Method, device, equipment, medium and program product for constructing comprehensive user portrait

Abstract

The application provides a comprehensive user portrait construction method and device, equipment, medium and program product, relates to the computer technical field, and is used for improving the accuracy of the constructed user portrait. The specific technical scheme is: obtaining communication service order information, payment behavior information and identity verification information of a first user in a first operator; constructing a user order portrait based on the communication service order information; constructing a user payment portrait based on the payment behavior information, the user payment portrait representing the consumption ability and consumption willingness strength of the first user; constructing a user identity portrait based on the identity verification information; and constructing a comprehensive user portrait based on the user order portrait, the user payment portrait and the user identity portrait. The application is applied to the scene of constructing a user portrait.

Description

Technical Field

[0001] This application relates to the field of computer technology, and in particular to a method, apparatus, device, medium and program product for constructing a comprehensive user profile. Background Technology

[0002] With the rapid development of the telecommunications industry, the proportion of new user orders in operators' telecommunications service contracts continues to increase. Therefore, how to accurately construct user profiles for new users in the absence of their historical consumption records, in order to achieve risk control, targeted marketing, and personalized services for them, has become an urgent problem to be solved.

[0003] In related technologies, based on information such as gender, age, occupation, and interests actively provided by new users when registering an account on the platform, collaborative filtering algorithms are used to find other existing users similar to the new users. Then, based on the user profiles of the other existing users found, a user profile of the new user is constructed.

[0004] However, during the implementation of the above scheme, due to the scarcity of information actively provided by users, the similarity between the other existing users found by the system and the new users is low. This results in low accuracy of the user profiles constructed for the new users, making it impossible for the system to effectively support risk control, accurate recommendations, and personalized services for new users based on these profiles. Thus, the user profiles constructed in this related technology have low accuracy. Summary of the Invention

[0005] This application provides a comprehensive user profile construction method, apparatus, device, medium, and program product to improve the accuracy of the constructed user profile.

[0006] In a first aspect, embodiments of this application provide a method for constructing a comprehensive user profile. The method includes: obtaining communication service order information, payment behavior information, and identity verification information of a first user from a first operator; constructing a user order profile based on the communication service order information, wherein the user order profile represents the first user's communication needs preferences and the service attributes of the services already processed; constructing a user payment profile based on the payment behavior information, wherein the user payment profile represents the first user's spending power and willingness to spend; constructing a user identity profile based on the identity verification information, wherein the user identity profile represents the first user's identity attributes, accessibility service needs, and family structure; and constructing a comprehensive user profile based on the user order profile, the user payment profile, and the user identity profile.

[0007] The technical solution provided in this application brings at least the following beneficial effects: by separately acquiring the communication service order information, payment behavior information, and identity verification information of new users, it is possible to construct user order profiles, user payment profiles, and user identity profiles of new users from multiple dimensions, and then merge the constructed multiple profiles to obtain a comprehensive user profile that can reflect user characteristics from multiple perspectives. Based on this comprehensive user profile, a target strategy corresponding to the user can be generated, avoiding the problem of low accuracy of user profiles caused by constructing user profiles based solely on information actively entered by new users when registering an account, and improving the accuracy of the constructed user profiles.

[0008] One possible implementation is that the aforementioned communication service order information includes: user address information; the construction of a user order profile based on the aforementioned communication service order information includes: parsing the aforementioned user address information using a natural language processing algorithm to obtain an address parsing result, which represents the actual geographical location corresponding to the aforementioned user address information; determining an address value tag based on the aforementioned address parsing result and a preset knowledge graph, which represents the consumption potential corresponding to the aforementioned actual geographical location; and constructing the aforementioned user order profile based on the aforementioned address value tag.

[0009] Another possible implementation is that the aforementioned payment behavior information includes: payment time information, which includes: service selection time information and payment completion time information; the aforementioned construction of a user payment profile based on the aforementioned payment behavior information includes: determining a user payment behavior tag based on the aforementioned service selection time information and the aforementioned payment completion time information, the user payment behavior tag representing the behavioral characteristics of the first user when paying for a communication service order; and constructing the aforementioned user payment profile based on the aforementioned user payment behavior tag.

[0010] Another possible implementation is that the aforementioned authentication information includes: an authentication image; the aforementioned construction of a user identity profile based on the aforementioned authentication information includes: recognizing the aforementioned authentication image using an optical character recognition algorithm to obtain consumption level information; determining a user identity tag based on the consumption level information, the user identity tag representing the social identity of the aforementioned first user; and constructing the aforementioned user identity profile based on the aforementioned user identity tag.

[0011] Another possible implementation is that the aforementioned communication service order information also includes at least one of the following: service package information, user age information, and user gender information; the aforementioned payment behavior information also includes at least one of the following: payment channel information, payment status information, and payment device and network environment information; the aforementioned identity verification information also includes at least one of the following: ID photo, occupation information, accessibility needs information, and family structure information.

[0012] Another possible implementation is that the aforementioned comprehensive user profile is used to generate the target strategy corresponding to the first user, and the target strategy includes at least one of the following: security protection strategy and business recommendation strategy.

