Data processing method and device, nonvolatile storage medium and electronic equipment

Abstract

The invention discloses a data processing method and device, a nonvolatile storage medium and electronic equipment. The method comprises the following steps: determining a plurality of user sets in a plurality of different users; connecting the plurality of user identifiers with the customer identifier through a bridge line, and connecting the customer identifier with the customer name; the customer identifier is connected with the contract identifier and the identity certificate identifier which are associated with the customer identifier through an anchor line, the contract identifier is connected with the customer name, the identity certificate identifier is connected with the customer name, and the weight of the anchor line is smaller than that of the bridge line; and connecting the customer identifier with the business object associated with the customer identifier through a frame line, and connecting the business object with the customer name, the weight of the frame line being smaller than the weight of the anchor line. According to the method and the device, the technical problem that a user group cannot be positioned in a message pushing scene and other scenes due to the fact that user data with a complex incidence relation cannot be effectively integrated in an operator database is solved.

Description

Technical Field

[0001] This application relates to the field of information technology applications, and more specifically, to a data processing method and apparatus, a non-volatile storage medium, and an electronic device. Background Technology

[0002] In the daily operations of telecommunications operators, databases accumulate massive amounts of user data. This data includes not only basic personal information such as name, address, and contact information, but also users' consumption habits, service preferences, device usage, and potential relationships with other users. However, due to historical issues, non-standard data entry, and diverse user scenarios, this data often exhibits complex relational structures, making the effective integration and parsing of user relationships within the operator's database a significant challenge.

[0003] Specifically, when a customer has multiple accounts or multiple products (such as SIM cards, broadband, IPTV, etc.) in an operator's system, these accounts and products may be scattered across different data tables or systems. Furthermore, due to limitations in early system design, accounts are not always linked by a unified standard key (such as customer ID), but may be indirectly associated through contract numbers, identification numbers, package codes, and other methods. In this situation, traditional direct matching or single-keyword search methods struggle to accurately identify all accounts belonging to the same customer or potential customer group. Consequently, in scenarios such as push notifications, personalized service recommendations, and market analysis, operators are unable to accurately locate and aggregate target user groups.

[0004] There is currently no effective solution to the above problems. Summary of the Invention

[0005] This application provides a data processing method and apparatus, a non-volatile storage medium, and an electronic device to at least solve the technical problem of being unable to locate user groups in scenarios such as message push due to the inability to effectively integrate user data with complex relationships in the operator's database.

[0006] According to one aspect of this application, a data processing method is provided, comprising: determining multiple user sets among multiple different users, wherein each user set includes: multiple user identifiers, each user identifier being associated with a consistent customer identifier; for each user set, connecting the multiple user identifiers and customer identifiers via bridge lines, and connecting customer identifiers with customer names; for each user set, connecting a customer identifier with a contract identifier and an identity credential identifier associated with the customer identifier via anchor lines, connecting the contract identifier with the customer name, and connecting the identity credential identifier with the customer name, wherein the weight of the anchor lines is less than the weight of the bridge lines; for each user set, connecting a customer identifier with a business object associated with the customer identifier via frame lines, and connecting the business object with the customer name, wherein the weight of the frame lines is less than the weight of the anchor lines.

[0007] Optionally, for each user set, before connecting the customer identifier and the business object associated with the customer identifier through the frame line, the business object associated with the customer identifier is determined by the following method, wherein the business object includes packages and products: Step S1, determine the package set associated with the customer identifier, and traverse the package set through the first cursor object; Step S2, determine whether the traversal of the package set is complete; if yes, determine the packages and products associated with the customer identifier; if no, proceed to step S3; Step S3, obtain the target package identifier currently pointed to by the first cursor object, and retrieve the package basic information corresponding to the target package identifier using the target package identifier as an index; Step S4, determine the associated packages of the target package identifier based on the package basic information. The product set is traversed using a second cursor object; Step S5: Determine if the traversal of the product set is complete; if yes, retrieve the next package in the package set and execute step S2; if no, execute step S6; Step S6: Obtain the target identifier of the current product using the second cursor object, query the product database using the target identifier as the key to obtain the product's basic information; Step S7: Based on the product's basic information, determine if there are any sub-products associated with the current product; if yes, recursively retrieve the sub-products, store the sub-products in the product relationship chain, and execute step S5; if no, execute step S8; Step S8: Store the current product in the product relationship chain, retrieve the next product in the product set, and execute step S5.

[0008] Optionally, the method further includes: for each user set, connecting the business object with the user installation address information, terminal registration address information, terminal usage address information, and call base station address associated with the business object through meridian lines, wherein the weight of the meridian lines is less than the weight of the frame lines; for each user set, determining the similarity between the address information of any two users to obtain one or more similarities; if the average of one or more similarities is greater than a first preset threshold, determining that the user set is labeled as users of the same family; if the average of one or more similarities is greater than a second preset threshold and less than the first preset threshold, determining that the user set is labeled as users of the same customer group.

[0009] Optionally, the method further includes: for each user set, connecting the call base station address and the call frequency information and call time information associated with the call base station address through auxiliary lines, wherein the weight of the auxiliary lines is less than the weight of the meridian lines.

[0010] Optionally, the method further includes: for each user set, determining whether the relationship between users belongs to a preset label, and determining the call frequency and call duration between users; determining the correlation index between users based on whether the relationship between users belongs to a preset label, the call frequency and call duration between users; if the correlation index is greater than a third preset threshold, determining that the user set is labeled as users of the same family; if the correlation index is greater than a fourth preset threshold and less than a third preset threshold, determining that the user set is labeled as users of the same customer group.

[0011] Optionally, determining the call frequency and call duration between users includes: for any two users in the user set, determining the call frequency as the ratio of the number of calls within a preset time window to the length of the preset time window, and determining the call duration as the ratio of the number of calls within a preset peak period to the total number of calls within the preset time window.

[0012] Optionally, determining whether the relationship between users belongs to a preset label includes: processing user personal information using a random forest classification model to determine whether the relationship between users belongs to a preset label, wherein the preset label includes: "same family". The random forest classification model is trained by: obtaining a training dataset, wherein the training dataset includes a set of user personal information, including: age field, gender field, occupation field, and surname field; calculating the global minimum and global maximum values ​​of the age field based on the training dataset, and linearly transforming the original age to a preset interval using the global minimum and global maximum values ​​to obtain the normalized age feature; obtaining the gender string. The system employs a mapping rule between gender and numerical values. Based on this rule, the gender field is converted into discrete numerical values, yielding numerical gender features. The occupation field is extracted from the training dataset, generating a binary vector representation of the occupation field to obtain one-hot encoded occupation features. A mapping dictionary from surname to family identifier is obtained, and the surname field is converted into a family identifier by querying the dictionary, yielding family identifier features. For any two users in the training dataset, if their family identifier features are identical, they are labeled positively; otherwise, they are labeled negatively. The normalized age feature, numerical gender feature, one-hot encoded occupation feature, and family identifier feature are concatenated into an input vector, and the positive and negative labels are used as supervisory signals to train a random forest classification model.

