Multi-granularity dynamic data pushing method and system

Abstract

The application provides a multi-granularity dynamic data pushing method and system, relates to the field of big data, and can be applied to the financial field and other fields.The method comprises the following steps: calling corresponding business data information according to identity information of a plurality of users in a preset area; constructing a business granularity mother tree through a full-amount business data set of the preset area, and generating a granularity subtree according to the business granularity mother tree and the business data information; extracting concepts of users by using formal concept analysis according to the granularity subtree and the identity information, and constructing a plurality of customer concept lattices with different scales; generating batch pushing data according to the customer concept lattices and service demands, and pushing data to the plurality of users according to the batch pushing data.

Description

Technical Field

[0001] This application relates to the field of big data and can be applied to the financial field and other fields, particularly a multi-granularity dynamic data push method and system. Background Technology

[0002] In recent years, the business model of bank branches has undergone significant changes, placing greater emphasis on customer service and marketing compared to traditional counter services. Currently, most bank branch marketing employs a one-on-one approach. For example, after a teller calls a customer's number, they select a product the customer doesn't currently own or recommend a product, sending the product's QR code to an external information processing device. The customer then scans the QR code to learn more when they come to the counter. Alternatively, lobby staff may approach customers during or after self-service transactions, aiming to attract or intercept them. However, without a comprehensive overview of the branch's customer base, current marketing models are somewhat random, and the one-on-one approach relies heavily on the experience and judgment of staff, resulting in significant manpower waste and low efficiency. Therefore, the industry urgently needs an efficient data analysis method to help staff quickly analyze potential user preferences and achieve precise, large-scale data delivery. Summary of the Invention

[0003] The purpose of this application is to provide a multi-granularity dynamic data push method and system. Taking on-site customers at bank branches as objects and marketing products as attributes, the system obtains and refines the relationships between customers and marketing products through a bank data warehouse. Formal concept analysis is used to extract formal concepts from objects, attributes, and the relationships between object attributes. Object clustering is performed by mining common attributes to generate concept lattices with duality. A batch of pre-selected attribute sets is then used to clean the concept lattices through evaluation methods, obtaining more effective conceptual information and constructing many-to-many object and attribute relationships, thereby achieving batch service and marketing effects.

[0004] To achieve the above objectives, this application provides a multi-granularity dynamic data push method, the method comprising: retrieving corresponding business data information based on the identity information of multiple users within a preset area; constructing a business granularity mother tree through the full business dataset of the preset area; generating granularity subtrees by matching the business granularity mother tree and the business data information; extracting user concepts using formal concept analysis based on the granularity subtrees and the identity information to construct multiple customer concept lattices of different sizes; generating batch push data based on the customer concept lattices and service requirements; and pushing data to multiple users based on the batch push data.

[0005] In the above multi-granularity dynamic data push method, optionally, retrieving corresponding business data information based on the identity information of multiple users within a preset area includes: obtaining the biometric information of users entering the preset area and / or the identification information provided by the users; analyzing the biometric information and / or the identification information to obtain the user's identity information; and retrieving the corresponding business data information from a preset database based on the identity information.

[0006] In the above multi-granularity dynamic data push method, optionally, constructing a business granularity mother tree through the full business dataset of a preset region includes: obtaining business information and business category of each business based on the full business dataset of the preset region; obtaining the affiliation relationship of each business based on the business category; and constructing a business granularity mother tree step by step based on the affiliation relationship and the business information.

[0007] In the above multi-granularity dynamic data push method, optionally, generating a granular subtree by matching the business granularity parent tree and the business data information includes: determining the corresponding leaf node in the business granularity parent tree according to the business data information; and generating a granular subtree according to the leaf node and the business granularity parent tree.

[0008] In the above-mentioned multi-granularity dynamic data push method, optionally, constructing multiple customer concept lattices of different scales by extracting user concepts using formal concept analysis based on the granularity subtree and the identity information includes: obtaining multi-level business categories based on the granularity subtree; and using formal concept analysis to extract user concepts based on the correspondence between the identity information and the business categories at each level, thereby obtaining customer concept lattices of different scales.

[0009] In the above multi-granularity dynamic data push method, optionally, generating batch push data based on the customer concept grid and service requirements includes: obtaining granularity information and push parameters based on the service requirements; retrieving the customer concept grid of the corresponding granularity based on the granularity information; obtaining corresponding customer information based on the corresponding customer concept grid; and generating batch push data based on the customer information and the push parameters.

