer in a target supply chain, specifically comprising:

[0029] Match corresponding marketing outlets for customers in the target supply chain;

[0030] The marketing tasks are issued to the electronic equipment of the corresponding marketing outlets, and the marketing tasks include: the marketing strategy plan for the corresponding customers.

[0031] In one embodiment, matching corresponding marketing outlets for customers in a target supply chain includes:

[0032] If there is a first customer belonging to the banking system in the target supply chain, the bank where the first customer opens an account will be used as the corresponding marketing outlet;

[0033] If there is a second customer who does not belong to the banking system in the target supply chain, then at least one candidate marketing outlet is determined based on the distance between the second customer and each marketing outlet;

[0034] Determine the marketing reliability of the candidate marketing outlets based on their basic data;

[0035] Based on the marketing reliability, a marketing outlet corresponding to the second customer is determined from the candidate marketing outlets.

[0036] In one embodiment, using multiple prediction models, obtaining multiple initial candidate supply chains based on preprocessed supply chain data information, and obtaining candidate supply chains from the multiple initial candidate supply chains specifically includes:

[0037] Each prediction model generates a data package based on the preprocessed supply chain data information, and the data package includes multiple initial candidate supply chains;

[0038] The alternative supply chains are obtained by finding the intersection of multiple data packets.

[0039] In a second aspect, the present application provides a marketing strategy processing device, comprising:

[0040] an acquisition module, used to acquire supply chain data information of the supply chain where the first customer is located; the supply chain data information includes financial data, financial interaction data and supply chain quality data, and the supply chain quality data is used to characterize the quality of the supply chain; and is also used to use multiple prediction models to acquire multiple initial candidate supply chains based on the preprocessed supply chain data information, and to acquire candidate supply chains from the multiple initial candidate supply chains;

[0041] A data processing module is used to pre-process the supply chain data information;

[0042] A determination module, used for determining a target supply chain based on quality data of the candidate supply chains;

[0043] The generation module is used to generate marketing strategies for customers in the target supply chain.

[0044] In a third aspect, the present application provides an electronic device, comprising: a processor, and a memory communicatively connected to the processor;

[0045] Memory stores computer-executable instructions;

[0046] The processor executes the computer-executable instructions stored in the memory to implement any of the above methods.

[0047] In a fourth aspect, the present application provides a computer-readable storage medium, in which computer-executable instructions are stored. When the computer-executable instructions are executed by a processor, they are used to implement any of the above methods.

[0048] In a fifth aspect, the present application provides a computer program product, including a computer program, which implements any of the above methods when executed by a processor.

[0049] The present application provides a marketing strategy processing method, device, equipment, storage medium and program product, the method comprising: obtaining supply chain data information of the supply chain where the first customer is located; the supply chain data information includes financial data, financial interaction data and supply chain quality data, and the supply chain quality data is used to characterize the quality of the supply chain; preprocessing the supply chain data information; using multiple prediction models, respectively, according to the preprocessed supply chain data information, obtaining multiple initial alternative supply chains, and obtaining alternative supply chains from multiple initial alternative supply chains; determining the target supply chain based on the quality data of the alternative supply chain to complete precision marketing for customers on the target supply chain. The present application obtains the supply chain data information of the supply chain where the first customer is located, screens out the alternative supply chain through data processing and multiple prediction models, determines the target supply chain in the alternative supply chain according to the quality data of the supply chain, selects high-value supply chains, and generates marketing strategies for customers in the target supply chain, so as to achieve precise delivery, high efficiency and accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0051] Figure 1 A flowchart of a marketing strategy processing method provided in one embodiment of the present application;

[0052] Figure 2 A flowchart of obtaining supply chain data information of a supply chain where a first customer is located, provided in an embodiment of the present application;

[0053] Figure 3 A flowchart of a method before generating a marketing strategy for a customer in a target supply chain provided by an embodiment of the present application;

[0054] Figure 4 A flow chart of a method for matching corresponding marketing outlets for customers in a target supply chain provided in an embodiment of the present application;

[0055] Figure 5 A schematic diagram of prediction model training is provided for an embodiment of the present application;

[0056] Figure 6 A marketing strategy processing device is provided in one embodiment of the present application.