[0013] According to another aspect of this application, a data processing apparatus is also provided, comprising: a determining module, configured to determine multiple user sets among multiple different users, wherein each user set includes: multiple user identifiers, and each user identifier is associated with a consistent customer identifier; a first connecting module, configured to connect multiple user identifiers and customer identifiers, and connect customer identifiers and customer names, for each user set, via bridge lines; a second connecting module, configured to connect customer identifiers and associated contract identifiers and identity credential identifiers, connect contract identifiers and customer names, and connect identity credential identifiers and customer names, for each user set, via anchor lines, wherein the weight of the anchor lines is less than the weight of the bridge lines; and a third connecting module, configured to connect customer identifiers and associated business objects, and connect business objects and customer names, for each user set, via frame lines, wherein the weight of the frame lines is less than the weight of the anchor lines.

[0014] According to another aspect of this application, a non-volatile storage medium is also provided, the storage medium including a stored program, wherein the program, when running, controls the device where the storage medium is located to execute the above data processing method.

[0015] According to another aspect of this application, an electronic device is also provided, comprising: a memory and a processor, the processor being configured to run a program stored in the memory, wherein the program executes the above-described data processing method during runtime.

[0016] According to another aspect of this application, a computer program is also provided, wherein the computer program, when executed by a processor, implements the above-described data processing method.

[0017] According to another aspect of this application, a computer program product is also provided, comprising a non-volatile computer-readable storage medium, wherein the non-volatile computer-readable storage medium stores a computer program, which, when executed by a processor, implements the above-described data processing method.

[0018] In this application, multiple user sets are identified from multiple different users. Each user set includes multiple user identifiers, and each user identifier is associated with a consistent customer identifier. For each user set, multiple user identifiers and customer identifiers are connected by bridge lines, and customer identifiers are connected by customer names. For each user set, customer identifiers are connected by anchor lines to contract identifiers and identity credential identifiers associated with the customer identifiers, contract identifiers are connected by customer names, and identity credential identifiers are connected by customer names, where the weight of anchor lines is less than the weight of bridge lines. For each user set, customer identifiers are connected by frame lines to business objects associated with the customer identifiers, and business objects are connected by customer names, where the weight of frame lines is less than the weight of anchor lines. This approach effectively integrates user data with complex relationships, thereby achieving the technical effect of accurately locating target user groups. It also solves the technical problem of being unable to locate user groups in scenarios such as message push due to the inability to effectively integrate user data with complex relationships in the operator's database. Attached Figure Description

[0019] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings:

[0020] Figure 1 This is a flowchart of a data processing method according to an embodiment of this application;

[0021] Figure 2a This is a schematic diagram of the connection between objects according to an embodiment of this application;

[0022] Figure 2b This is a schematic diagram of another connection between objects according to an embodiment of this application;

[0023] Figure 2c This is a schematic diagram of another connection between objects according to an embodiment of this application;

[0024] Figure 2dThis is a schematic diagram of another connection between objects according to an embodiment of this application;

[0025] Figure 3 This is a flowchart illustrating a method for determining a business object associated with a customer identifier according to an embodiment of this application;

[0026] Figure 4 This is a flowchart of a method for determining the tags corresponding to a user set according to an embodiment of this application;

[0027] Figure 5 This is a structural diagram of a data processing apparatus according to an embodiment of this application;

[0028] Figure 6 This is a hardware structure block diagram of a computer terminal according to an embodiment of the present application for a data processing method. Detailed Implementation

[0029] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort should fall within the scope of protection of the present application.

[0030] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0031] According to an embodiment of this application, a method embodiment for data processing is provided. It should be noted that 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. Furthermore, 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.

[0032] Figure 1 This is a flowchart of a data processing method according to an embodiment of this application, such as... Figure 1As shown, the method includes the following steps:

[0033] Step S102: Determine multiple user sets among multiple different users, wherein each user set includes: multiple user identifiers, and the customer identifier associated with each user identifier is the same.

[0034] Step S102 is the first step in building a user association network. Its purpose is to identify and aggregate multiple user identifiers that actually belong to the same customer from numerous isolated user records in the database. For example, a customer is a family, and users are family members of that family.

[0035] In a typical telecommunications operator's database, each customer may have multiple unique user identifiers due to the different types of telecommunications services they have subscribed to (such as mobile phones, broadband, IPTV, etc.) or the different times and locations they conduct business. These user identifiers are closely linked to their respective service contracts, bills, and usage records, but the information directly showing the customer's affiliation may not be complete or consistent.

[0036] To solve the above problems, it is first necessary to extract a series of basic information fields from the database, including but not limited to user identifiers (such as mobile phone numbers, broadband accounts, IPTV usernames, etc.), customer identifiers (a number assigned by the operator's internal system to identify each customer), user names, and any available identification numbers. Then, using the key field of customer ID, query or data mining algorithms can be used to filter out the customer IDs associated with all user identifiers to determine whether these user identifiers point to the same customer.

[0037] Suppose there are three users in the database, with user IDs U1, U2, and U3. Initial data queries reveal that U1, U2, and U3 are each associated with customer ID C1, meaning these three user IDs actually represent different services for the same customer. At this stage, U1, U2, and U3 can be considered a user set, denoted as user set A. The characteristic of set A is that regardless of the number of user IDs, as long as they are associated with the same customer ID, they belong to the same set.

[0038] Step S104: For each user set, connect multiple user identifiers and customer identifiers via bridge lines, and connect customer identifiers and customer names.

[0039] The bridging element is the most direct channel connecting user identifiers and customer identifiers. In practice, for each user set determined in step S102, the customer ID field is looked up to associate multiple user identifiers and customer identifiers within the set. This association ensures that each user identifier is firmly bound to its respective customer, establishing a clear hierarchical relationship. It also connects the customer identifier with the customer name, thus giving each numeric ID a specific entity meaning, such as "Zhang San's". Assume set A contains user identifiers U1, U2, and U3, which are collectively associated with customer ID CIDA. In this case, the bridging element acts like a bridge, directly connecting U1, U2, and U3 to CIDA, while also ensuring that CIDA is connected to the customer name "Zhang San," thus constructing a direct bridge between users and customers.