[0010] In the above multi-granularity dynamic data push method, optionally, retrieving the customer concept grid of the corresponding granularity according to the granularity information includes: constructing an attribute set according to the granularity information, calculating the set similarity and separation degree of the attribute set and the customer concept grid; and obtaining the customer concept grid of the corresponding granularity according to the set similarity and separation degree.

[0011] This application also provides a multi-granularity dynamic data push system, the system comprising: a data acquisition module, a concept grid generation module, and a data push module; the data acquisition module is used to retrieve corresponding business data information based on the identity information of multiple users within a preset area; the concept grid generation module is used to construct a business granularity mother tree through the full business dataset of the preset area, and generate granularity subtrees by matching the business granularity mother tree and the business data information; based on the granularity subtrees and the identity information, it uses formal concept analysis to extract concepts from users and construct multiple customer concept grids of different sizes; the data push module is used to generate batch push data based on the customer concept grids and service requirements, and push data to multiple users based on the batch push data.

[0012] This application also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the above-described method.

[0013] This application also provides a computer-readable storage medium storing a computer program that performs the above-described methods.

[0014] This application also provides a computer program product, including a computer program / instructions that, when executed by a processor, implement the steps of the above-described method.

[0015] The beneficial technical effects of this application are as follows: from the moment a customer enters the branch, the customer is incorporated into the overall service marketing grid of the branch. By extracting the formal concepts of customers and attributes, the customer and attributes are double-clustered to construct a customer concept grid. The massive concept grid is cleaned through evaluation methods, enabling branch business personnel to efficiently obtain the association information between branch customers and products, select nodes to push service information or marketing information, and merge attributes and push them in batches to all customers in the object set, achieving a one-to-many service marketing effect. Using this invention can greatly improve the accuracy and efficiency of branch service marketing. Attached Figure Description

[0016] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, do not constitute a limitation thereof. In the drawings:

[0017] Figure 1 This is a flowchart illustrating a multi-granularity dynamic data push method provided in an embodiment of this application;

[0018] Figure 2 This is a schematic diagram of the process for obtaining business data information provided in an embodiment of this application;

[0019] Figure 3This is a schematic diagram illustrating the construction process of a business granularity mother tree provided in an embodiment of this application;

[0020] Figure 4 This is a schematic diagram illustrating the construction process of a granular subtree provided in an embodiment of this application;

[0021] Figure 5 A schematic diagram illustrating the construction process of a customer concept lattice provided in an embodiment of this application;

[0022] Figure 6 This is a schematic diagram of the batch push data generation process provided in an embodiment of this application;

[0023] Figure 7 A schematic diagram of the customer concept grid retrieval process provided in an embodiment of this application;

[0024] Figure 8 This is a schematic diagram of the application structure of a multi-granularity dynamic data push system provided in an embodiment of this application;

[0025] Figure 9 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0026] The following will describe in detail the implementation methods of this application with reference to the accompanying drawings and embodiments, so as to fully understand how this application uses technical means to solve technical problems and achieve technical effects, and to implement it accordingly. It should be noted that, as long as there is no conflict, the various embodiments and features in each embodiment of this application can be combined with each other, and the resulting technical solutions are all within the protection scope of this application.

[0027] Furthermore, the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although a logical order is shown in the flowchart, in some cases the steps shown or described may be executed in a different order than that shown here.

[0028] It is worth noting that the acquisition, storage, use, and processing of data in the technical solution of this application all comply with relevant laws and regulations. The user information in the embodiments of this application was obtained through legal and compliant means, and the acquisition, storage, use, and processing of user information have been authorized and agreed upon by the client.

[0029] Please refer to Figure 1 As shown, this application provides a multi-granularity dynamic data push method, the method comprising:

[0030] S101 retrieves corresponding business data information based on the identity information of multiple users within a preset area;

[0031] S102 constructs a service granularity mother tree through the full service dataset of a preset region, and generates a granularity subtree by matching the service granularity mother tree and the service data information;

[0032] S103 Based on the granularity subtree and the identity information, formal concept analysis is used to extract concepts from the user and construct multiple customer concept lattices of different sizes;

[0033] S104 generates batch push data based on the customer concept grid and service requirements, and pushes data to multiple users based on the batch push data.