[0057] The above drawings have shown clear embodiments of the present application, which will be described in more detail later. These drawings and text descriptions are not intended to limit the scope of the present application in any way, but to illustrate the concept of the present application to those skilled in the art by referring to specific embodiments. DETAILED DESCRIPTION

[0058] Exemplary embodiments will be described in detail herein, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The implementations described in the following exemplary embodiments do not represent all implementations consistent with the present application. Instead, they are merely examples of devices and methods consistent with some aspects of the present application as detailed in the appended claims.

[0059] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, storage, use, processing, transmission, provision, disclosure and application of the relevant data comply with the relevant laws, regulations and standards of the relevant countries and regions, take necessary confidentiality measures, do not violate public order and good morals, and provide corresponding operation entrances for users to choose to authorize or refuse.

[0060] In addition, this application involves conducting big data analysis of user information (including but not limited to personal biometrics, identity data, consumption data, asset data, electronic terminal operation data, etc.), and using artificial intelligence technology to make automated decisions, and making technical solutions that have a significant impact on personal rights and interests based on the results of automated decisions. The application provides users with corresponding operation entrances for them to choose to agree or reject the results of automated decisions; if the user chooses to reject, the expert decision-making process will be entered.

[0061] It should be noted that the marketing strategy processing method, device, equipment, storage medium and program product provided in this application can be used in the field of big data, and can also be used in any field other than big data. The application field of the marketing strategy processing method, device, equipment, storage medium and product in this application is not limited.

[0062] Commercial banks should fully tap the core value of the supply chain, formulate appropriate marketing strategies, and accurately deliver appropriate financial products to the required nodes in the supply chain. On the premise of meeting the financing and asset appreciation needs of the entire supply chain, they should accurately analyze the value of the supply chain and better serve the companies in the supply chain.

[0063] The marketing strategy processing method provided in this application includes: obtaining supply chain data information of the supply chain where the first customer is located; the supply chain data information includes financial data, financial interaction data and supply chain quality data, and the supply chain quality data is used to characterize the quality of the supply chain; preprocessing the supply chain data information; using multiple prediction models, respectively, according to the preprocessed supply chain data information, obtaining multiple initial alternative supply chains, and obtaining alternative supply chains from multiple initial alternative supply chains; determining the target supply chain based on the quality data of the alternative supply chain to complete precision marketing for customers on the target supply chain. This application obtains the supply chain data information of the supply chain where the first customer is located, screens out alternative supply chains through data processing and multiple prediction models, determines the target supply chain in the alternative supply chain according to the quality data of the supply chain, selects high-value supply chains, and generates marketing strategies for customers in the target supply chain to achieve precise delivery, high efficiency and accuracy.

[0064] The technical solution of the present application and how the technical solution of the present application solves the above-mentioned technical problems are described in detail below with specific embodiments. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments. The embodiments of the present application will be described below in conjunction with the accompanying drawings.

[0065] like Figure 1 As shown, Figure 1 A flowchart of a marketing strategy processing method provided in an embodiment of the present application. The marketing strategy processing method includes the following steps:

[0066] Step S102: Acquire supply chain data information of the supply chain where the first customer is located; the supply chain data information includes financial data, financial interaction data and supply chain quality data, and the supply chain quality data is used to characterize the quality of the supply chain.

[0067] Specifically, collect and integrate the relevant data of the supply chain where the first customer is located, and the supply chain data information is recorded as matrix D n*m , the “rows” in the matrix are all the extracted supply chain samples, and the “columns” are the data characteristics of the entire supply chain (including Internet data).