[0040] Ideally, customer IDs and identification numbers, as identifiers in user data, should be consistent and clear across all user records, serving as the link connecting users and customers. However, due to limitations in the early system architecture, coupled with potential oversights in standardized operations by frontline sales staff when entering customer information, not all user data in the database completely and correctly includes identification information. Given this limitation, the customer ID, with its relatively higher completeness and consistency, becomes the initial and most reliable connection point in building the user relationship network—the cornerstone of the so-called "bridge." Therefore, to ensure the user relationship network starts on the most solid foundation, the customer ID is first used as the core reference to identify and aggregate all user identifiers directly or indirectly belonging to the same customer. This not only effectively avoids the uncertainty caused by missing identification information but also, leveraging the ubiquitous data field of the customer ID, lays a solid foundation for subsequent relationship analysis and network expansion. Through priority consideration, the customer ID, as a building element of the bridge, ensures the accuracy and reliability of the user relationship network in its initial stage.

[0041] Figure 2a This is a schematic diagram of the connection between objects according to an embodiment of this application, such as... Figure 2a As shown, User 1, User 2, User 3, etc. are connected to the same Customer ID through a bridge cable, and the same Customer ID is connected to Customer A through a bridge cable.

[0042] Step S106: For each user set, connect the customer identifier and the contract identifier and identity credential identifier associated with the customer identifier through anchor lines, connect the contract identifier and the customer name, and connect the identity credential identifier and the customer name, wherein the weight of the anchor line is less than the weight of the bridge line.

[0043] For example, based on the existing set A, it is further discovered that U1, U2, and U3 are associated with contract IDs CA1, CA2, and CA3 respectively, and CA1, CA2, and CA3 all point to CIDA. Simultaneously, the identity credential IDs of U1, U2, and U3 are also associated with CIDA. The establishment of these anchor lines means that even without direct customer ID information, indirect association paths can be found through contract IDs or identity credential IDs, closely linking U1, U2, and U3 to "Zhang San's" family.

[0044] Understandably, in the process of building a user network, a series of robust anchor lines are further constructed on top of the existing customer ID bridge, aiming to deepen the network's coherence and accuracy. The construction of these anchor lines focuses on utilizing key information, such as ID numbers and contract numbers, which, while not as ubiquitous as customer IDs, possess equally high identifiable capabilities, to confirm deep connections between users. ID numbers and contract numbers, as important components of user data, are second only to customer IDs in stability, providing an additional dimension of verification to ensure the reliability and breadth of network connections.

[0045] Figure 2b This is a schematic diagram illustrating another connection between objects according to an embodiment of this application, such as... Figure 2b As shown, the same customer ID_1 is connected to the same contract number and the same document through an anchor line, and the same contract number and the same document are connected to customer A through an anchor line.

[0046] Step S108: For each user set, connect the customer identifier and the business object associated with the customer identifier through frame lines, and connect the business object and the customer name through frame lines, wherein the weight of the frame line is less than the weight of the anchor line.

[0047] For example, members U1, U2, and U3 in set A use packages P1, P2, and P3 respectively, and these packages are associated with CIDA. The role of the frame lines is not only to connect CIDA with P1, P2, and P3, but also to reveal the package preferences of "Zhang San's" family members, such as the type and price of the packages, further refining the details of the user association network.

[0048] Understandably, the introduction of frame lines essentially deepens and expands the established anchor relationships, forming a stable triangular framework by linking at least two anchor lines. Figure 2c This is a schematic diagram illustrating another connection between objects according to an embodiment of this application, such as... Figure 2c As shown, the same customer ID_1 is connected to the same package member through the frame line, and the same package member is connected to customer A through the frame line.

[0049] It should be noted that the user data used in steps S102 to S108 includes: basic user information, payment information, terminal information, package information, and behavioral information. Basic user information includes, but is not limited to: customer name, customer ID, user address, and contact number; payment information includes, but is not limited to: contract name, contract ID, and user account; terminal information includes, but is not limited to: broadband and dedicated line terminals, IPTV terminals, and smart home terminals; package information includes, but is not limited to: bundled packages, single-product packages, discount packages, and gift packages; behavioral information includes, but is not limited to: changes in service acceptance, changes in terminal registration and login information, and the activity range of GPS base stations.

[0050] In the above steps, the bridging line connects user and customer identifiers based on the most direct and stable association information, such as customer ID. Since the customer ID is considered the most accurate identifier for a customer, the bridging line is given the highest weight. This means that in scenarios like push notifications, the bridging line connection will be prioritized, and the association information it provides is considered the most reliable foundation. The high weight of the bridging line ensures the primary goal of network construction—accurately identifying the direct affiliation between users and customers. The anchor line establishes a deeper connection based on secondary association information, such as contract number and identity credential number. Compared to the customer ID, this information is less consistent or subject to change in some cases, but it still provides important supplementary information to help confirm indirect relationships between users. The anchor line has a lower weight than the bridging line, indicating that in scenarios like push notifications, the anchor line's association strength and stability are inferior to the bridging line, but it provides necessary support when direct association through customer ID is not possible. The frame line goes a step further, utilizing business data, such as service packages and device usage, to establish relationships between users, focusing on revealing the inherent connections between specific user behaviors. The weight of frame lines is lower than that of anchor lines, which means they are less relevant and direct, but their value lies in their ability to depict user patterns and preferences when using services.

[0051] In summary, the "spider web construction principle" proposed in this application provides a novel perspective and methodology. Starting with the most solid bridge in building a user network (such as customer ID), it gradually constructs a layered and structurally robust network by building anchor lines (such as ID numbers and contract numbers) and framework lines (such as package associations). This strategy of moving from simple to complex and from the core to the edge ensures the comprehensiveness and accuracy of network construction, while also laying a solid foundation for the introduction of a weighted scoring model, enhancing the stability and reliability of the entire network. Furthermore, by accurately identifying relationships between users, including family members, social groups, or corporate customers, more personalized and effective message push strategies can be developed, improving customer service response speed and satisfaction.

[0052] The following are Figure 1The steps shown are illustrated and explained by way of example.