[0034] Therefore, taking on-site customers at branch offices as the objects and key customer business data as attributes, key customer business data is obtained from the bank's data warehouse and refined. Formal concept analysis is used to extract formal concepts from objects, attributes, and the relationships between object attributes. Object clustering is performed by mining common attributes to generate customer concept lattices. The full business dataset from the data warehouse is extracted to construct a financial business granularity parent tree. Current branch customer attributes are matched with the parent tree to extract granular subtrees. Different scales of customer concept lattices are obtained by transforming the granularity of the subtrees. In practice, these concept lattices provide a comprehensive and visual clustered representation of branch customer information. The concept lattices are rich in information, not only showing the clustering of objects but also analyzing the co-occurrence probability of products from an attribute perspective. Business personnel can select different granularities and categories of customer attributes and redraw concept lattices to achieve better clustering results. By batch pushing services and marketing to clustered customer groups, more accurate and efficient classified services and marketing effects can be obtained.

[0035] Formal concept analysis, as described above, refers to using formal context to describe objects, attributes, and the relationships between them. It involves mining concepts from the formal context and representing the generalization and instantiation relationships of these concepts using concept lattices. Here, the extension of a concept can be considered the set of all objects, while the intension is the set of attributes common to objects within that extension. The basic idea of ​​formal concept analysis is to describe a domain using the formal context of objects and attributes. In formal concept analysis, formal context is used to represent data; a specific definition will be given below.

[0036] Definition 1: A formal context is a triple K = (G, M, I), where G represents the set of object elements, and M represents the set of all attributes contained in the objects in G. It is a set of relationships between objects and attributes. (g,m)∈I means that object g has attribute m, which can also be represented as gIm.

[0037] According to definition 1, the formal background can be represented as a cross table as shown in Table 1.

[0038] Table 1

[0039]

[0040] In this table, the columns and rows represent the objects and attributes of the background, respectively. An "×" is marked at the intersection of a row and a column, indicating that the object in that row possesses the attribute corresponding to that column; otherwise, it indicates that the object does not possess the attribute corresponding to that column. As shown in Table 1, object a has attributes 2, 4, and 5, and object b has attribute 5.

[0041] Definition 2: Suppose there exists a formal background K = (G, M, I), and a set The common attribute of all objects in set A is then defined as:

[0042]

[0043] Correspondingly, for sets Define the set of objects containing all attributes of set B as:

[0044]

[0045] As shown in Table 1, if the set of objects A is A = {a, c, f}, then the common attribute A' of the objects in the set is A' = {2, 5}. If the set of attributes B is B = {1, 2, 3, 4}, then B' = {d}.

[0046] Definition 3: Let the formal background be K = (G, M, I), if If A' = B and B' = A, then (A, B) is called a concept in the formal background K, and A and B are called the extension and intension of the concept, respectively. β(G, M, I) is the set of all concepts in background K.

[0047] As shown in Table 1, let A = {c,d,e,f} and B = {1,2}. According to the table, A' = B and B' = A. Then ({c,d,e,f},{1,2}) is called a concept of K, where A is the extension of the concept and B is the intension of the concept.

[0048] Proposition 1: If K = (G, M, I) is a formal background, For a collection of objects, If it is a set of attributes, then: (5)A'=A"'; (6)B'=B"';

[0049] Corollary 1: A" is the smallest extension containing A, and B" is the smallest intension containing B. Corollary 2: If there exists and only one A = A", then... It is the extension of the formal background; similarly, if there exists and only one "B = B", then It is the connotation of the formal background.

[0050] Definition 4: Let (A1, B1) and (A2, B2) be two concepts of the formal background K, respectively. Then (A1,B1) is called a subconcept of (A2,B2), and (A2,B2) is called a superconcept of (A1,B1), denoted as (A1,B1)≤(A2,B2), where the relation ≤ is called the order. The set of all order relations on the formal background K is called the concept lattice of that background, which can be represented as β(G,M,I). The concept lattice is a complete lattice, and concept elements and order relations can be displayed through Hasse diagrams.

[0051] Definition 5: Let C1 and C2 be two concepts with a known formal background K. If C1 ≤ C2 and there is no other concept C3 (C3 ≠ C1, C3 ≠ C2) that satisfies C1 ≤ C3 ≤ C2, then C1 is called the lower neighbor of C2 and C2 is called the upper neighbor of C1. The relationship between C1 and C2 is C1 < C2.