[0013] Secondly, embodiments of this application provide a comprehensive user profile construction apparatus, comprising: an acquisition module and a processing module; the acquisition module is used to acquire communication service order information, payment behavior information, and identity verification information of a first user from a first operator; the processing module is used to construct a user order profile based on the communication service order information acquired by the acquisition module, the user order profile representing the communication needs preferences and service attributes of the services already processed by the first user; and, based on the payment behavior information, construct a user payment profile, the user payment profile representing the consumption capacity and consumption willingness of the first user; and, based on the identity verification information, construct a user identity profile, the user identity profile representing the identity attributes, accessibility service needs, and family structure of the first user; and, based on the user order profile, the user payment profile, and the user identity profile, construct a comprehensive user profile, the comprehensive user profile being used to generate a target strategy corresponding to the first user, the target strategy including at least one of the following: a security protection strategy and a service recommendation strategy.

[0014] One possible implementation is that the aforementioned communication service order information includes: user address information; the aforementioned processing module is specifically used to: parse the aforementioned user address information using a natural language processing algorithm to obtain an address parsing result, the address parsing result representing the actual geographical location corresponding to the aforementioned user address information; determine an address value tag based on the address parsing result and a preset knowledge graph, the address value tag representing the consumption potential corresponding to the aforementioned actual geographical location; and construct the aforementioned user order profile based on the aforementioned address value tag.

[0015] Another possible implementation is that the aforementioned payment behavior information includes: payment time information, which includes: service selection time information and payment completion time information; the aforementioned processing module is specifically used to: determine user payment behavior tags based on the aforementioned service selection time information and the aforementioned payment completion time information, the user payment behavior tags representing the behavioral characteristics of the aforementioned first user when paying for a communication service order; and construct the aforementioned user payment profile based on the aforementioned user payment behavior tags.

[0016] Another possible implementation is that the above processing module is specifically used to: identify the above identity verification image through an optical character recognition algorithm to obtain consumption level information; determine a user identity tag based on the consumption level information, the user identity tag representing the social identity of the above first user; and construct the above user identity profile based on the above user identity tag.

[0017] Another possible implementation is that the aforementioned communication service order information also includes at least one of the following: service package information, user age information, and user gender information; the aforementioned payment behavior information also includes at least one of the following: payment channel information, payment status information, and payment device and network environment information; the aforementioned identity verification information also includes at least one of the following: ID photo, occupation information, accessibility needs information, and family structure information.

[0018] Another possible implementation is that the aforementioned comprehensive user profile is used to generate the target strategy corresponding to the first user, and the target strategy includes at least one of the following: security protection strategy and business recommendation strategy.

[0019] Thirdly, this application provides an electronic device comprising: a processor and a memory; the memory stores a program or instructions executable on the processor, wherein the program or instructions, when executed by the processor, implement the method of the first aspect described above.

[0020] Fourthly, this application provides a readable storage medium on which a program or instructions are stored, which, when executed by a computer, implement the method of the first aspect described above.

[0021] Fifthly, this application provides a computer program product stored in a storage medium, which, when executed by a computer, implements the method described in the first aspect.

[0022] In a sixth aspect, embodiments of this application provide a chip including a processor and a communication interface, wherein the communication interface is coupled to the processor, and the processor is used to run programs or instructions to implement the method described in the first aspect.

[0023] The beneficial effects of the second to sixth aspects mentioned above are described in the corresponding description of the first aspect and will not be repeated here. Attached Figure Description

[0024] Figure 1 A schematic diagram of the network architecture for an application of a comprehensive user profile construction method provided in this application embodiment;

[0025] Figure 2 A flowchart illustrating a comprehensive user profile construction method provided in this application embodiment;

[0026] Figure 3 A flowchart illustrating another comprehensive user profile construction method provided in this application embodiment;

[0027] Figure 4 A flowchart illustrating another comprehensive user profile construction method provided in this application embodiment;

[0028] Figure 5 A flowchart illustrating another comprehensive user profile construction method provided in this application embodiment;

[0029] Figure 6 A schematic diagram of the application layer of a comprehensive user profile construction device provided in this application embodiment;

[0030] Figure 7 A schematic diagram of the image inference layer of a comprehensive user profile construction device provided in this application embodiment;

[0031] Figure 8 A schematic diagram of a multimodal fusion layer of a comprehensive user profile construction device provided in this application embodiment;

[0032] Figure 9 A schematic diagram of the feature extraction layer of a comprehensive user profile construction device provided in this application embodiment;

[0033] Figure 10 A schematic diagram of the data access layer of a comprehensive user profile construction device provided in this application embodiment;

[0034] Figure 11 A schematic diagram of a comprehensive user profile construction device provided in this application embodiment;

[0035] Figure 12 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0036] The following section will describe in detail the comprehensive user profile construction method, apparatus, equipment, medium, and program products provided in this application, with reference to the accompanying drawings.

[0037] The technical solutions of the embodiments of this application will be clearly described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application are within the scope of protection of this application.