[0053] According to some optional embodiments of this application, for each user set, before connecting the customer identifier and the business object associated with the customer identifier through a frame line, the business object associated with the customer identifier is determined by the following method, wherein the business object includes packages and products: Step S1, determine the package set associated with the customer identifier, and traverse the package set through a first cursor object; Step S2, determine whether the traversal of the package set is complete; if yes, determine the package and product associated with the customer identifier; if no, proceed to step S3; Step S3, obtain the target package identifier currently pointed to by the first cursor object, and retrieve the package basic information corresponding to the target package identifier using the target package identifier as an index; Step S4, determine the target package based on the package basic information. The package identifier is associated with a set of products, and the product set is traversed using a second cursor object; Step S5: Determine whether the traversal of the product set is complete; if yes, obtain the next package in the package set and execute step S2; if no, execute step S6; Step S6: Obtain the target identifier of the current product using the second cursor object, query the product database using the target identifier as the key to obtain the product's basic information; Step S7: Based on the product's basic information, determine whether there are any sub-products associated with the current product; if yes, recursively retrieve the sub-products, store the sub-products in the product relationship chain, and execute step S5; if no, execute step S8; Step S8: Store the current product in the product relationship chain, obtain the next product in the product set, and execute step S5.

[0054] Cursors are an advanced feature in database operations that allow programs to access data in a query result set in a specified order without loading all results at once. The first and second cursor objects are used to iterate through the package set and product set, respectively, and they control the flow and data access within the algorithm. Cursors enable efficient location and processing of data in the database, avoiding performance bottlenecks caused by loading large amounts of data at once. The first cursor object is invoked in steps S1 and S3, while the second cursor object is called in steps S4 and S6.

[0055] The product relationship chain is a data structure used to store and represent the relationships between all business objects (packages and products) used by a customer. During traversal and retrieval, once a product associated with a customer's identifier is identified, including sub-products, they are added one by one to the product relationship chain, forming a chain structure that reflects the customer service's usage history. The product relationship chain not only contains detailed information about the products themselves but also reflects the hierarchical relationships and combinations between products.

[0056] In this embodiment, step S1: First, determine the set of packages associated with a specific customer identifier. This set contains information on all packages currently or historically used by the customer. By introducing a first cursor object, the entire set of packages can be traversed in an orderly manner.

[0057] Step S2: During the traversal of the package set, continuously check whether all packages have been accessed. If the traversal of the package set is complete, it means that all package information related to the customer identifier has been fully collected. At this point, the work of determining the association between packages and products officially begins. Otherwise, if the traversal is not yet complete, proceed to step S3 to continue the retrieval and analysis of package information.

[0058] Step S3: Guided by the first cursor object, determine the identifier of the target package currently being accessed. This identifier serves as a key index for retrieving basic package information. After obtaining the basic package information, retrieve the detailed content of the target package, including but not limited to package type, service scope, and pricing standards.

[0059] Step S4: Based on the detailed information of the target package, further determine the set of all products associated with this package, and introduce a second cursor object to traverse the product set. This operation ensures a thorough exploration of all products related to each package, preparing for the construction of a complete product relationship chain.

[0060] Step S5: Similar to Step S2, determine whether all associated products have been traversed. If the product set has been traversed, the algorithm will move forward, reposition to the next package in the package set, and repeat Step S2 to continue exploring the association between packages and products. If the product set has not been traversed, the algorithm will execute Step S6 to delve into product details.

[0061] Step S6: Obtain the current product identifier, i.e. the target identifier, through the second cursor object, and use this identifier as the query key to retrieve its basic information in the product database.

[0062] Step S7: After obtaining the basic product information, assess whether the current product is associated with any sub-products. If sub-products exist, a recursive approach will be used to explore the information of these sub-products in depth, and they will be stored in the product relationship chain to reflect the hierarchical relationship between products. Then, return to step S5 and continue traversing the product set. If the current product has no sub-products, the algorithm will proceed to step S8.

[0063] Step S8: After confirming that the current product is not associated with any sub-products, store the product information in the product relationship chain to form a coherent service usage record. Then, locate the next product in the product set and repeat the above process until the product set has been traversed, ultimately constructing a complete chain structure that includes all packages and products used by the customer and their potential sub-product relationships.

[0064] on the other hand, Figure 3 This is a flowchart illustrating a method for determining a business object associated with a customer identifier according to an embodiment of this application, such as... Figure 3 As shown, the business object associated with the customer identifier can be determined through the following steps:

[0065] 1. Start: Initiate the algorithm process and prepare for the correlation analysis between packages and products.

[0066] 2. Load Package List: Read all available package information from the database or other data sources. Package information includes the basic attributes of the package and its association with the user.

[0067] 3. Initialize Package Cursor: Set a pointer or marker to track the position of the package record being processed by the current algorithm, ensuring that each package in the package list can be accessed one by one.

[0068] 4. Determine if all packages have been traversed: Check if all packages have been processed. If not, continue with the next steps; if completed, proceed to the output stage. No: Continue execution; Yes: Proceed to step 11.

[0069] 5. Get Current Package ID: Read the unique identifier of the package currently pointed to by the cursor.

[0070] 6. Search basic package information: Get more detailed information about the current package, including package type, price, and included services.

[0071] 7. Initialize Product Cursor: Similar to the package cursor, set a product cursor to track the position of product records.

[0072] 8. Determine if all related products have been traversed: Check if all products related to the current package have been processed. If not, continue with the next steps; if completed, return to step 4 and process the next package. No: Continue; Yes: Proceed to step 10.

[0073] 9. Get Current Product ID: Reads the unique identifier of the product pointed to by the current product cursor.

[0074] 10. Retrieve detailed product information: Obtain all information associated with the current product, including but not limited to product type, functions, and specific association methods with users or packages.

[0075] 11. Check nested association conditions: Analyze whether there are multi-level associations between products and packages. For example, a package may include basic services and additional services, and these services may be associated with different products.

[0076] 12. Determine if sub-products exist: Determine if the current product has any affiliated or bundled products. Yes: Recursively search for sub-products, i.e., return to step 8 and repeat the association search process for sub-products; No: Proceed to step 13.

[0077] 13. Store product relationship chain: Save the association between the current product and related packages.

[0078] 14. Get the next product: Move the product cursor to the next product, and prepare to process the association between the current package and the next product in the package list.

[0079] 15. Get the next package: After completing the processing of the current package and all associated products, the mobile package cursor moves to the next unprocessed package, ready to perform the same association analysis on the next package.

