Bank customer expansion model application method based on knowledge graph

Abstract

The invention discloses a bank customer expansion model application method based on a knowledge graph, and belongs to the technical field of computer technology application, and the method comprises the steps: obtaining to-be-screened customer data, inputting the to-be-screened customer data into a pre-trained bank customer expansion model, recognizing high-value potential customers, and storing the high-value potential customers in a database; the information of the high-value potential customers is output so as to carry out precise marketing; according to the invention, the trained bank customer expansion model is obtained by deeply mining and constructing the customer transaction association relationship graph, so that the bank customer expansion model is utilized to quickly identify and expand potential customers, realize accurate reaching and marketing, improve the customer expansion efficiency, avoid invalid operation and waste, reduce the marketing cost, and improve the market competitiveness. And through customized marketing strategies and personalized service recommendation, customer requirements are met, and customer satisfaction and loyalty are improved.

Description

Technical Field

[0001] The present invention relates to a knowledge graph-based bank customer development model application method, and belongs to the technical field of computer technology application. Background Art

[0002] With the rapid development of the financial market, banks are facing increasingly fierce competition for customers. Traditional manual analysis and customer development methods rely on manual screening and promotion, which are inefficient and difficult to accurately locate target customers. At the same time, with the rise of big data and artificial intelligence technologies, how to effectively use massive customer data, explore potential connections between customers, and achieve precise customer marketing has become an urgent problem for banks to solve. Summary of the invention

[0003] The purpose of the present invention is to overcome the shortcomings of the prior art, provide a bank customer development model application method based on knowledge graph, establish a data-driven customer development model, explore target marketing customers, integrate the bank's internal customer data, use Kcore and PageRank graph algorithms to deeply explore customer transaction associations, construct a customer transaction association graph, and train a model based on this that can accurately identify and expand potential customers.

[0004] To achieve the above object, the present invention is implemented by adopting the following technical solutions: In a first aspect, the present invention provides a method for applying a bank customer development model based on a knowledge graph, comprising: Obtain customer data to be screened, input the data into the pre-trained bank customer development model, identify high-value potential customers, and output the information of high-value potential customers for precision marketing; The training method of the bank customer development model includes: Obtain basic customer information from the bank's internal system and third-party data sources as a training set; Use customers as nodes and relationships between customers as edges to build a knowledge graph; Establishing a bank customer development model based on the knowledge graph; Use the training set to complete the training of the knowledge graph-based banking customer development model.

[0005] Furthermore, the bank customer development model adopts PageRank algorithm or K-core subgraph algorithm.

[0006] Furthermore, when the bank customer development model adopts the PageRank algorithm, the training process is as follows: Assign the same initial importance value to each node in the knowledge graph, set to 1 / N, where N is the total number of customers; Count the number of outgoing links of each node and construct a transfer matrix, where each row of the transfer matrix represents a node and each column represents a possible jump target; According to the transfer matrix and the importance value of the current node, the importance of each node is iteratively updated, where the new importance value of a node is the sum of the importance of all its incoming nodes, and the importance value of each incoming node is weighted according to its outgoing number. In the iterative process, a damping factor is introduced, and the iteration is continued until the importance of all nodes converges. Output the iterative results so that each node customer obtains a stable importance value to reflect the importance of the customer in the entire knowledge graph.

[0007] Furthermore, when the bank customer development model adopts the K-core subgraph algorithm, the training process is as follows: Determine the minimum degree K required for each node in the specified subgraph in the K-core subgraph algorithm; Apply the K-core subgraph algorithm on the knowledge graph, traverse all nodes in the graph, and calculate the degree of each node; Remove nodes and related edges with degrees less than K until the degrees of all remaining nodes are not less than K; Extract the subgraph whose node degree is greater than or equal to K, which contains the most compact and core set of nodes in the graph; Analyze the properties of the extracted subgraph nodes to identify key nodes, important relationship chains, and potential high-value customer groups.

[0008] Furthermore, the method also includes: preprocessing the training set, including data cleaning, deduplication, and standardization, and extracting key information.

[0009] Furthermore, the method also includes: adding feedback data from precision marketing and newly applied customer data to sample data to update and iterate the bank's customer development model.

[0010] In a second aspect, the present invention provides a bank customer development model application device based on a knowledge graph, comprising: The application module is used to obtain customer data to be screened, input the customer data to be screened into the pre-trained bank customer development model, identify high-value potential customers, and output the information of high-value potential customers for precision marketing; The training method of the bank customer development model includes: Obtain basic customer information from the bank's internal system and third-party data sources as a training set; Use customers as nodes and relationships between customers as edges to build a knowledge graph; Establishing a bank customer development model based on the knowledge graph; Use the training set to complete the training of the knowledge graph-based banking customer development model.