[0052] The duality principle of concept lattices: Suppose there exists a formal background K = (G, M, I). By swapping the rows and columns of background K, we get K' = (M, G, I). -1 ), then K' is also a formal background, making yes β (G,M,I) to β (G,M,I) d isomorphic mappings.

[0053] The duality principle of concept lattices explains that if the objects and attributes in the formal context are swapped, a concept lattice can also be generated, and its hierarchical order is reversed compared to the original concept lattice. Since the concept lattice itself is a clustering process, it can also be seen through the duality principle that it can cluster not only objects but also attributes. Therefore, concept lattice clustering is a bi-clustering method.

[0054] For applications of formal concept analysis, the core content is constructing a formal context and generating concept lattices. Constructing the concept lattice is a crucial step; regardless of the order of elements within the same formal context, the generated concept lattice will be consistent.

[0055] Please refer to Figure 2 As shown, in one embodiment of this application, retrieving corresponding business data information based on the identity information of multiple users within a preset area includes:

[0056] S201 Obtains the biometric information of users entering the preset area and / or the identification information provided by the users;

[0057] S202 Analyzes the biometric information and / or the document information to obtain the user's identity information;

[0058] S203 retrieves the corresponding business data information from the preset database based on the identity information.

[0059] Specifically, in practice, this business data information can be used to verify user identity through methods such as facial recognition, bank card registration, and ID card registration, and then the user's business data can be retrieved based on the user's identity. Among them, bank card or ID card information can be obtained through the branch queuing machine system after the customer obtains a number online or offline; facial information can be obtained by capturing and recognizing the faces of users entering the branch area based on the existing facial recognition system.

[0060] Please refer to Figure 3 As shown, in one embodiment of this application, constructing a business-granularity parent tree using the full business dataset of a preset region includes:

[0061] S301 obtains the business information and business category of each business based on the full business dataset of the preset area;

[0062] S302 obtains the attribution relationship of each business according to the business category, and constructs a business granularity mother tree step by step according to the attribution relationship and the business information.

[0063] For further details, please refer to... Figure 4 As shown, in one embodiment of this application, generating a granular subtree based on the matching of the business granularity parent tree and the business data information includes:

[0064] S401 determines the corresponding leaf node in the service granularity parent tree based on the service data information;

[0065] S402 generates a granular subtree based on the leaf nodes and the business granularity parent tree.

[0066] Specifically, in practice, the granularity tree is primarily determined by the hierarchical relationships within the business structure. For example, insurance products can be subdivided into auto insurance, health insurance, and accident insurance, while wealth management products can be subdivided into product categories with different risk levels and holding periods. The closer the granularity tree is to the leaf nodes, the finer the attribute granularity. For instance, in the granularity tree of financial products, the granularity tree needs to be periodically reconstructed because financial business products are constantly being updated. Granular subtrees can be constructed on the granular parent tree based on user information. For example, if a branch customer holds products such as Jiejiegao No. 1 and Jiejiegao No. 2, which belong to car insurance type A, then the extracted subtree would be car insurance type A branching to Jiejiegao No. 1 and Jiejiegao No. 2. The granularity tree setting allows for more flexible adjustment of the size of the customer concept lattice. The closer the granularity tree is to the leaf node, the finer the attribute granularity. Finer granular attributes will contain more detailed information in the formal context, resulting in a larger concept lattice size and more refined clustering results. Conversely, coarser granular attributes will reduce the exposure of information in the formal context, resulting in a smaller concept lattice size and fewer clustering results.

[0067] Please refer to Figure 5 As shown, in one embodiment of this application, the construction of multiple customer concept lattices of different sizes by using formal concept analysis to extract user concepts based on the granular subtree and the identity information includes:

[0068] S501 obtains multi-level service categories based on the granularity subtree;

[0069] S502 uses formal concept analysis to extract concepts from users based on the correspondence between the identity information and business categories at all levels, and obtains customer concept grids corresponding to different scales.

[0070] In practice, due to different business needs, the business level to be extracted can be determined based on business differences during the concept extraction process, and customer concept lattices of different granularities and scales can be constructed based on the business level. The construction method of the customer concept lattice has been described in detail in the foregoing embodiments and will not be described in detail here.