[0080] 16. Output all package-product relationship tree: After all packages have been traversed and processed, summarize and output the complete relationship between packages and products, presented in a tree or network format, to facilitate subsequent data analysis and utilization.

[0081] 17. Generate complete package structure: For each package, build a detailed package structure view based on its relationship with all products.

[0082] 18. End: The algorithm process ends after iteratively searching and constructing the relationship between packages and products.

[0083] According to some optional embodiments of this application, the above data processing method further includes the following steps: for each user set, connecting the business object with the user installation address information, terminal registration address information, terminal usage address information, and call base station address associated with the business object through meridian lines, wherein the weight of the meridian lines is less than the weight of the frame lines; for each user set, determining the similarity between the address information of any two users to obtain one or more similarities; if the average of one or more similarities is greater than a first preset threshold, determining that the user set is labeled as users of the same family; if the average of one or more similarities is greater than a second preset threshold and less than the first preset threshold, determining that the user set is labeled as users of the same customer group.

[0084] In this embodiment, the geographic coordinate data, including the user's installation address, terminal registration address, terminal usage address, and call base station address, are closely linked to the user's service plan and services through meridian lines. Compared to frame lines, meridian lines are given a lower weight in the connection structure. This indicates that although address information is an important reference in association analysis, it provides a supplementary or derivative association clue compared to core connection points such as ID number, contract number, and service plan association.

[0085] Then, to quantify the strength of geographical location associations among members within a user set, an address similarity calculation method is used. This method determines the address distance or similarity between any two users based on their installation address, terminal registration address, terminal usage address, and call base station address. The result of this process may be one or more similarity values, reflecting the spatial density of members within the user set.

[0086] Furthermore, two preset thresholds are set to define whether users belong to the same family or the same customer group. When the calculated similarity value or the average of all similarities is higher than the first preset threshold, it indicates that the members within the user set exhibit extremely high geographical homogeneity, sufficient to identify them as part of the same family unit, and the user set is thus labeled as "users from the same family." Conversely, when the average similarity is between the second and first preset thresholds, although the geographical connection between users is not as strong as that between family members, it still shows a certain degree of clustering, suggesting that they share similar living areas or work environments. Therefore, this type of user set is labeled as "users from the same customer group." This labeling classification provides telecom operators with richer data dimensions, facilitating more in-depth data analysis and improving the accuracy of message push notifications.

[0087] On the other hand, the labels of user sets can also be determined by the following method: Calculate the Si value (similarity index) of user address information, terminal registration address information, terminal usage address information, and user call base station address information using a string matching function to preliminarily infer the user's family and customer group. Extensive data analysis and comparison results show that: when the Si value is between [99% and 100%], users belong to the same address or the same floor; when the Si value is between [95% and 98%], users belong to the same neighborhood (community, campus, factory, company, etc.); when the Si value is between [90% and 94%], users belong to the same street address; and when the Si value is between [80% and 89%], users belong to the same area. It can be understood that the smaller the Si value, the lower the similarity. Specifically, when the Si value is between [99% and 100%], it is determined that users at this address belong to the same customer family. When the Si value is between [95% and 98%], it is determined that users at this address belong to the same customer group.

[0088] The feature data required for the above calculation process, such as basic user information, payment records, and behavioral actions, are all routinely collected and stored, making data collection efficient and inexpensive. Meanwhile, by making implicit features explicit (e.g., quantifying geographical similarity as a Si value) and adopting an intuitive weighted scoring model, the calculation logic becomes simple and easy to understand, reducing the difficulty of technical implementation and facilitating rapid deployment and widespread application.

[0089] Preferably, Figure 2d This is a schematic diagram of the connection between objects according to another embodiment of this application, by Figure 2d It can be seen that the above data processing method also includes the following steps: For each user set, the call base station address and the call frequency information and call time information associated with the call base station address are connected by auxiliary lines, wherein the weight of the auxiliary lines is less than the weight of the meridian lines.

[0090] It should be noted that the auxiliary line connects a user's base station address with their call frequency and time information, providing in-depth insights into the user's communication habits. Compared to the meridian line, the auxiliary line carries a lighter weight in the structure of the user's association network. This reflects that although call behavior is one of the important reference indicators for user association, its position in the overall association judgment is slightly lower than that of geographical location information, which is regarded as more direct and solid evidence of association.

[0091] As some optional embodiments of this application, the above data processing method further includes the following steps: for each user set, determining whether the relationship between users belongs to a preset tag, and determining the call frequency and call duration between users; determining the correlation index between users based on whether the relationship between users belongs to a preset tag, the call frequency and call duration between users; if the correlation index is greater than a third preset threshold, determining that the user set is labeled as users of the same family; if the correlation index is greater than a fourth preset threshold and less than a third preset threshold, determining that the user set is labeled as users of the same customer group.

[0092] Specifically, the frequency and duration of calls between users can be determined as follows: For any two users in the user set, the ratio of the number of calls within a preset time window to the length of the preset time window is determined as the call frequency, and the ratio of the number of calls within a preset peak period to the total number of calls within the preset time window is determined as the call duration.

[0093] Specifically, determining whether the relationship between users belongs to a preset label can be achieved as follows: A random forest classification model is used to process user personal information to determine whether the relationship between users belongs to a preset label. The preset label includes: belonging to the same family. The random forest classification model is trained as follows: A training dataset is obtained, which includes a set of user personal information, including age, gender, occupation, and surname fields. Based on the training dataset, the global minimum and global maximum values ​​of the age field are calculated. These global minimum and global maximum values ​​are then used to linearly transform the original age to a preset interval, obtaining the normalized age feature. [Further details on age feature acquisition are needed.] The system identifies the mapping rules between strings and numbers, converts the gender field into discrete numbers based on these rules, and obtains numerical gender features. It extracts the occupation field from the training dataset, generates a binary vector representation of the occupation field, and obtains one-hot encoded occupation features. It obtains a mapping dictionary from the surname field to family identifiers, converts the surname field into family identifiers by querying the dictionary, and obtains family identifier features. For any two users in the training dataset, they are labeled positively if their family identifier features are the same, and negatively otherwise. The normalized age features, numerical gender features, one-hot encoded occupation features, and family identifier features are concatenated into an input vector, and the positive and negative labels are used as supervision signals to train a random forest classification model.

[0094] On the other hand, the above-mentioned method for determining the tags corresponding to a user set can also be used by... Figure 4 The steps shown are to be followed.