[0068] Step S104: pre-process the supply chain data information.

[0069] Specifically, due to D n*m The data features used in are all continuous numerical types, so there is no need to use encoding technology for conversion. If there are missing values, they are supplemented by prediction, that is, the part with values ​​in the dimension where the missing value is located is used as "Y", and other features are trained to fit "Y", and finally the prediction filling is completed. Then the data envelopment analysis (DEA) method is used to analyze the statistical information and distribution of the data in each dimension. For data with many outliers, square root or log is used for transformation. If necessary, continuous features can also be binned. At the end of data preprocessing, the Z-score method is used to standardize the data. The formula is shown below. Without changing the distribution of the original data, the feature value space is compressed, the model stability is increased, and the training efficiency is improved.

[0070]

[0071] Where Z is the converted score of data x, μ is the average of the feature values ​​of the dimension where x is located, and δ is the standard deviation. The collected supply chain data information is cleaned and standardized to ensure the accuracy and consistency of the data. Using the embedded feature selection method, feature selection and model training are completed in the same optimization process, and the machine learning algorithm automatically determines the importance of the feature.

[0072] Step S106: using multiple prediction models, respectively obtaining multiple initial candidate supply chains according to the preprocessed supply chain data information, and obtaining alternative supply chains from the multiple initial candidate supply chains.

[0073] Specifically, different prediction models (such as machine learning models and statistical models) are used to analyze the supply chain data. Different prediction models generate multiple initial lists of candidate supply chains, and the candidate supply chains that meet specific criteria are screened out from these initial candidate supply chains.

[0074] Step S108: Determine the target supply chain based on the quality data of the candidate supply chains.

[0075] Step S110: Generate a marketing strategy for customers in the target supply chain.

[0076] Specifically, we develop effective marketing strategies for customers in the target supply chain, design customized financial products and services to meet customers' needs in financing, risk management, and operational optimization; we implement precision marketing and promote these products and services to customers through appropriate channels and methods.

[0077] This application screens out high-value target supply chains and generates corresponding marketing strategies for customers in the target supply chains to achieve precise marketing delivery. Compared with traditional analysis and marketing for a single enterprise, it is highly efficient and can accurately identify customers in the target supply chain to promote business growth.

[0078] In one embodiment, step S102 specifically includes the following steps: Figure 2 As shown, Figure 2 A flowchart of obtaining supply chain data information of a first customer's supply chain provided in an embodiment of the present application:

[0079] Step S202: construct a supply chain graph data model of the first customer based on the transaction data of the first customer.

[0080] Specifically, commercial banks mostly store the transaction data of corporate customers in relational databases, which makes it difficult and time-consuming to extract the overall supply chain data. This application constructs a supply chain graph data model that is more suitable for expressing supply relationships.

[0081] Step S204, obtaining the supply chain of the first customer according to the supply chain graph data model; a node on the supply chain of the first customer is a customer, wherein the attribute data of the node is financial data, and the edges between adjacent nodes are used to represent financial interaction data.

[0082] Specifically, the mathematical essence of the supply chain is an open-loop directed graph, which radiates from a certain enterprise to upstream and downstream nodes. The nodes are connected by upstream and downstream relationships (called edges). In the supply chain graph data model, this upstream and downstream relationship is specifically represented by capital flow. The nodes on the supply chain represent the enterprises on the supply chain. The attribute data of the nodes are financial data. The edges between adjacent nodes are used to represent financial interaction data, which includes the number of transactions and transaction amounts.

[0083] Step S206: Acquire supply chain quality data based on the attribute data of the nodes in the supply chain of the first customer and the edges between adjacent nodes.