[0071] Please refer to Figure 6 As shown, in one embodiment of this application, generating batch push data based on the customer concept grid and service requirements includes:

[0072] S601 obtains granularity information and push parameters according to the service requirements, and retrieves the customer concept grid of the corresponding granularity according to the granularity information;

[0073] S602 obtains the corresponding customer information based on the corresponding customer concept grid, and generates batch push data based on the customer information and the push parameters.

[0074] Specifically, in practice, there are currently two main types of methods for constructing concept lattices. One is batch processing algorithms, which can generate all nodes at once. These algorithms can be further divided into top-down and bottom-up construction algorithms. Top-down algorithms, such as the Bordat algorithm, start building from the top of the concept lattice, first finding the largest concept among all concepts, and then building the concept lattice downwards. Bottom-up construction algorithms, on the other hand, do the opposite, starting from the smallest concept and building the concept lattice upwards, such as the Chein algorithm. The other type of algorithm is incremental construction algorithms, which differ in that they construct the concept lattice by continuously adding nodes.

[0075] The batch processing algorithm is described as follows:

[0076] First, initialize a concept lattice L = (H, I, R). Let the queue P = (H, I, R). Then, generate sub-concepts A for all concepts A in the queue P in sequence. i If A i If it is a new concept, then new concept A i Add to cell L and add new concept A. i The concept is related to the order of its parent concept. If queue P is not empty, continue adding concepts to queue P until P is empty, and finally output the concept cell L.

[0077] The incremental algorithm first initializes an empty concept grid, inserts an object into the concept grid one at a time, and determines the operation on the concept grid by the result of the intersection operation between the object and the existing concepts in the concept grid. The operations include adding, modifying and deleting nodes.

[0078] The incremental algorithm is described as follows:

[0079] First, initialize an empty concept lattice L, and let the concepts in the existing concept lattice L be Ci = (Ai, Bi). Then, extract the objects g from the formal background in sequence. Then add object g to the object collection A of this concept. i In the middle, if Bi∩f(g)≠Bi, and there does not exist a parent node (Aj,Bj) of (Ai,Bi) that satisfies Then, a new node is created in the existing concept grid L. If there are still unprocessed objects g in the formal context, the process is repeated until the end, and finally the concept grid L is output.

[0080] Please refer to Figure 7 As shown, in one embodiment of this application, retrieving the customer concept grid with the corresponding granularity based on the granularity information includes:

[0081] S701 constructs an attribute set based on the granularity information, and calculates the set similarity and separation degree between the attribute set and the customer concept lattice;

[0082] S702 obtains the customer concept lattice with corresponding granularity based on the set similarity and separation degree.

[0083] Specifically, in practice, due to the diverse product types and numerous customer attributes extracted by banks, the resulting concept grids can be enormous, with excessive concept nodes and related information, making rapid judgment by business personnel challenging. Therefore, business personnel can set expected attribute sets based on current best-selling items and personal recommendations. After generating the customer concept grid, they can evaluate each concept node using an evaluation method, retaining those that meet the evaluation criteria and removing those that significantly deviate from expectations. This process cleanses the concept grid, resulting in a more effective customer concept grid.

[0084] Evaluation metric 1: Set similarity. Calculate the similarity between the preset set and the attribute set in each concept node. The higher the similarity, the higher the attribute overlap and the more consistent the concept node.

[0085]

[0086] Where S exp Let S be a predefined set. n This refers to the set of attributes within a concept node.

[0087] Evaluation index 2: Set separation degree. Calculate the separation degree between the preset set and the attribute set in each concept node. The smaller the separation degree, the smaller the difference between the two attribute sets, and the more consistent the concept node is.

[0088]

[0089] Where S exp Let S be a predefined set. n This refers to the set of attributes within a concept node.

[0090] Set validity: The validity of the attribute set in the concept node is obtained by combining evaluation index one and evaluation index two.

[0091]

[0092] When the set in the node is exactly the same as the preset set attribute, Eva(S) exp ,S n Eva(S) = 1, when the set in the node is completely inconsistent with the preset set attributes. exp ,S n = -1.

[0093] By retaining concept nodes with similarity greater than dissimilarity, i.e., Eva(S) exp ,S nA value greater than 0 can filter out a lot of irrelevant conceptual information, greatly optimize the concept grid, and make it easier for business personnel to effectively identify target customer sets and related service information.