[0095] First, call frequency is calculated using the following formula: Call Frequency = Total Number of Calls / Total Time Span. Call frequency directly reflects a user's call activity level within a given time window (such as a week or a month), providing a basis for identifying users with high-frequency calls.

[0096] Secondly, the call duration is calculated using a specific formula: Call Duration = Total Peak-Hour Calls / Total Calls. The call duration measures the percentage of a user's calls made during peak hours.

[0097] Then, the static attributes of users, such as age, gender, occupation, and surname, are converted into quantifiable feature vectors (*personal information score). For age, normalization is used to limit its range to 0 to 1, highlighting its relative importance across different age groups; gender is represented by concise numerical encoding (male = 0, female = 1); occupation information uses one-hot encoding, transforming categorical variables into multiple binary variables, with each occupation corresponding to an independent encoding bit; and surname features are directly mapped to potential family branches through string mapping. Based on this foundation, a random forest learning model, named "personal information score," is trained using the real label of "whether or not they belong to the same family," to capture the intrinsic relationship between user personal information and family affiliation.

[0098] Based on the above indicators, the weighted scoring model organically integrates call behavior scores and personal information scores, expressed by the following formula: S = w1 * call frequency + w2 * call duration + w3 * personal information score, where w1 + w2 + w3 = 1. In the model, the determination of each weight coefficient needs to be based on the actual distribution of the data, determined through multiple experimental comparisons and parameter optimization to maximize evaluation accuracy. The S value, as the final comprehensive score, directly correlates with the strength of the connection between users. A higher S value indicates that users exhibit high similarity in call frequency, peak-hour call preferences, and personal information, suggesting they may belong to the same family or social group. Conversely, a lower S value indicates a looser connection between users, possibly belonging to a more generalized customer group.

[0099] In summary, by combining IT technology and big data models, this application can deeply mine the user relationships hidden in telecommunications data. It not only leverages direct, explicit features (such as shared IDs, contracts, or document information) but also utilizes indirect behavioral patterns (such as call frequency, time of day, and geographical location) to construct deeper connections between users. This progressive association method starts with basic user information and gradually delves into user behavior habits and social connections, achieving a comprehensive exploration and precise characterization of user relationships. Compared to traditional association analysis techniques, this application significantly improves both identification accuracy and coverage. Furthermore, the weighted scoring model allows for dynamic adjustment of the scoring function weights (w1, w2, w3) based on the actual distribution of the data, providing flexibility and adaptability. Through comparative verification with real-world scenarios and continuous optimization of model parameters, the accuracy and efficiency of identification can be gradually improved in practical applications, ensuring that the technical solution can adapt to the ever-changing market environment and user behavior, and continuously provide high-quality user association analysis services.

[0100] Figure 5 This is a structural diagram of a data processing apparatus according to an embodiment of this application, such as... Figure 5 As shown, the device includes:

[0101] The determination module 52 is used to determine multiple user sets among multiple different users, wherein each user set includes: multiple user identifiers, and the customer identifier associated with each user identifier is the same.

[0102] The first connection module 54 is used to connect multiple user identifiers and customer identifiers via bridge lines for each user set, and to connect customer identifiers and customer names.

[0103] The second connection module 56 is used to connect, for each user set, a customer identifier and the contract identifier and identity credential identifier associated with the customer identifier through anchor lines, connect the contract identifier and the customer name, and connect the identity credential identifier and the customer name, wherein the weight of the anchor line is less than the weight of the bridge line.

[0104] The third connection module 58 is used to connect customer identifiers and business objects associated with customer identifiers, and to connect business objects and customer names, for each user set, through frame lines, wherein the weight of the frame line is less than the weight of the anchor line.

[0105] Optionally, for each user set, before connecting the customer identifier and the business object associated with the customer identifier through the frame line, the business object associated with the customer identifier is determined by the following method, wherein the business object includes packages and products: Step S1, determine the package set associated with the customer identifier, and traverse the package set through the first cursor object; Step S2, determine whether the traversal of the package set is complete; if yes, determine the packages and products associated with the customer identifier; if no, proceed to step S3; Step S3, obtain the target package identifier currently pointed to by the first cursor object, and retrieve the package basic information corresponding to the target package identifier using the target package identifier as an index; Step S4, determine the associated packages of the target package identifier based on the package basic information. The product set is traversed using a second cursor object; Step S5: Determine if the traversal of the product set is complete; if yes, retrieve the next package in the package set and execute step S2; if no, execute step S6; Step S6: Obtain the target identifier of the current product using the second cursor object, query the product database using the target identifier as the key to obtain the product's basic information; Step S7: Based on the product's basic information, determine if there are any sub-products associated with the current product; if yes, recursively retrieve the sub-products, store the sub-products in the product relationship chain, and execute step S5; if no, execute step S8; Step S8: Store the current product in the product relationship chain, retrieve the next product in the product set, and execute step S5.

[0106] Optionally, the data processing device is further configured to perform the following steps: for each user set, connecting the business object with the user installation address information, terminal registration address information, terminal usage address information, and call base station address associated with the business object through a meridian line, wherein the weight of the meridian line is less than the weight of the frame line; for each user set, determining the similarity between the address information of any two users to obtain one or more similarities; if the average of one or more similarities is greater than a first preset threshold, determining that the user set is labeled as users of the same family; if the average of one or more similarities is greater than a second preset threshold and less than the first preset threshold, determining that the user set is labeled as users of the same customer group.

[0107] Optionally, the data processing device is further configured to perform the following steps: for each user set, connecting the call base station address and the call frequency information and call time information associated with the call base station address via auxiliary lines, wherein the weight of the auxiliary lines is less than the weight of the meridian lines.

[0108] Optionally, the data processing device is further configured to perform the following steps: for each user set, determine whether the relationship between users belongs to a preset label, and determine the call frequency and call duration between users; based on whether the relationship between users belongs to a preset label, the call frequency and call duration between users, determine the correlation index between users; if the correlation index is greater than a third preset threshold, determine that the user set is labeled as users of the same family; if the correlation index is greater than a fourth preset threshold and less than a third preset threshold, determine that the user set is labeled as users of the same customer group.

[0109] Optionally, determining the call frequency and call duration between users includes the following steps: for any two users in the user set, the ratio of the number of calls within a preset time window to the length of the preset time window is determined as the call frequency, and the ratio of the number of calls within a preset peak period to the total number of calls within the preset time window is determined as the call duration.