[0011] In a third aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of any of the aforementioned methods.

[0012] In a fourth aspect, the present invention provides a computer device, comprising: Memory, for storing computer programs / instructions; A processor is used to execute the computer program / instructions to implement the steps of any of the aforementioned methods.

[0013] In a fifth aspect, the present invention provides a computer program product, comprising a computer program / instruction, which implements the steps of any one of the aforementioned methods when executed by a processor.

[0014] Compared with the prior art, the present invention has the following beneficial effects: The present invention provides a bank customer development model application method based on knowledge graph, which obtains a trained bank customer development model by deeply mining and constructing a customer transaction association relationship graph, thereby using the bank customer development model to break the traditional marketing model from the bank's existing customers, penetrate and analyze customer transactions and equity investment links, screen out upstream and downstream related companies as target customers for marketing development, and provide new channels for customer development. This method can quickly identify and expand potential customers, achieve accurate contact and marketing, improve customer development efficiency, avoid ineffective operations and waste, reduce marketing costs, meet customer needs through customized marketing strategies and personalized service recommendations, and improve customer satisfaction and loyalty. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 It is a schematic diagram of the method flow of the present invention. DETAILED DESCRIPTION

[0016] The present invention will be further described below in conjunction with the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solution of the present invention, and cannot be used to limit the protection scope of the present invention.

[0017] Embodiment 1: This embodiment introduces a method for applying a bank customer development model based on a knowledge graph, including: Obtain customer data to be screened, input the data into the pre-trained bank customer development model, identify high-value potential customers, and output the information of high-value potential customers for precision marketing; The training method of the bank customer development model includes: Obtain basic customer information from the bank's internal system and third-party data sources as a training set; Take key customers as nodes and transaction relationships between customers as edges, continuously search for companies associated with them, select and apply another type of equity relationship, search, identify and mark the associated companies of this type of relationship, and build a knowledge graph; Establishing a bank customer development model based on the knowledge graph; Use the training set to complete the training of the knowledge graph-based banking customer development model.

[0018] like Figure 1 As shown, the application method of the bank customer development model based on the knowledge graph provided in this embodiment specifically involves the following steps: After determining the design accuracy and training time, the model is trained based on the training set and the test set to obtain the bank customer expansion model of the trained knowledge graph; Through the bank customer development model based on the knowledge graph, high-value potential customers are screened out from the bank's core credit customers, strategic customers and fund customers as target customers for marketing, so as to achieve accurate identification and development of potential customers.

[0019] The training set and test set acquisition method are as follows: The valid customer sample data is divided into a training set and a test set; the training set construction process is as follows: Data collection: Collect basic customer information from the bank's internal system and third-party public data sources, including company name, legal person, registered capital, business scope, equity relationship, etc.; Data preprocessing: Clean, remove duplicates, and standardize the collected data to ensure data quality and consistency. At the same time, extract key information, such as customer number, transaction information, etc. Relationship extraction: Identify and extract relationships between entities, such as transaction relationships, equity relationships, etc.

[0020] Knowledge graph construction: Take key customers as nodes, transaction relationships between customers as edges, and key enterprises as the starting point. Select the association method, and continuously search downward through the association direction to find all associated enterprises of this type of association method, and obtain the relationship graph of this type of association. After completion, select another association method, use the same method to search, identify and mark, obtain the corresponding association relationship graph, repeat the above steps until all association methods are selected, thus forming a complete customer relationship graph, and clean and denoise the graph to ensure data accuracy and consistency.

[0021] Training set annotation: Annotate the nodes or edges in the graph, such as marking potential customers, important customer chains, etc.

[0022] The bank customer development model of the knowledge graph adopts PageRank or K-core subgraph algorithm.

[0023] The training process of the sample data is as follows: PageRank algorithm application: The Pagerank algorithm is introduced to evaluate the influence of customers in the network and identify core customers with high influence. The specific process is as follows: Initialization: Assign the same initial importance value to each node in the graph, set to 1 / N, where N is the total number of customers.

[0024] Construct a transfer matrix: Count the number of outgoing links of each node and construct a transfer matrix. Each row of the transfer matrix represents a node, and each column represents a possible jump target.