[0094] In one embodiment of this application, a multi-granularity dynamic data push system is also provided. The system includes: a data acquisition module, a concept lattice generation module, and a data push module. The data acquisition module is used to retrieve corresponding business data information based on the identity information of multiple users within a preset area. The concept lattice generation module is used to construct a business granularity mother tree through the full business dataset of the preset area, and generate granularity subtrees by matching the business granularity mother tree and the business data information. Based on the granularity subtrees and the identity information, the module uses formal concept analysis to extract concepts from users and constructs multiple customer concept lattices of different sizes. The data push module is used to generate batch push data based on the customer concept lattices and service requirements, and push data to multiple users based on the batch push data.

[0095] In practical work, the aforementioned multi-granularity dynamic data push system can be referenced. Figure 8 As shown, its structure may include: a customer identification system 1, terminal peripherals 2, a business processing system 3, a branch service marketing system 4, a data preprocessing unit 41, a concept grid construction unit 42, an evaluation unit 43, a data storage unit 44, a front-end interaction unit 45, and a business processing unit 46. Among these, the data preprocessing unit 41, concept grid construction unit 42, evaluation unit 43, data storage unit 44, front-end interaction unit 45, and business processing unit 46 are sub-units of the branch service marketing system 4.

[0096] Customer Identification System 1: Existing customer identification systems at the branch can be used, such as the branch queuing machine system, which can identify customers after they take a number online or offline; or the smart branch facial recognition system, which captures and identifies the faces of customers entering the branch.

[0097] Terminal peripheral 2: Supports smart terminal devices such as mobile phones, platforms, and PCs for front-end page display and human-computer interaction.

[0098] Business Processing System 3: Supports business personnel to perform operations such as object and attribute pre-selection and batch marketing initiation. Branch Service Marketing System 4 calls the existing business processing system to perform business processing.

[0099] The branch service marketing system 4 includes a data preprocessing unit, a concept grid construction unit, an evaluation unit, a data storage unit, a front-end interaction unit, and a business processing unit. It connects to the customer identification system and the business processing system. After the customer identification system 1 identifies a customer, it provides the customer information to the service marketing system. The service marketing system extracts attributes from the data warehouse and performs data cleaning, clustering, and other processing to generate customer concept grids. It uses evaluation coefficients to calculate the concept nodes that are closest to the pre-selected ones and sends them in batches to the business processing system 3 for service, marketing, and other business processing.

[0100] Data preprocessing unit 41: Receives customer information data from customer identification system 1 and interactive data returned by terminal peripherals 2, and performs data organization, storage, and related preprocessing operations. The customer identification system provides customer elements, extracting full customer attributes from the bank's data warehouse. Customer attributes include those of customers who do not hold products, those who are recommended products, and those whose products are nearing maturity. For marketing, there are often recommendation principles and preferences. For example, products matching the risk level of customers with different risk assessment levels should be recommended. For complex financial products, there is a preference for recommending them to younger customers. Therefore, pre-setting data cleaning rules to clean and purify the data can improve the effectiveness of marketing.

[0101] Concept lattice construction unit 42: This system takes customers as its objects. Customer attributes include pre-processed customer elements, non-held products, recommended products, and near-expiration products. Customers and their attributes are described through formal context. Through formal concept analysis, concepts are mined from the formal context, and the generalization and instantiation relationships of these concepts are constructed into concept lattices. Constructing concept lattices is a bi-clustering process; it can cluster not only objects but also attributes, resulting in concept lattices with duality. Business personnel can extract rich information from the customer concept lattice and initiate batch services and marketing recommendations. This concept lattice construction unit 42 may also include a granular tree construction unit, which extracts the full business dataset from the data warehouse, cleans and deduplicates the data, and constructs a financial business granularity mother tree based on the hierarchical relationships of the business. For example, insurance products can be subdivided into car insurance, health insurance, and accident insurance, while wealth management products can be subdivided into product categories with different risk levels and holding periods. The closer the granularity tree is to the leaf node, the finer the attribute granularity. Then, the network customer attributes purified by the data preprocessing unit 41 are matched on the granularity mother tree to extract the granularity subtree. For example, if the products held by the network customer are: Jie Jie Gao No. 1, Jie Jie Gao No. 2, and Car Insurance A, then the subtree extracted from the granularity mother tree is obtained.