[0110] Optionally, determining whether the relationship between users belongs to a preset label includes the following steps: processing user personal information using a random forest classification model to determine whether the relationship between users belongs to a preset label, wherein the preset label includes: "same family". The random forest classification model is trained using the following method: obtaining a training dataset, wherein the training dataset includes a set of user personal information, including: age field, gender field, occupation field, and surname field; calculating the global minimum and global maximum values ​​of the age field based on the training dataset, and using the global minimum and global maximum values ​​to linearly transform the original age to a preset interval to obtain the normalized age feature; obtaining the gender... The system employs a mapping rule between strings and numbers to convert the gender field into discrete numerical values, resulting in numerical gender features. It extracts the occupation field from the training dataset, generates a binary vector representation of the occupation field, and obtains the occupational one-hot encoded features. A mapping dictionary from the surname field to family identifiers is obtained, and the surname field is converted into family identifiers by querying the dictionary, yielding family identifier features. For any two users in the training dataset, they are labeled positively if their family identifier features are the same, and negatively otherwise. The normalized age features, numerical gender features, occupational one-hot encoded features, and family identifier features are concatenated into an input vector, and the positive and negative labels are used as supervisory signals to train a random forest classification model.

[0111] It should be noted that the above Figure 5 The modules in the above can be program modules (e.g., a set of program instructions that implement a specific function) or hardware modules. For the latter, they can be represented in the following forms, but are not limited to these: each of the above modules is represented by a processor, or the functions of each of the above modules are implemented by a processor.

[0112] It should be noted that, Figure 5 Preferred embodiments of the shown examples can be found in [reference needed]. Figure 1 The relevant descriptions of the embodiments shown will not be repeated here.

[0113] Figure 6 A hardware structure block diagram of a computer terminal for implementing a data processing method is shown. Figure 6 As shown, the computer terminal 60 may include one or more processors 602 (shown as 602a, 602b, ..., 602n in the figure) 602 (processor 602 may include, but is not limited to, a microprocessor MCU or a programmable logic device FPGA, etc.), a memory 604 for storing data, and a transmission module 606 for communication functions. In addition, it may also include: a display, an input / output interface (I / O interface), a universal serial bus (USB) port (which may be included as one of the ports of a BUS bus), a network interface, a power supply, and / or a camera. Those skilled in the art will understand that... Figure 6 The structure shown is for illustrative purposes only and does not limit the structure of the aforementioned electronic device. For example, the computer terminal 60 may also include... Figure 6 The more or fewer components shown, or having the same Figure 6 The different configurations shown.

[0114] It should be noted that the aforementioned one or more processors 602 and / or other data processing circuits are generally referred to herein as "data processing circuits". These data processing circuits may be embodied, in whole or in part, in software, hardware, firmware, or any other combination thereof. Furthermore, the data processing circuits may be a single, independent processing module, or may be integrated, in whole or in part, into any other element within the computer terminal 60. As involved in the embodiments of this application, the data processing circuits serve as a form of processor control (e.g., selection of a variable resistor termination path connected to an interface).

[0115] The memory 604 can be used to store software programs and modules of application software, such as the program instructions / data storage device corresponding to the data processing method in this embodiment. The processor 602 executes various functional applications and data processing by running the software programs and modules stored in the memory 604, thereby implementing the aforementioned data processing method. The memory 604 may include high-speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory 604 may further include memory remotely located relative to the processor 602, and these remote memories can be connected to the computer terminal 60 via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0116] The transmission module 606 is used to receive or send data via a network. Specific examples of the network described above may include a wireless network provided by the communication provider of the computer terminal 60. In one example, the transmission module 606 includes a Network Interface Controller (NIC), which can connect to other network devices via a base station to communicate with the Internet. In another example, the transmission module 606 may be a Radio Frequency (RF) module, used for wireless communication with the Internet.

[0117] The display may be, for example, a touchscreen liquid crystal display (LCD) that allows the user to interact with the user interface of the computer terminal 60.

[0118] It should be noted here that, in some optional embodiments, the above... Figure 6 The computer terminal shown may include hardware elements (including circuitry), software elements (including computer code stored on a computer-readable medium), or a combination of both hardware and software elements. It should be noted that... Figure 6 This is only one instance of a specific particular instance, and is intended to illustrate the types of components that may exist in the aforementioned computer terminal.

[0119] It should be noted that, Figure 6 The computer terminal shown is used to execute Figure 1 The data processing method shown above is also applicable to this electronic device, and will not be repeated here.

[0120] This application also provides a non-volatile storage medium, which includes a stored program, wherein the program controls the device where the storage medium is located to execute the above data processing method when it runs.

[0121] A non-volatile storage medium performs the following functions: Identifying multiple user sets among multiple distinct users, wherein each user set includes: multiple user identifiers, each associated with a consistent customer identifier; for each user set, connecting the multiple user identifiers and customer identifiers via bridges, and connecting the customer identifiers and customer names; for each user set, connecting the customer identifier with its associated contract identifier and identity credential identifier via anchors, connecting the contract identifier with the customer name, and connecting the identity credential identifier with the customer name, wherein the anchors have a smaller weight than the bridges; for each user set, connecting the customer identifier with its associated business object via frame lines, and connecting the business object with the customer name, wherein the frame lines have a smaller weight than the anchors.

[0122] This application also provides an electronic device, including a memory and a processor, wherein the processor is used to run a program stored in the memory, wherein the program executes the above-described data processing method during runtime.

[0123] The processor is used to run a program that performs the following functions: identifying multiple user sets among multiple distinct users, wherein each user set includes: multiple user identifiers, each associated with a consistent customer identifier; for each user set, connecting multiple user identifiers and customer identifiers via bridge lines, and connecting customer identifiers and customer names; for each user set, connecting a customer identifier with its associated contract identifier and identity credential identifier via anchor lines, connecting the contract identifier with the customer name, and connecting the identity credential identifier with the customer name, wherein the weight of the anchor lines is less than the weight of the bridge lines; for each user set, connecting a customer identifier with its associated business object via frame lines, and connecting the business object with the customer name, wherein the weight of the frame lines is less than the weight of the anchor lines.

[0124] The sequence numbers of the embodiments in this application are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0125] In the above embodiments of this application, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0126] In the above embodiments of this application, the information collected is 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 all comply with relevant laws, regulations and standards, take necessary protective measures, do not violate public order and good morals, and provide corresponding operation entry points for users to choose to authorize or refuse.

[0127] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. The device embodiments described above are merely illustrative; for example, the division of units can be a logical functional division, and in actual implementation, there may be other division methods. For instance, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the displayed or discussed mutual coupling, direct coupling, or communication connection may be through some interfaces; the indirect coupling or communication connection between units or modules may be electrical or other forms.