[0025] Iterative calculation: In each iteration, the importance of each node is updated according to the transfer matrix and the importance value of the current node. Specifically, the new importance value of a node is the sum of the importance of all its incoming nodes, and the importance value of each incoming node is weighted according to its outgoing number. During the iteration process, a damping factor (set to 0.85) is introduced, and the iteration is continued until the importance of all nodes converges.

[0026] Result output: Ultimately, each node customer obtains a stable importance value, which reflects the importance of the customer in the entire graph.

[0027] K-core subgraph algorithm training process: The K-core subgraph algorithm is used to identify closely related customer groups and form potential high-value customer groups. The specific process is as follows: Parameter setting: Determine the K value, which is used to specify the minimum degree required for each node in the subgraph in the K-core subgraph algorithm; Apply the K-core subgraph algorithm on the knowledge graph, traverse all nodes in the graph, and calculate the degree of each node; Remove nodes and related edges with degrees less than K until the degrees of all remaining nodes are not less than K; Extract the subgraph whose node degree is greater than or equal to K, which contains the most compact and core set of nodes in the graph; Analyze the attributes of the extracted subgraph nodes, such as basic enterprise information, upstream and downstream enterprises with high-frequency transactions, etc., to identify key nodes, important relationship chains and potential high-value customer groups.

[0028] The bank customer development model updates and iterates the existing bank customer development model by analyzing the marketing results after a period of time and adding newly applied customers to the sample data.

[0029] The contents involved in the above embodiment are described below in conjunction with a preferred embodiment.

[0030] Step 1: Data integration and preprocessing Data source: Collect internal customer data of the bank, including basic information of corporate customers (company name, legal person, registered capital, business scope, equity relationship), transaction records (such as transaction time, transaction amount, counterparty, etc.) and other information.

[0031] Data cleaning: Clean the collected data to remove duplicate data, invalid data and abnormal data to ensure the accuracy and consistency of the data.

[0032] Data formatting: The cleaned data is stored in a unified format for subsequent processing and analysis.

[0033] Step 2: Build a knowledge graph Entity recognition: Identify company customer entities from cleaned data and annotate datasets Relationship extraction: Extract the relationships between customers, such as transactions, equity, and upstream and downstream relationships, based on transaction records and equity relationships; Graph construction: Represent customer entities and relationships in the form of a graph to build a customer transaction relationship graph; Step 3: Algorithm Application Use PageRank algorithm and K-core subgraph algorithm; Algorithm parameter setting: set the parameters of PageRank algorithm and K-core subgraph algorithm, such as number of iterations, damping factor, and minimum coreness threshold; PageRank value calculation: Calculate the value of each customer node in the graph to evaluate the importance of the customer; calculate the coreness of each customer node in the graph to identify customers with higher coreness; Result sorting: Sort customers according to PageRank values, and give priority to customers with higher PageRank values; select core customers as the bank's key focus and development targets based on core degree calculation results; Step 4: Model training and optimization Feature selection: Based on the analysis results of the above graph algorithm, select customer coreness, PageRank value, etc. as model features; Model building: Use machine learning algorithms to build customer precision expansion models and adjust parameters to improve prediction performance and accuracy; Step 5: Accurately reach and expand Customer screening: Use the trained model to screen potential customers and determine the priority customer list; Effect evaluation: Use online platforms to accurately reach and market to customers, and evaluate and analyze the effects of marketing activities, including indicators such as customer conversion rate. Adjust marketing strategies and model parameters based on evaluation results, and optimize and iterate subsequent marketing activities.

[0034] Embodiment 2: This embodiment provides a bank customer development model application device based on knowledge graph, including: The application module is used to obtain customer data to be screened, input the customer data to be screened into the pre-trained bank customer development model, identify high-value potential customers, and output the information of high-value potential customers for precision marketing; The training method of the bank customer development model includes: Obtain basic customer information from the bank's internal system and third-party data sources as a training set; Use customers as nodes and relationships between customers as edges to build a knowledge graph; Establishing a bank customer development model based on the knowledge graph; Use the training set to complete the training of the knowledge graph-based banking customer development model.

[0035] The specific functional implementation of each of the above modules can be found in the relevant contents of the method in Example 1 and will not be elaborated here.

[0036] Embodiment 3, this embodiment provides a computer-readable storage medium, on which a computer program is stored, and when the program is executed by a processor, the steps of any one of the methods described in Embodiment 1 are implemented.

[0037] Embodiment 4: This embodiment provides a computer device, including: Memory, for storing computer programs / instructions; A processor, configured to execute the computer program / instructions to implement the steps of any one of the methods described in Example 1.