[0102] Evaluation Unit 43: Due to the diverse product types and numerous customer attributes extracted by banks, the resulting concept grids are extremely large, with excessive concept nodes and related information, posing a challenge for business personnel in making rapid judgments. Therefore, business personnel can set expected attribute sets based on current best-selling items and personal recommendations. After generating the customer concept grid, they can evaluate each concept node using evaluation methods, retaining those that meet the evaluation criteria and removing those that significantly deviate from expectations. This process cleanses the concept grid, resulting in a more effective customer concept grid.

[0103] Data storage unit 44: Uses a MySQL database for data storage, provides database interaction interfaces, database connection pool management, etc.

[0104] Front-end Interaction Unit 45: Front-end page display, human-computer interaction processing, etc.

[0105] Business processing unit 46: processes batch business requests, classifies business requests, organizes call information, and interacts with business processing system 3.

[0106] The beneficial technical effects of this application are as follows: from the moment a customer enters the branch, the customer is incorporated into the overall service marketing grid of the branch. By extracting the formal concepts of customers and attributes, the customer and attributes are double-clustered to construct a customer concept grid. The massive concept grid is cleaned through evaluation methods, enabling branch business personnel to efficiently obtain the association information between branch customers and products, select nodes to push service information or marketing information, and merge attributes and push them in batches to all customers in the object set, achieving a one-to-many service marketing effect. Using this invention can greatly improve the accuracy and efficiency of branch service marketing.

[0107] This application also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the above-described method.

[0108] This application also provides a computer-readable storage medium storing a computer program that performs the above-described methods.

[0109] This application also provides a computer program product, including a computer program / instructions that, when executed by a processor, implement the steps of the above-described method.

[0110] like Figure 9 As shown, the electronic device 600 may also include: a communication module 110, an input unit 120, an audio processing unit 130, a display 160, and a power supply 170. It is worth noting that the electronic device 600 does not necessarily need to include these components. Figure 9All components shown; in addition, the electronic device 600 may also include Figure 9 For components not shown, please refer to existing technologies.

[0111] like Figure 9 As shown, the central processing unit 100, sometimes also referred to as a controller or operating control, may include a microprocessor or other processor device and / or logic device. The central processing unit 100 receives inputs and controls the operation of various components of the electronic device 600.

[0112] The memory 140 may be, for example, one or more of a cache, flash memory, hard drive, removable media, volatile memory, non-volatile memory, or other suitable devices. It may store the aforementioned failure-related information, and also store a program for executing that information. The central processing unit 100 may execute the program stored in the memory 140 to perform information storage or processing, etc.

[0113] Input unit 120 provides input to central processing unit 100. Input unit 120 may be, for example, a keypad or touch input device. Power supply 170 provides power to electronic device 600. Display 160 displays images and text. Display may be, for example, an LCD display, but is not limited thereto.

[0114] The memory 140 can be a solid-state memory, such as a read-only memory (ROM), random access memory (RAM), a SIM card, etc. It can also be a memory that retains information even when power is off, can be selectively erased, and contains more data; examples of this type of memory are sometimes referred to as EPROMs. The memory 140 can also be some other type of device. The memory 140 includes a buffer memory 141 (sometimes referred to as a buffer). The memory 140 may include an application / function storage unit 142 for storing application programs and function programs or processes for executing the operation of the electronic device 600 via the central processing unit 100.

[0115] The memory 140 may also include a data storage unit 143 for storing data, such as contacts, digital data, pictures, sounds, and / or any other data used by the electronic device. The driver storage unit 144 of the memory 140 may include various drivers for the electronic device's communication functions and / or for performing other functions of the electronic device (such as messaging applications, address book applications, etc.).

[0116] The communication module 110 is a transmitter / receiver 110 that transmits and receives signals via antenna 111. The communication module (transmitter / receiver) 110 is coupled to the central processing unit 100 to provide input signals and receive output signals, which can be the same as in a conventional mobile communication terminal.

[0117] Based on different communication technologies, multiple communication modules 110 can be configured in the same electronic device, such as cellular network modules, Bluetooth modules, and / or wireless LAN modules. The communication module (transmitter / receiver) 110 is also coupled to a speaker 131 and a microphone 132 via an audio processor 130 to provide audio output via the speaker 131 and receive audio input from the microphone 132, thereby enabling typical telecommunications functions. The audio processor 130 may include any suitable buffer, decoder, amplifier, etc. Additionally, the audio processor 130 is coupled to a central processing unit 100, enabling on-device recording via the microphone 132 and on-device playback of stored audio via the speaker 131.