[0128] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0129] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0130] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to related technologies, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, read-only memory (ROM), random access memory (RAM), portable hard drives, magnetic disks, or optical disks.

[0131] The above description is only a preferred embodiment of this application. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of this application, and these improvements and modifications should also be considered within the scope of protection of this application.

Claims

1. A data processing method, characterized in that, include: Multiple user sets are identified among multiple different users, wherein each user set includes: multiple user identifiers, and each user identifier is associated with the same customer identifier; For each user set, multiple user identifiers and customer identifiers are connected via bridge lines, and the customer identifiers and customer names are connected; For each user set, the customer identifier is connected to the contract identifier and identity credential identifier associated with the customer identifier via anchor lines, the contract identifier is connected to the customer name, and the identity credential identifier is connected to the customer name, wherein the weight of the anchor line is less than the weight of the bridge line. For each user set, the customer identifier and the business object associated with the customer identifier are connected by a frame line, and the business object and the customer name are connected by a frame line, wherein the weight of the frame line is less than the weight of the anchor line.

2. The method according to claim 1, characterized in that, For each user set, before connecting the customer identifier and the business object associated with the customer identifier via a frame line, the business object associated with the customer identifier is determined by the following method, wherein the business object includes packages and products: Step S1: Determine the set of packages associated with the customer identifier, and traverse the set of packages using the first cursor object; Step S2: Determine whether the traversal of the package set has been completed; if yes, determine the packages and products associated with the customer identifier; if no, proceed to step S3. Step S3: Obtain the target package identifier currently pointed to by the first cursor object, and retrieve the basic package information corresponding to the target package identifier using the target package identifier as an index; Step S4: Based on the basic information of the package, determine the set of products associated with the target package identifier, and traverse the set of products using the second cursor object; Step S5: Determine whether the traversal of the product set is complete; if yes, obtain the next package in the package set and execute step S2; if no, execute step S6. Step S6: Obtain the target identifier of the current product through the second cursor object, and query the product database with the target identifier as the key to obtain the basic product information; Step S7: Based on the product basic information, determine whether there is a sub-product associated with the current product; if yes, recursively search for the sub-product, store the sub-product in the product relationship chain, and execute step S5; if no, execute step S8. Step S8: Store the current product in the product relationship chain, obtain the next product in the product set, and execute step S5.

3. The method according to claim 1, characterized in that, The method further includes: For each user set, the service object and the user installation address information, terminal registration address information, terminal usage address information and call base station address associated with the service object are connected by meridian lines, wherein the weight of the meridian lines is less than the weight of the frame lines; For each set of users, determine the similarity between the address information of any two users to obtain one or more similarities; If one or more similarities are greater than a first preset threshold, the user set is identified as users from the same family. If one or more similarities are greater than a second preset threshold and less than the first preset threshold, the user set is identified as a user belonging to the same customer group.

4. The method according to claim 3, characterized in that, The method further includes: For each user set, the call base station address and the call frequency information and call time information associated with the call base station address are connected by an auxiliary line, wherein the weight of the auxiliary line is less than the weight of the meridian line.

5. The method according to claim 1, characterized in that, The method further includes: For each user set, determine whether the relationship between users belongs to a preset tag, and determine the call frequency and call duration between users; Based on whether the relationship between users belongs to the preset tag, the frequency of calls between users, and the call duration, the correlation index between users is determined; If the correlation index is greater than a third preset threshold, the user set is identified as users belonging to the same family. If the correlation index is greater than the fourth preset threshold and less than the third preset threshold, the user set is identified as users belonging to the same customer group.

6. The method according to claim 5, characterized in that, Determine the frequency and duration of calls between users, including: For any two users in the user set, the ratio of the number of calls within a preset time window to the length of the preset time window is determined as the call frequency, and the ratio of the number of calls during a preset peak period to the total number of calls within the preset time window is determined as the call period.

7. The method according to claim 5, characterized in that, Determining whether the relationship between users falls under a preset tag includes: A random forest classification model is used to process user information to determine whether relationships between users belong to a preset label, wherein the preset label includes: belonging to the same family. The random forest classification model is trained using the following method: Obtain a training dataset, wherein the training dataset includes a set of user personal information, which includes: an age field, a gender field, an occupation field, and a surname field; Based on the training dataset, calculate the global minimum and global maximum values ​​of the age field, and use the global minimum and global maximum values ​​to linearly transform the original age to a preset range to obtain normalized age features. Obtain the mapping rules between gender strings and numerical values, and convert the gender field into discrete numerical values ​​according to the mapping rules to obtain numerical gender features; Extract the occupation field from the training dataset, generate the binary vector representation corresponding to the occupation field, and obtain the occupation one-hot encoded feature; Obtain a mapping dictionary from the surname field to the family identifier, and convert the surname field into the family identifier by querying the mapping dictionary to obtain the family identifier feature; For any two users in the training dataset, if the family identifier features are the same, they are labeled as positive; otherwise, they are labeled as negative. The normalized age feature, the numerical gender feature, the occupational one-hot encoding feature, and the family identifier feature are concatenated into an input vector, and the positive and negative labels are used as supervision signals to train a random forest classification model.

8. A data processing apparatus, characterized in that, include: The determination module is used to determine multiple user sets among multiple different users, wherein each user set includes: multiple user identifiers, and each user identifier is associated with the same customer identifier; The first connection module is used to connect multiple user identifiers and customer identifiers via bridge lines for each user set, and to connect the customer identifiers and customer names; The second connection module is used to connect the customer identifier and the contract identifier and identity credential identifier associated with the customer identifier, connect the contract identifier and the customer name, and connect the identity credential identifier and the customer name for each user set via anchor lines, wherein the weight of the anchor line is less than the weight of the bridge line. The third connection module is used to connect the customer identifier and the business object associated with the customer identifier through a frame line for each user set, and to connect the business object and the customer name, wherein the weight of the frame line is less than the weight of the anchor line.

9. A non-volatile storage medium, characterized in that, The non-volatile storage medium includes a stored program, wherein, when the program is executed, it controls the device containing the non-volatile storage medium to perform the data processing method according to any one of claims 1 to 7.

10. An electronic device, characterized in that, include: A memory and a processor, the processor being configured to run a program stored in the memory, wherein the program, when running, performs the data processing method according to any one of claims 1 to 7.

11. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the data processing method according to any one of claims 1 to 7.

Images