[0084] In one embodiment, the supply chain graph data model includes the primary relationship graph data, the secondary relationship graph data, etc. of the first customer. This embodiment is for the primary relationship graph data of the first customer i (i=1,2...n), that is, the direct supply chain of the first customer i. Assuming that the direct upstream and downstream of all enterprises are not empty, then the jth direct supply chain of i can be defined as:

[0085]

[0086] Among them, S ij is a 5-tuple, u ij , d ij They represent the upstream and downstream enterprises of the first customer i in the supply chain j. u Refers to the capital flow relationship between i and upstream enterprises, with the direction of capital outflow from i. d It indicates the relationship in which funds flow from the downstream enterprise to the first customer i.

[0087] S ij Each node and relationship above contains attribute data. The attributes of the nodes include financial data in the fields of deposits, loans, wealth management, insurance, funds, foreign exchange, etc. The node attributes of this application are illustrated by taking "annual average daily assets (A)", "annual average daily deposits (S)", "annual average daily contribution of intermediary business (M)", "loan overdue days (O)", and "number of loan defaults (D)" as examples.

[0088] First, we need to set the time period T, which is usually in years. For example, T=[-3,0] represents the past three years. Then the calculation method for the annual average daily asset A of a supply chain is as follows:

[0089]

[0090] in , and They represent the total assets of the first customer's upstream, the first customer, and the first customer's downstream on the i-th day in the past T years. The calculation method of the annual average daily deposit S is similar to that of the annual average daily assets A, and will not be repeated here.

[0091] The intermediary business contribution of the first customer is also one of the core indicators commonly used by commercial banks to measure customer value. The difference earned by commercial banks when customers purchase products sold by commercial banks (such as insurance, precious metals, etc.) is the intermediary business income. The calculation method of the annual average daily intermediary business contribution M is as follows:

[0092]

[0093] Among them, I and C represent the expenditure and the bank's cost of the bank's agency sales products purchased by the first customer and upstream and downstream customers on the i-th day in the past T years, respectively.

[0094] O is the total number of days overdue on-chain customer loans in the past T years. The number of days overdue loans is generally counted on an annual basis. The calculation formula is as follows:

[0095]

[0096] in, , , They respectively represent whether the three adjacent customers in the supply chain are in a loan overdue state on the i-th day in the past T years. If they are in a loan overdue state O, it takes 1, otherwise it takes 0.

[0097] D is the total number of on-chain corporate loan defaults in the past T years, calculated as follows:

[0098]

[0099] The countD() function is used to calculate the number of loan defaults of three adjacent customers in the supply chain within T years, where u represents the upstream enterprise of the first customer, i represents the first customer, and d represents the downstream enterprise of the first customer.

[0100] The edges between adjacent nodes are financial interaction data, which includes supply chain S ij The annual average daily net capital inflow N, supply chain S ij The formula for the annual average daily net capital inflow N is as follows:

[0101]

[0102] Among them, in represents the total amount of funds that have net flowed into a supply chain in the past T years; out represents the total amount of funds that have net flowed out of a supply chain in the past T years.

[0103] In one embodiment, the attribute data of the node includes first financial data related to credit and second financial data related to financial capability; step S206 specifically includes the following steps:

[0104] If the first financial data related to credit indicates that there is a credit problem in the supply chain, the quality data is determined according to a quotient obtained by dividing the sum of the second financial data by the product of the first financial data.

[0105] If the first financial data related to credit indicates that there is no credit problem in the supply chain, the quality data is determined based on the sum of the second financial data.

[0106] In one embodiment, the first financial data includes the number of days overdue for loans (O) and the number of loan defaults (D), and the second financial data includes the annual average daily assets (A), the annual average daily deposits (S), the annual average daily contribution of intermediary business (M) and the annual average daily net capital inflow N. The quality data V of the supply chain is calculated by the objective function, assuming that A, S, M∈[0,+∞), N∈(-∞,+∞), O, D∈[0,+∞). Among them, the annual average daily assets (A), the annual average daily deposits (S), the annual average daily contribution of intermediary business (M) and the annual average daily net capital inflow N are benefit-type characteristics, that is, the larger their values, the better; while the number of days overdue for loans (O) and the number of loan defaults (D) are cost-type characteristics, the smaller their values, the better. The first financial data and the second financial data both refer to the data of the entire supply chain, not just the data of a customer in the supply chain. The calculation formula for quality data is as follows:

[0107]

[0108] If there are customers in the supply chain who do not belong to the bank system, the financial data and financial interaction data of the customer will be partially missing. In this case, the quality data of the supply chain calculated by the objective function will be lower than the actual value. Optionally, the data of customers who do not belong to the bank system can be missing through "graph federated learning" to obtain the prediction information of the required data.

[0109] In one embodiment, the supply chain data information further includes public data of customers in the supply chain, and the method further includes the following steps:

[0110] Get web page data related to customers in the supply chain within a preset historical period.

[0111] The content of web page data is analyzed through large-scale language analysis models or language processing technology to obtain public data of customers in the supply chain.

[0112] Specifically, by analyzing web page content through the "LLM (Large-Scale Language Analysis Model)" or language processing technology, S ij The number of state-owned enterprises, the number of listed companies, the number of specialized and innovative enterprises, the number of positive public opinions, the number of negative public opinions, the total number of awards, the total number of patents, the total number of penalties, etc. The more public data obtained, the easier it is to accurately predict the output of the model.

[0113] In one embodiment, step S108 specifically includes the following steps:

[0114] Query quality data of alternative supply chains;

[0115] Sort the quality data of the alternative supply chains from large to small, and select the alternative supply chains that meet the requirements as the target supply chains.

[0116] Specifically, the target supply chain is a screened high-value supply chain, and the target supply chain is taken as the priority marketing supply chain.

[0117] In one embodiment, before step S110, the following steps are also included: Figure 3 As shown, Figure 3 A flowchart of a method for generating a marketing strategy for a customer in a target supply chain provided in an embodiment of the present application:

[0118] Step S302: If there is a first customer belonging to the bank system in the target supply chain, the bank's private portrait data and public data are supplemented for the first customer.

[0119] Step S304: If there is a second customer in the target supply chain who does not belong to the banking system, then supplement the public data for the second customer;

[0120] Step S306: Generate a marketing strategy for customers in the target supply chain based on the supplemented target supply chain.

[0121] This application provides a more comprehensive decision-making basis for generating marketing strategies by providing targeted supplementary data for the first customer and the second customer.

[0122] In one embodiment, step S110 specifically includes the following steps:

[0123] Match corresponding marketing outlets to customers in the target supply chain.

[0124] The marketing tasks are issued to the electronic equipment of the corresponding marketing outlets, and the marketing tasks include: the marketing strategy plan for the corresponding customers.

[0125] Specifically, the marketing strategy includes financial product and financial business recommendations. The marketing strategy is determined based on actual conditions, and this application does not limit it here.

[0126] In one embodiment, matching corresponding marketing outlets for customers in a target supply chain includes the following steps: Figure 4 As shown, Figure 4 A flow chart of a method for matching corresponding marketing outlets for customers in a target supply chain provided in an embodiment of the present application:

[0127] Step S402: If there is a first customer belonging to the banking system in the target supply chain, the bank where the first customer opens an account is used as the corresponding marketing outlet.

[0128] Step S404: If there is a second customer who does not belong to the banking system in the target supply chain, at least one candidate marketing outlet is determined based on the distance between the second customer and each marketing outlet.

[0129] Step S406: Determine the marketing reliability of the candidate marketing outlets based on the basic data of the candidate marketing outlets.

[0130] Step S408: Determine the marketing outlet corresponding to the second customer from the candidate marketing outlets based on the marketing reliability.