[0038] The terms "first," "second," etc., used in the specification and claims of this application are used to distinguish similar objects and not to describe a specific order or sequence. It should be understood that such use of data can be interchanged where appropriate so that embodiments of this application can be implemented in orders other than those illustrated or described herein, and the objects distinguished by "first," "second," etc., are generally of the same class and the number of objects is not limited; for example, a first object can be one or more. Furthermore, in the specification and claims, "and / or" indicates at least one of the connected objects, and the character " / " generally indicates that the preceding and following objects are in an "or" relationship.

[0039] The terms "at least one," "at least one of," etc., used in the specification and claims of this application refer to any one, any two, or a combination of two or more of the included items. For example, at least one of a, b, and c can mean: "a," "b," "c," "a and b," "a and c," "b and c," and "a, b, and c," where a, b, and c can be single or multiple. Similarly, "at least two" refers to two or more items, and its meaning is similar to that of "at least one."

[0040] In the description of this application, unless otherwise stated, "a plurality of" means two or more.

[0041] The embodiments of this application provide a comprehensive user profile construction method, apparatus, device, medium, and program product, which can be applied to scenarios of constructing user profiles.

[0042] In the telecommunications operator business, new mobile phone and broadband orders account for a significant proportion. These new users typically lack historical consumption data, resulting in a "cold start" problem. Among related technologies, user profiling techniques largely rely on long-term behavioral sequences, making them unsuitable for this scenario. Currently, operators' understanding of new users is often limited to basic real-name information and selected service plans, leading to issues such as reliance on simplistic rules for risk control, crude marketing recommendations, and one-size-fits-all service.

[0043] In related technologies, initial recommendations are made using interest tags filled in by users during registration and collaborative filtering algorithms. However, users actively provide very little interest information, and the communication products are not very different, limiting the effectiveness of collaborative filtering. This results in low similarity between the existing users found by the system and the new users, leading to low accuracy in the user profiles constructed for the new users. Consequently, the system cannot effectively support risk control, accurate recommendations, and personalized services for new users based on these user profiles. Thus, the user profiles constructed in these related technologies have low accuracy.

[0044] To address the aforementioned technical issues, embodiments of this application provide a comprehensive user profile construction method, apparatus, device, medium, and program product. This method acquires a first user's communication service order information, payment behavior information, and identity verification information from a first operator. Based on the communication service order information, a user order profile is constructed, representing the first user's communication needs preferences and the service attributes of the services already subscribed to. Based on the payment behavior information, a user payment profile is constructed, representing the first user's spending power and willingness to spend. Based on the identity verification information, a user identity profile is constructed, representing the first user's identity attributes, accessibility service needs, and family structure. Based on the user order profile, user payment profile, and user identity profile, a comprehensive user profile is constructed. This comprehensive user profile is used to generate a target strategy corresponding to the first user, and the target strategy includes at least one of the following: a security protection strategy and a service recommendation strategy. In this solution, by separately acquiring the new user's communication service order information, payment behavior information, and identity verification information, user order profiles, user payment profiles, and user identity profiles are constructed from multiple dimensions. These multiple profiles are then merged to obtain a comprehensive user profile that reflects user characteristics from multiple perspectives. Based on this comprehensive user profile, corresponding target strategies for the user are generated. This avoids the problem of low accuracy in user profiles caused by relying solely on information actively entered by new users during account registration, and improves the accuracy of the constructed user profiles.

[0045] The following description, in conjunction with the accompanying drawings, details the comprehensive user profile construction method, apparatus, device, medium, and program products provided in the embodiments of this application.

[0046] Figure 1 This illustration shows the network architecture of a comprehensive user profile construction method provided in an embodiment of this application. For example... Figure 1 As shown, the network architecture includes a comprehensive user profile building device 101 and a terminal device 102. The comprehensive user profile building device 101 and the terminal device 102 are interconnected.

[0047] In some embodiments, the comprehensive user profile building apparatus 101 may be a server, a computer, or a processor or processing unit within a server or computer. The server may be a single server or a server cluster consisting of multiple servers. It should be noted that the embodiments of this application do not limit the specific device form of the comprehensive user profile building apparatus 101. Figure 1 The example of the integrated user profile building device 101 is a single server.

[0048] In some embodiments, the terminal device may be a mobile phone, tablet computer, laptop computer, handheld computer, in-vehicle electronic device, mobile internet device (MID), augmented reality (AR) / virtual reality (VR) device, robot, wearable device, personal computer (PC), ultra-mobile personal computer (UMPC), netbook, or personal digital assistant (PDA), etc., and the embodiments of this application do not specifically limit it. Figure 1 The example shown is a mobile phone, with terminal device 102 as an example.

[0049] It should be noted that the network architecture described in the embodiments of this application is for the purpose of more clearly illustrating the technical solutions of the embodiments of this application, and does not constitute a limitation on the technical solutions provided in the embodiments of this application. As network architectures evolve, the technical solutions provided in the embodiments of this application are also applicable to similar technical problems.