[0038] Embodiment 5: This embodiment provides a computer program product, including a computer program / instruction, which implements the steps of any method described in Embodiment 1 when executed by a processor.

[0039] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the technical principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

[0040] It should be understood by those skilled in the art that the embodiments of the present disclosure may be provided as methods, systems or computer program products. Therefore, the present disclosure may take the form of a complete hardware embodiment, a complete software embodiment or an embodiment combining software and hardware aspects. Moreover, the present disclosure may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0041] The present disclosure is described with reference to flowcharts and / or block diagrams of methods, devices (systems) and computer program products according to embodiments of the present disclosure. It should be understood that each process and / or block in the flowchart and / or block diagram and the combination of processes and / or blocks in the flowchart and / or block diagram can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0042] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0043] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0044] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present disclosure rather than to limit its protection scope. Although the present disclosure has been described in detail with reference to the above embodiments, ordinary technicians in the relevant field should understand that after reading the present disclosure, those skilled in the art can still make various changes, modifications or equivalent substitutions to the specific implementation methods of the invention, but these changes, modifications or equivalent substitutions are all within the protection scope of the disclosed claims to be approved.

Claims

1. A method for applying a bank customer development model based on knowledge graph, characterized in that: include: Obtain customer data to be screened, input the data into the pre-trained bank customer development model, identify high-value potential customers, and output the information of high-value potential customers for precision marketing; The training method of the bank customer development model includes: Obtain basic customer information from the bank's internal system and third-party data sources as a training set; Use customers as nodes and relationships between customers as edges to build a knowledge graph; Establishing a bank customer development model based on the knowledge graph; Use the training set to complete the training of the knowledge graph-based banking customer development model.

2. The bank customer development model application method based on knowledge graph according to claim 1 is characterized in that: The bank customer development model adopts PageRank algorithm or K-core subgraph algorithm.

3. The bank customer development model application method based on knowledge graph according to claim 2 is characterized in that: When the bank customer development model adopts the PageRank algorithm, the training process is as follows: Assign the same initial importance value to each node in the knowledge graph, set to 1 / N, where N is the total number of customers; Count the number of outgoing links of each node and construct a transfer matrix, where each row of the transfer matrix represents a node and each column represents a possible jump target; According to the transfer matrix and the importance value of the current node, the importance of each node is iteratively updated, where the new importance value of a node is the sum of the importance of all its incoming nodes, and the importance value of each incoming node is weighted according to its outgoing number. In the iterative process, a damping factor is introduced, and the iteration is continued until the importance of all nodes converges. Output the iterative results so that each node customer obtains a stable importance value to reflect the importance of the customer in the entire knowledge graph.

4. The bank customer development model application method based on knowledge graph according to claim 2 is characterized in that: When the bank customer development model adopts the K-core subgraph algorithm, the training process is as follows: Determine the minimum degree K required for each node in the specified subgraph in the K-core subgraph algorithm; Apply the K-core subgraph algorithm on the knowledge graph, traverse all nodes in the graph, and calculate the degree of each node; Remove nodes and related edges with degrees less than K until the degrees of all remaining nodes are not less than K; Extract the subgraph whose node degree is greater than or equal to K, which contains the most compact and core set of nodes in the graph; Analyze the properties of the extracted subgraph nodes to identify key nodes, important relationship chains, and potential high-value customer groups.

5. The bank customer development model application method based on knowledge graph according to claim 1 is characterized in that: The method also includes: preprocessing the training set, including data cleaning, deduplication, and standardization, and extracting key information.

6. The bank customer development model application method based on knowledge graph according to claim 1 is characterized in that: The method also includes: adding feedback data from precision marketing and newly applied customer data to sample data to update and iterate the bank's customer development model.

7. A bank customer development model application device based on knowledge graph, characterized in that: include: The application module is used to obtain customer data to be screened, input the customer data to be screened into the pre-trained bank customer development model, identify high-value potential customers, and output the information of high-value potential customers for precision marketing; The training method of the bank customer development model includes: Obtain basic customer information from the bank's internal system and third-party data sources as a training set; Use customers as nodes and relationships between customers as edges to build a knowledge graph; Establishing a bank customer development model based on the knowledge graph; Use the training set to complete the training of the knowledge graph-based banking customer development model.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the steps of the method described in any one of claims 1 to 6 are implemented.

9. A computer device, characterized in that: include: Memory, for storing computer programs / instructions; A processor, configured to execute the computer program / instructions to implement the steps of the method according to any one of claims 1 to 6.

10. A computer program product comprising a computer program / instructions, characterized in that When the computer program / instructions are executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images