[0131] Specifically, the formula for determining the candidate marketing outlet j is as follows:

[0132]

[0133] Among them, d(x,j) represents the spherical distance between the office address of the second customer x and the candidate marketing outlet j, and θ is the constraint condition (can be multiple) on the candidate marketing outlet, which is used to characterize the marketing reliability of the candidate marketing outlet. For example: the total assets managed by the branch must be greater than a certain value, the size (number of people) of the branch must be greater than a certain value, the evaluation of the branch's corporate marketing ability must be higher than a certain value, and the number of people with credit qualifications in the branch must exceed a certain value, etc.

[0134] The calculation formula for spherical distance is as follows:

[0135]

[0136] Among them, lat and lon are latitude and longitude respectively, and this information can be obtained through query.

[0137] In one embodiment, step S106 specifically includes the following steps:

[0138] Each prediction model generates a data package based on the preprocessed supply chain data information, and the data package includes multiple initial candidate supply chains. The candidate supply chains are obtained by finding the intersection of multiple data packages.

[0139] The supply chain data information is divided into data sets and positive and negative samples are selected. n*m The number of supply chain samples in the matrix is ​​usually large. This application uses random sampling to generate training sets, validation sets, and test sets. For large-scale data sets, it is impossible to manually label positive and negative samples. The calculation formula of the quality data in this application is used as the basis for dividing positive and negative samples. On the premise of ensuring the balance of positive and negative samples in the training set (such as: 1:1, 1:2 or 2:3), the threshold is selected for division. In order to avoid the bias caused by a single model, multiple prediction models (such as: random forest, logistic regression, XGBoost, lightGBT, etc.) are used to train the data set, such as Figure 5 As shown, Figure 5The figure below shows a schematic diagram of the prediction model training for an embodiment of the present application. It is required that the binary classification metrics of each model should meet the expectations (e.g., F1-Score > 0.7 and 0.85 < AUC < 0.95). After the training is completed, the models that do not meet the standards will be iteratively optimized until all models meet the standards. Finally, the initial alternative supply chain samples classified as true by multiple prediction models are intersected to obtain the alternative supply chain.

[0140] The present application provides a marketing strategy processing device, as Figure 6 shown Figure 6 The marketing strategy processing device provided for an embodiment of the present application includes: an acquisition module, configured to acquire the supply chain data information of the first customer's supply chain; the supply chain data information includes financial data, financial interaction data, and supply chain quality data, and the supply chain quality data is used to characterize the quality of the supply chain; it is also configured to use multiple prediction models to respectively obtain multiple initial alternative supply chains according to the preprocessed supply chain data information, and obtain the alternative supply chain from the multiple initial alternative supply chains; a data processing module, configured to perform data preprocessing on the supply chain data information; a determination module, configured to determine the target supply chain based on the quality data of the alternative supply chain; a generation module, configured to generate a marketing strategy for the customers on the target supply chain.

[0141] The present application provides an electronic device, including: a processor, and a memory communicatively connected to the processor; the memory stores computer-executable instructions; the processor executes the computer-executable instructions stored in the memory to implement the method as described in any of the above.

[0142] The present application provides a computer-readable storage medium, in which computer-executable instructions are stored, and when the computer-executable instructions are executed by a processor, they are used to implement the method as described in any of the above.

[0143] The present application provides a computer program product, including a computer program, and when the computer program is executed by a processor, it implements the method as described in any of the above.

[0144] It should be noted that for the foregoing method embodiments, for the sake of simple description, they are all expressed as a series of action combinations. However, those skilled in the art should know that the present application is not limited by the described action sequence, because according to the present application, certain steps can be performed in other sequences or simultaneously. Secondly, those skilled in the art should also know that the embodiments described in the specification are all optional embodiments, and the actions and modules involved are not necessarily essential to the present application.

[0145] It should be further noted that, although the various steps in the flowchart are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence in the order indicated by the arrows. Unless there is a clear description in this article, the execution of these steps is not strictly limited in order, and these steps can be executed in other orders. Moreover, at least a portion of the steps in the flowchart may include multiple sub-steps or multiple stages, and these sub-steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these sub-steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a portion of the sub-steps or stages of other steps.