[0050] See Figure 2 This is a flowchart illustrating a comprehensive user profile construction method provided in an embodiment of this application. Figure 2 As shown, the comprehensive user profile construction method provided in this application embodiment can be implemented by the above-mentioned comprehensive user profile construction device, specifically including the following steps 201 to 205.

[0051] Step 201: The integrated user profile building device obtains the first user's communication service order information, payment behavior information, and identity verification information from the first operator.

[0052] In some embodiments, the first user may be a new user of the first operator.

[0053] In some embodiments, a new user of the first operator refers to a user who has not previously engaged in any communication services with the first operator.

[0054] In some embodiments, the above-mentioned communication service order information may include: user address information.

[0055] In some embodiments, the payment behavior information may include: payment time information.

[0056] In some embodiments, the payment time information may include: business selection time information and payment completion time information.

[0057] In some embodiments, the aforementioned service selection time information refers to the time taken from when a user enters the service package selection page to when they finally confirm the selection of the target communication service package.

[0058] In some embodiments, the aforementioned payment completion time information refers to the time taken from when a user confirms the selection of a target communication service package to when the payment operation is completed and a payment success notification is received.

[0059] In some embodiments, the authentication information may include an authentication image.

[0060] In some embodiments, the above-mentioned communication service order information may also include at least one of the following: service package information, user age information, and user gender information; the above-mentioned payment behavior information may also include at least one of the following: payment channel information, payment status information, and payment device and network environment information; the above-mentioned identity verification information may also include at least one of the following: ID photo, occupation information, accessibility needs information, and family structure information.

[0061] Step 202: The comprehensive user profile construction device constructs a user order profile based on communication service order information.

[0062] In some embodiments, the user order profile described above represents the communication needs preferences and business attributes of the first user who has completed transactions.

[0063] In some embodiments, the above-mentioned communication demand preferences characterize a user's preferred choices of service parameters such as communication service package type, data usage duration allocation, value-added service content, and network quality standards.

[0064] In some embodiments, the aforementioned communication demand preferences may include at least one of the following: voice duration requirements and data traffic requirements. Of course, the aforementioned communication demand preferences may also include other communication demand preferences, which can be determined according to actual needs, and this application does not limit them.

[0065] In some embodiments, the aforementioned business attributes may include at least one of the following: business consumption level and business processing time. Of course, other business attributes may also be included, which can be determined according to actual needs, and this application does not limit this.

[0066] In some embodiments, the comprehensive user profile building apparatus can determine address value tags based on communication service order information and build a user order profile based on the address value tags.

[0067] It should be noted that the specific implementation process of the comprehensive user profile construction device, which determines address value tags based on communication service order information and constructs user order profiles based on address value tags, can be found in the relevant descriptions in the following embodiments. To avoid repetition, this application will not elaborate further here.

[0068] In some embodiments, the aforementioned communication service order information includes: user address information. For example, in conjunction with... Figure 2 ,like Figure 3 As shown, step 202 above can be implemented through steps 202a to 202c.

[0069] Step 202a: The integrated user profile construction device uses a natural language processing algorithm to parse user address information and obtain address parsing results.

[0070] In some embodiments, the above-mentioned natural language processing algorithm refers to a set of technologies that use computer technology to analyze, understand and process human natural language. Through natural language processing algorithms, structured parsing and semantic mining of unstructured text information can be achieved.

[0071] In some embodiments, the address resolution result represents the actual geographical location corresponding to the user address information.

[0072] In some embodiments, the comprehensive user profile building device can parse user address information using natural language processing algorithms to obtain the aforementioned address parsing results. Specifically, the electronic device first uses natural language processing algorithms to segment and recognize entities in the address text, accurately separating structured elements such as city, region, and park. Then, it processes ambiguous expressions through semantic disambiguation and finally combines an address standard library to complete format verification and completion, outputting address parsing results containing detailed geographical location attributes, building type, floor inference, and other information.

[0073] Step 202b: The integrated user profile construction device determines the address value tag based on the address resolution results and the preset knowledge graph.

[0074] In some embodiments, the aforementioned preset knowledge graph refers to a pre-constructed structured knowledge base that covers the geographical location, business district attributes, industrial distribution, communication demand characteristics, and other related information of at least one geographical region.

[0075] In some embodiments, the address value tag represents the consumption potential corresponding to the actual geographical location.

[0076] In some embodiments, the address value label includes at least one of the following: a high-value consumption area label, a medium-value consumption area label, and a low-value consumption area label. Of course, the address value label may also include other address value labels, which can be determined according to actual needs, and this application does not limit this.

[0077] Step 202c: The comprehensive user profile building device constructs a user order profile based on address value tags.

[0078] In some embodiments, the comprehensive user profile building device can construct a user order profile based on address value tags through the following process: First, the address resolution results are associated with a preset knowledge graph to generate address value tags; then, these address value tags are integrated with basic order features such as service package type, bandwidth level, and value-added service requirements in the order, and order profile dimension scores are generated through rule mapping and weighted calculation. For example, when the high consumption potential tag is matched with a gigabit broadband package, the "high-end demand tendency" score is improved; finally, the score is transformed into an interpretable tag system, including high-end service adaptability, cost-effectiveness priority, and enterprise demand-driven, forming a complete user order profile to provide a basis for subsequent business recommendations and risk assessments.