[0118] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied 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.

[0119] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0120] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1The function specified in one or more boxes.

[0121] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0122] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of this application. It should be understood that the above descriptions are merely specific embodiments of this application and are not intended to limit the scope of protection of this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A multi-granularity dynamic data push method, characterized in that, The method includes: Retrieve corresponding business data information based on the identity information of multiple users within a preset area; A business granularity mother tree is constructed using the full business dataset of a preset region, and a granularity subtree is generated by matching the business granularity mother tree with the business data information. Based on the granular subtree and the identity information, formal concept analysis is used to extract user concepts and construct multiple customer concept lattices of different sizes. Generate batch push data based on the customer concept grid and service requirements, and push data to multiple users based on the batch push data; Among them, based on the granular subtree and the identity information, formal concept analysis is used to extract concepts from users and construct multiple customer concept lattices of different sizes, including: Multi-level business categories are obtained based on the granularity subtree; Based on the correspondence between the identity information and business categories at all levels, formal concept analysis is used to extract concepts from users and obtain customer concept lattices corresponding to different scales. The batch push data generated based on the customer concept grid and service requirements includes: Based on the service requirements, obtain granularity information and push parameters, and retrieve the corresponding granularity of the customer concept grid based on the granularity information; Obtain corresponding customer information based on the corresponding customer concept grid, and generate batch push data based on the customer information and the push parameters.

2. The multi-granularity dynamic data push method according to claim 1, characterized in that, Based on the identity information of multiple users within a preset area, the corresponding business data information is retrieved, including: Obtain biometric information and / or identification information provided by users entering the preset area; The user's identity information is obtained by analyzing the biometric information and / or the document information. The corresponding business data information is retrieved from the preset database based on the identity information.

3. The multi-granularity dynamic data push method according to claim 1, characterized in that, The business-granularity parent tree is constructed using the full business dataset of the preset region, including: Obtain the business information and business category of each business based on the full business dataset of the preset region; The attribution relationship of each business is obtained according to the business category, and a business granularity mother tree is constructed step by step according to the attribution relationship and the business information.

4. The multi-granularity dynamic data push method according to claim 3, characterized in that, Generating a granular subtree by matching the business granularity parent tree and the business data information includes: Based on the business data information, determine the corresponding leaf node in the business granularity parent tree; A granular subtree is generated based on the leaf nodes and the business granularity parent tree.

5. The multi-granularity dynamic data push method according to claim 1, characterized in that, The customer concept grid retrieved based on the granularity information includes: An attribute set is constructed based on the granularity information, and the set similarity and separation degree between the attribute set and the customer concept lattice are calculated. The customer concept lattice with the corresponding granularity is obtained based on the set similarity and separation degree.

6. A multi-granularity dynamic data push system, characterized in that, The system includes: a data acquisition module, a concept grid generation module, and a data push module; The data acquisition module is used to retrieve corresponding business data information based on the identity information of multiple users within a preset area; The concept grid generation module is used to construct a business granularity mother tree through the full business dataset of a preset region, and generate granularity subtrees by matching the business granularity mother tree and the business data information; and to extract customer concepts from users using formal concept analysis based on the granularity subtrees and the identity information to construct multiple customer concept grids of different sizes. The data push module is used to generate batch push data according to the customer concept grid and service requirements, and push data to multiple users according to the batch push data; Among them, based on the granular subtree and the identity information, formal concept analysis is used to extract concepts from users and construct multiple customer concept lattices of different sizes, including: Multi-level business categories are obtained based on the granularity subtree; Based on the correspondence between the identity information and business categories at all levels, formal concept analysis is used to extract concepts from users and obtain customer concept lattices corresponding to different scales. The batch push data generated based on the customer concept grid and service requirements includes: Based on the service requirements, obtain granularity information and push parameters, and retrieve the corresponding granularity of the customer concept grid based on the granularity information; Obtain corresponding customer information based on the corresponding customer concept grid, and generate batch push data based on the customer information and the push parameters.

7. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the method of any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that enables a computer to execute the method of any one of claims 1 to 5.

9. A computer program product comprising a computer program / instructions, characterized in that, When the computer program / instructions are executed by the processor, they implement the steps of the method described in any one of claims 1 to 5.

Images