[0146] It should be understood that the above-mentioned device embodiments are only illustrative, and the device of the present application can also be implemented in other ways. For example, the division of units / modules in the above-mentioned embodiments is only a logical function division, and there may be other division methods in actual implementation. For example, multiple units, modules or components can be combined, or can be integrated into another system, or some features can be ignored or not executed.

[0147] In addition, unless otherwise specified, each functional unit / module in each embodiment of the present application may be integrated into one unit / module, each unit / module may exist physically separately, or two or more units / modules may be integrated together. The above-mentioned integrated unit / module may be implemented in the form of hardware or in the form of a software program module.

[0148] If the integrated unit / module is implemented in the form of hardware, the hardware may be a digital circuit, an analog circuit, etc. The physical implementation of the hardware structure includes but is not limited to transistors, memristors, etc. Unless otherwise specified, the processor may be any appropriate hardware processor, such as a CPU, a GPU, an FPGA, a DSP, an ASIC, etc. Unless otherwise specified, the storage unit may be any appropriate magnetic storage medium or magneto-optical storage medium, such as a resistive random access memory (RRAM), a dynamic random access memory (DRAM), a static random access memory (SRAM), an enhanced dynamic random access memory (EDRAM), a high-bandwidth memory (HBM), a hybrid memory cube (HMC), etc.

[0149] If the integrated unit / module is implemented in the form of a software program module and sold or used as an independent product, it can be stored in a computer-readable memory. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a memory, including a number of instructions to enable a computer device (which can be a personal computer, server or network device, etc.) to execute all or part of the steps of the various embodiments of the present application. The aforementioned memory includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, disk or optical disk and other media that can store program codes.

[0150] In the above embodiments, the description of each embodiment has its own emphasis. For the part not described in detail in a certain embodiment, please refer to the relevant description of other embodiments. The technical features of the above embodiments can be combined arbitrarily. In order to make the description concise, all possible combinations of the technical features in the above embodiments are not described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0151] Those skilled in the art will readily appreciate other embodiments of the present application after considering the specification and practicing the invention disclosed herein. The present application is intended to cover any modification, use or adaptation of the present application, which follows the general principles of the present application and includes common knowledge or customary techniques in the art that are not disclosed in the present application. The specification and examples are intended to be exemplary only, and the true scope and spirit of the present application are indicated by the following claims.

[0152] It should be understood that the present application is not limited to the precise structures that have been described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present application is limited only by the appended claims.

Claims

1. A marketing strategy processing method, characterized in that: The method comprises: Acquire supply chain data information of the supply chain where the first customer is located; the supply chain data information includes financial data, financial interaction data and supply chain quality data, and the supply chain quality data is used to characterize the quality of the supply chain; Performing data preprocessing on the supply chain data information; Using multiple prediction models, respectively, based on the preprocessed supply chain data information, multiple initial candidate supply chains are obtained, and alternative supply chains are obtained from the multiple initial candidate supply chains; determining a target supply chain based on the quality data of the alternative supply chains; Generate marketing strategies for customers in the target supply chain.

2. The marketing strategy processing method according to claim 1, characterized in that: Obtain supply chain data information of the first customer's supply chain, including: Building a supply chain graph data model of the first customer based on the transaction data of the first customer; Acquire the supply chain of the first customer according to the supply chain graph data model; a node on the supply chain of the first customer is a customer, wherein the attribute data of the node is financial data, and the edges between adjacent nodes are used to represent financial interaction data; The supply chain quality data is acquired based on the attribute data of the nodes in the supply chain of the first customer and the edges between the adjacent nodes.