[0079] In this way, by parsing user address information through natural language processing algorithms, the user's actual geographical location is obtained. Then, based on the user's actual geographical location and a pre-defined knowledge graph, the regional value of the user's actual geographical location is determined. Based on the regional value of the user's actual geographical location, an address value tag representing the consumption potential of the actual geographical location is determined. Based on the determined address value tag, a user order profile is constructed. Subsequently, based on this user order profile, a comprehensive user profile that can reflect user characteristics from multiple perspectives is obtained. Based on this comprehensive user profile, a target strategy corresponding to the user is generated. This avoids the problem of low accuracy in user profiles caused by building user profiles based solely on information actively entered by new users when registering an account, and improves the accuracy of the constructed user profile.

[0080] Step 203: The comprehensive user profile building device constructs a user payment profile based on payment behavior information.

[0081] In some embodiments, the comprehensive user profile building apparatus can determine user payment behavior tags based on payment behavior information, and then build a user payment profile based on the user payment behavior tags.

[0082] It should be noted that the specific implementation process of the comprehensive user profile building device, which determines user payment behavior tags based on payment behavior information and then builds a user payment profile based on these tags, can be found in the relevant descriptions in the following embodiments. To avoid repetition, this application will not elaborate further here.

[0083] In some embodiments, the aforementioned user payment profile represents the spending power and willingness to spend of the first user.

[0084] In some embodiments, the spending power of the first user represents the first user's ability to bear the price of communication services.

[0085] In some embodiments, the intensity of the first user's willingness to consume represents the degree of the first user's proactive demand for communication services or their willingness to accept value-added services.

[0086] In some embodiments, the payment behavior information includes payment time information, which includes service selection time information and payment completion time information. For example, in combination with... Figure 2 ,like Figure 4 As shown, step 203 above can be implemented through steps 203a and 203b.

[0087] Step 203a: The comprehensive user profile building device determines the user's payment behavior tags based on the business selection time information and the payment completion time information.

[0088] In some embodiments, the aforementioned user payment behavior tag characterizes the behavioral features of the first user when paying for a communication service order.

[0089] In some embodiments, the aforementioned behavioral characteristics include at least one of the following: the user's decision-making time for selecting a communication service package, the interval between order submission and payment completion, the frequency of repeatedly entering and exiting the order page before payment, payment method preference, and whether there is a cancellation after payment. Of course, the aforementioned behavioral characteristics may also include other behavioral characteristics, which can be determined according to actual needs, and this application does not limit them.

[0090] In some embodiments, the aforementioned user payment behavior tags include at least one of the following: rational decision-making tag, impulsive but cautious tag, highly hesitant tag, high credit and low risk tag, medium to low credit risk tag, price-sensitive tag, high consumption potential tag, and technology early adopter tag. Of course, the aforementioned user payment behavior tags may also include other types of user payment behavior tags, which can be determined according to actual needs, and this application does not limit this.

[0091] Step 203b: The comprehensive user profile construction device constructs a user payment profile based on user payment behavior tags.

[0092] In some embodiments, the comprehensive user profile building device can construct a user payment profile based on user payment behavior tags through the following process: First, the entire payment behavior data (such as payment tools, time intervals, success rates, cancellation records, device environment, etc.) is decomposed into multi-dimensional features, and accurate tags such as credit risk level, decision-making style type, price sensitivity, device scenario preference, and abnormal fraud signals are generated through a pre-trained artificial intelligence (AI) model; then, these tags are weighted, integrated, and cross-validated, and corresponding service strategy suggestions are associated, ultimately forming a complete user payment profile covering dimensions such as credit assessment, consumption tendency, risk warning, and service adaptation, directly supporting business decisions such as intelligent review, personalized recommendation, and fraud interception.

[0093] In this way, by acquiring payment behavior information including business selection time information and payment completion time information, payment behavior tags that characterize the behavior of the first user when paying for a communication service order are determined, and a user payment profile is constructed based on the determined payment behavior tags. Subsequently, based on the constructed user payment profile, a comprehensive user profile that can reflect user characteristics from multiple perspectives is obtained, and a target strategy corresponding to the user is generated based on this comprehensive user profile. This avoids the problem of low accuracy in user profiles caused by constructing user profiles based solely on information actively entered by new users when registering an account, and improves the accuracy of the constructed user profile.

[0094] Step 204: The integrated user profile construction device constructs a user identity profile based on the identity verification information.

[0095] In some embodiments, the aforementioned user profile represents the first user's identity attributes, accessibility service needs, and family structure.

[0096] In some embodiments, the authentication information described above includes: an authentication image. Exemplarily, in conjunction with... Figure 2 ,like Figure 5 As shown, step 204 above can be implemented through steps 204a to 204c.

[0097] Step 204a: The integrated user profile construction device uses an optical character recognition algorithm to identify the identity verification image and obtain consumption level information.

[0098] In some embodiments, the aforementioned optical character recognition algorithm refers to the technology of converting character information in printed text, handwritten text, or images into editable and storable digital text content through computer vision technology. Optical character recognition algorithms can efficiently process unstructured visual data such as ID photos, certificates, and on-site images.

[0099] Step 204b: The comprehensive user profile construction device determines user identity tags based on consumption level information.

[0100] In some embodiments, the aforementioned consumption level information may include the user's consumption level.

[0101] In some embodiments, the aforementioned user identity tag represents the social identity of the first user.

[0102] In some embodiments, the aforementioned user identity tags include at least one of the following: low-credit user tag, medium-credit user tag, high-credit user tag, accessible service user tag, high-value user tag, medium-value user tag, and low-value user tag. Of course, the aforementioned user identity tags may also include other user identity tags, which can be determined according to actual needs, and this application does not limit this.

[0103] Step 204c: The comprehensive user profile construction device constructs a user identity profile based on the user identity tags.

[0104] In some embodiments, the comprehensive user profile building device first extracts basic features such as identity attributes, occupational information, special needs, and family structure from identity verification materials using technologies such as optical character recognition algorithms, computer vision analysis, and knowledge graph association. Then, it maps these features into precise tags such as low-risk, high-credit users, special care service recipients, and users who consume collaboratively with family members. Next, it weights and integrates the tags according to the logical framework of "identity attributes - social credit - special needs - family characteristics" and cross-verifies the consistency of multi-dimensional information, thereby constructing the aforementioned user identity profile based on user identity tags.

[0105] In this way, by recognizing the user's identity verification image through an optical character recognition algorithm, the user's consumption level information is obtained. Based on the user's consumption level information, a user identity tag representing the social identity of the first user is determined. Based on the determined user identity tag, a user identity profile is constructed. Subsequently, a comprehensive user profile that can reflect the user's characteristics from multiple perspectives is obtained based on the user identity profile. Based on this comprehensive user profile, a target strategy corresponding to the user is generated. This avoids the problem of low accuracy in user profiles caused by constructing user profiles based solely on information actively entered by new users when registering an account, and improves the accuracy of the constructed user profile.

[0106] Step 205: The comprehensive user profile building device constructs a comprehensive user profile based on the user order profile, user payment profile, and user identity profile.

[0107] In some embodiments, the comprehensive user profile described above is used to generate a target strategy corresponding to the first user, and the target strategy includes at least one of the following: a security protection strategy and a business recommendation strategy.

[0108] In some embodiments, the above security protection strategy is used to determine the probability of abnormal user risk in real time, trigger control actions such as strengthening real-name verification, limiting prepaid amount, and blocking orders for high-risk users, and automatically execute a fast review process for medium and low-risk users.

[0109] In some embodiments, the above-mentioned business recommendation strategy is used to accurately recommend communication service packages and value-added services to users in different scenarios. For example, recommending high-value smart home kits to users with high consumption potential, pushing campus-exclusive nighttime data packages to university dormitory users, and recommending cloud storage services and international network acceleration services to users in industrial parks.

[0110] The comprehensive user profile construction method provided in this application obtains new users' communication service order information, payment behavior information, and identity verification information respectively to construct user order profiles, user payment profiles, and user identity profiles from multiple dimensions. The constructed profiles are then merged to obtain a comprehensive user profile that reflects user characteristics from multiple perspectives. Based on this comprehensive user profile, a target strategy corresponding to the user is generated. This avoids the problem of low accuracy in user profiles caused by constructing user profiles based solely on information actively entered by new users when registering an account, and improves the accuracy of the constructed user profiles.

[0111] The following specific embodiments illustrate the comprehensive user profile construction method of this application.

[0112] This application provides a method for generating user profiles for operators, aiming to achieve: deep fusion and feature extraction of multimodal data throughout the entire order process; real-time generation of accurate, multidimensional, and interpretable user profiles during the order placement process; and seamless embedding the profiles into the order processing flow to achieve automated intelligent decision-making based on the profiles, thereby improving risk control, operational efficiency, and user value, and realizing an end-to-end closed loop from raw data to business value.

[0113] Specifically, the operator user profile generation method provided in this application can be implemented through the following steps 1 to 5:

[0114] Step 1: Data access.

[0115] Specifically, in the operator user profile generation method provided in this application, after an order is created, the system accesses structured and unstructured data in real time. The structured data includes: package type (e.g., gigabit bundled package), payment method (e.g., balance payment, credit card payment), payment time sequence, and the address entered by the user. The unstructured data includes: photos of the front and back of the user's ID card uploaded by the user, a live video of the user holding their ID card, and photos of their employment certificate.

[0116] Step 2: Feature extraction.

[0117] Specifically, the operator user profile generation method provided in this application uses an optical character recognition module to extract birth date and address from identity verification images; extract company name and position information from employment certificates; analyze the background of on-site videos using a computer vision module to identify the office environment (such as workstations and computers); analyze clothing using a computer vision module to determine business casual style; and perform liveness detection and person comparison. A natural language processing module deeply analyzes user addresses, links them to a knowledge graph to obtain tags such as "high-tech park" and "high-income population cluster," and analyzes payment behavior sequences to calculate a "decision hesitation index."