3. The marketing strategy processing method according to claim 2, characterized in that: The attribute data of the node includes first financial data related to credit and second financial data related to financial capability; the acquiring the supply chain quality data according to the attribute data of the node in the supply chain of the first customer and the edges between the adjacent nodes specifically includes: If the first financial data related to credit indicates that the supply chain has a credit problem, determining the quality data according to a quotient obtained by dividing the sum of the second financial data by the product of the first financial data; If the first financial data related to credit indicates that the supply chain has no credit problem, the quality data is determined according to the sum of the second financial data.

4. The marketing strategy processing method according to claim 2, characterized in that: The supply chain data information also includes public data of customers in the supply chain, and the method further includes: Obtaining web page data related to customers in the supply chain within a preset historical period; The content of the web page data is analyzed by a large-scale language analysis model or language processing technology to obtain the public data of the customers in the supply chain.

5. The marketing strategy processing method according to claim 3, characterized in that: The determining of the target supply chain based on the quality data of the candidate supply chain specifically includes: Querying quality data of the alternative supply chain; The quality data of the candidate supply chains are sorted from large to small, and the candidate supply chains that meet the requirements are used as the target supply chains.

6. The marketing strategy processing method according to claim 1, characterized in that: Before generating a marketing strategy for the customers in the target supply chain, the method further includes: If there is a first customer belonging to the bank system in the target supply chain, the bank's private portrait data and public data are supplemented for the first customer; If there is a second customer in the target supply chain who does not belong to the banking system, then supplement the public data for the second customer; Based on the supplemented target supply chain, a marketing strategy is generated for customers on the target supply chain.

7. The marketing strategy processing method according to claim 1, characterized in that: The generating of marketing strategies for customers in the target supply chain specifically includes: Matching corresponding marketing outlets for customers in the target supply chain; A marketing task is issued to the electronic equipment of the corresponding marketing outlet, wherein the marketing task includes: a marketing strategy plan for the corresponding customer.

8. The marketing strategy processing method according to claim 7, characterized in that: The matching of corresponding marketing outlets for customers in the target supply chain includes: If there is a first customer belonging to the banking system in the target supply chain, the bank where the first customer opens an account is used as the corresponding marketing outlet; If there is a second customer who does not belong to the banking system in the target supply chain, determining at least one candidate marketing outlet based on the distance between the second customer and each marketing outlet; Determining the marketing reliability of the candidate marketing outlets based on the basic data of the candidate marketing outlets; Based on the marketing reliability, a marketing outlet corresponding to the second customer is determined from the candidate marketing outlets.

9. The marketing strategy processing method according to claim 1, characterized in that: The method of using multiple prediction models to obtain multiple initial candidate supply chains based on the preprocessed supply chain data information, and obtaining candidate supply chains from the multiple initial candidate supply chains, specifically includes: Each of the prediction models generates a data package based on the preprocessed supply chain data information, wherein the data package includes a plurality of initial candidate supply chains; The candidate supply chain is obtained by finding the intersection of multiple data packets.

10. A marketing strategy processing device, characterized in that: include: An acquisition module, used to acquire supply chain data information of the supply chain where the first customer is located; The supply chain data information includes financial data, financial interaction data and supply chain quality data, wherein the supply chain quality data is used to characterize the quality of the supply chain; and is also used to obtain multiple initial candidate supply chains based on the preprocessed supply chain data information using multiple prediction models, and to obtain candidate supply chains from the multiple initial candidate supply chains; A data processing module, used for preprocessing the supply chain data information; A determination module, configured to determine a target supply chain based on the quality data of the candidate supply chains; A generation module is used to generate marketing strategies for customers in the target supply chain.

11. An electronic device, characterized in that: include: A processor, and a memory communicatively connected to the processor; The memory stores computer-executable instructions; The processor executes the computer-executable instructions stored in the memory to implement the method according to any one of claims 1 to 9.

12. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-executable instructions, which are used to implement the method according to any one of claims 1 to 9 when executed by a processor.

13. A computer program product, characterized in that The invention comprises a computer program, which implements the method according to any one of claims 1 to 9 when being executed by a processor.

Images