[0118] Step 3: Multimodal fusion.

[0119] Specifically, in the operator user profile generation method provided in this application, the extracted feature vectors are aligned and fused through a cross-modal attention fusion network. For example, spatial consistency checks and fusion are performed on "user address", "work unit address", "order delivery address" and "workplace inferred from video background" to obtain more accurate "permanent residence" and "occupational stability" features.

[0120] Step 4: Deduction based on the profile.

[0121] Specifically, in the operator user profile generation method provided in this application, the fused feature vector is input into a pre-trained profile reasoning model (such as a multi-task deep learning model), and the model outputs profile labels in multiple dimensions in parallel, including: credit risk profile: low risk (based on: stable occupation, successful real-name verification, good payment record), consumption capacity profile: high consumption potential (based on: address located in a high-value area, use of credit payment, occupation as a technician), family / need profile: possible home office / gaming needs (based on: selection of gigabit broadband, application during weekdays), and accessibility service needs profile (whether accessibility services are needed).

[0122] Step 5: Apply the output.

[0123] Specifically, the operator user profile generation method provided in this application outputs the generated profile to downstream systems in real time via an application interface. These downstream systems include an approval system, a scheduling system, and a marketing system. The approval system automatically and quickly approves profiles based on a "low-risk" label. The scheduling system prioritizes experienced engineers based on "high-value" and "potential home office" labels, scheduling appointments at times convenient for the user (e.g., evening). The marketing system recommends matching value-added products such as "high-end routers" and "international acceleration packages" in real time on the order confirmation page, based on the user profile.

[0124] The method for generating operator user profiles provided in this application will be described in detail below through specific embodiments.

[0125] Category 1: User order profile generation.

[0126] Step 1: Deep AI analysis of the address.

[0127] Specifically, the integrated user profiling device can obtain new users' order information and personal information actively entered by them. This includes four data dimensions. The first dimension is address information, specifically the complete delivery text address. AI can extract value from this, including obtaining geographical coordinates (latitude and longitude), address hierarchy (province, city, district, street, community, building, unit), distinguishing address types (residential buildings, office buildings, schools, hospitals, shops), and extracting semantic keywords such as "international community," "entrepreneurial park," and "university town." The second dimension is package information, specifically the name and tier of the main package selected by the user. AI can extract information from this to differentiate price sensitivity (low, medium, and high tiers) and identify high-volume transactions. The first data dimension is the user's demand for high-speed broadband and bundled packages, as well as the value in determining whether the user is in a new user acquisition phase. The third data dimension is identity information, specifically the user's age and gender after real-name authentication. AI can extract demographic characteristics from this, and also identify generational labels such as Generation Z, Millennials, and Senior Citizens. The fourth data dimension is time information, specifically the user's order time and scheduled installation time. AI can extract time-of-day, weekend, and late-night characteristics from this, and also identify seasonal factors such as the start of the school year, year-end, and promotional seasons. Through this process, intelligent address resolution and geographic value mining can be achieved.

[0128] For example, assuming the address information is "Building E, Floor F, District C, City A, District B, Park C, Community D", the following parsing result can be obtained by parsing the address information using natural language processing technology: Parsing result = {"City":"A","District":"B","Park":"C", "Community":"D", "Building Number":"E", "Building Type":"Office Building", "Floor": "F" floor}. Based on this analysis result, an external knowledge graph is linked to obtain the knowledge graph query result: {"Regional Characteristics":{"Average Housing Price":"120,000 RMB per square meter","Enterprise Density":"Extremely High","Population Characteristics":"Technology Practitioners, High Income, Young","Park Characteristics": "Enterprise Type": "Internet Giants","Network Demand":"Extremely High Bandwidth, Low Latency","Competitive Landscape": "Full Coverage by Operators, Intense Competition"}. Then, based on this knowledge graph query result, an "Address Value Profile" can be generated: Address Profile = {"Value Level": "High-Value Area","Core Demand":"Enterprise-Grade Broadband","International Acceleration","Cloud Services","Consumption Potential": "Extremely High","Competitive Sensitivity": "High (Requires Differentiated Services)"}.

[0129] Step 2: How a real-time recommendation engine works.

[0130] Specifically, the comprehensive user profile building device can accurately recommend services to target users based on real-time recommendation decision tree logic. For example, assuming the address type is a high-end residential area and the package tier is gigabit broadband, the comprehensive user profile building device can recommend high-end services such as Fiber to the Room (FTTR), smart home security packages, and high-definition video services. Alternatively, assuming the package tier is 500M broadband, the comprehensive user profile building device can recommend mid-range services such as upgrading to a gigabit experience package or smart door lock services. Or, assuming the address type is a university dormitory, the comprehensive user profile building device can recommend low-end services such as campus-exclusive nighttime data packages and video memberships. Or, assuming the address type is an industrial park or office building, and the order time is weekdays from 9:00 AM to 6:00 PM, the comprehensive user profile building device can recommend commonly used enterprise services such as enterprise cloud storage business edition, international network acceleration, and multi-device simultaneous online packages.

[0131] Step 3: The technical architecture of the recommendation algorithm.

[0132] Specifically, the comprehensive user profile building device can employ a three-layer hybrid recommendation strategy to accurately recommend services to users. These three layers include a rule layer, a collaborative filtering layer, and a deep learning layer. The rule layer ensures a basic conversion rate based on explicit business rules; for example, recommending 5G network acceleration services when subscribing to a 5G plan, or recommending international roaming packages when the address contains "international" or "overseas." The collaborative filtering layer identifies similar user groups. Similar user groups refer to users with a historical address feature similarity greater than 0.8, the same initial plan, and who ultimately purchased value-added service X. The recommendation strength is equal to the proportion of this group that purchased value-added service X. The deep learning layer outputs the click or purchase probability of each candidate value-added service based on the input address vector, plan vector, user attributes, and contextual features.

[0133] Category 2: User payment profile generation.

[0134] Specifically, the comprehensive user profile building device can acquire data from five different dimensions. The first data dimension is in-depth information on payment channels, specifically divided into three points: balance payment and credit card payment. AI can extract value from this information, reflecting users' payment habits, credit instrument usage preferences, and cash flow. The second data dimension is payment failure behavior, specifically including two parts: first, recording the reasons for payment failure, categorized into insufficient balance, incorrect password, and risk control interception; second, recording user behavior after payment failure, categorized into immediate retry, changing payment method, and abandoning payment. AI can extract value by identifying users' payment ability issues, operational proficiency, and purchase intention strength. The third data dimension is phased payment data, specifically recording the interval between the three phases. The first data dimension is the time from package selection to order submission, from order submission to payment initiation, and from payment initiation to payment completion. The value of AI in this dimension is to pinpoint the user's decision-making hesitation points, clarifying whether the user is struggling when selecting a package or hesitating during payment. The second data dimension is the device and network environment, which includes three points: the type of device used for payment (i.e., the phone model), the type of network used (Wi-Fi, 5G, 4G), and the geographical location information corresponding to the network address. AI can use this information to determine the user's digital consumption capacity and whether the user's current consumption scenario is at home or outside. The third data dimension is the historical payment success rate, which refers to the percentage of successful payments for the user's historical orders. The value of AI in this dimension is to assess the reliability of the user's payment.

[0135] Category 3: User Profile Generation.

[0136] Specifically, the comprehensive user profile building device can acquire data in four dimensions: facial features including age, gender, and emotional state; environmental background including shooting location and decoration style; clothing accessories including clothing brand, jewelry, and eyeglass type; and behavioral features including cooperation level and operational proficiency. The core technologies for this type of material are computer vision, scene recognition, and micro-expression analysis. The corresponding profile value is the analysis of the subject's real-time state, living environment, and consumption level. The second type of material is the front and back of an ID card. AI can extract information in three dimensions: structured information including name, gender, ethnicity, date of birth, and address; document authenticity information including anti-counterfeiting features, printing quality, and chip information; and address depth analysis information including issuing authority, address level, and regional characteristics. The corresponding core technologies are OCR, anti-counterfeiting point detection, and address NLP analysis. The corresponding profile value is to clarify basic identity, geographical origin, and document risk. The third type of material is occupational certificates. The information dimensions that AI can extract include occupational types such as civil servants, teachers, doctors, and corporate employees; the nature of the employer such as state-owned enterprises, private enterprises, and foreign-funded enterprises; and the job titles of ordinary employees and management. The first category is information on disability, including income range inferred from occupation and region. The core technologies for this are document classification, key information extraction, and industry knowledge. The corresponding profile value is the analysis of occupational stability, income level, and creditworthiness. The second category is disability certificates. AI can extract information on disability categories such as visual, hearing, physical, and intellectual disabilities, disability levels from one to four, and special needs related to communication assistance. The core technologies for this are seal recognition, table parsing, and classification models. The corresponding profile value is the clarification of special care needs and eligibility for policy benefits. The third category is income certificates for minors. AI can extract information on the guardian's occupation, family income level, and the authority of the issuing unit. The core technologies for this are handwriting / print recognition, number extraction, and unit credibility verification. The corresponding profile value is the analysis of family economic status and ability to fulfill obligations. The fourth category is relationship certificates for minors. AI can extract information on the guardianship relationship between parents and other relatives, the family structure (single-parent / double-parent / other), and household registration information. The core technologies for this are relationship graph construction and document verification. The corresponding profile value is the analysis of family structure and guardianship stability.

[0137] This solution leverages various AI technologies to perform multi-dimensional user profiling analysis. First, it analyzes occupational credentials. For user ID U12345, a public school teacher—a user with a high-stability occupation—the following characteristics are identified: the employing unit is a municipal key high school; based on the professional title, the estimated years of service are 8 to 12 years (intermediate professional title); based on the regional average, the estimated monthly income is 8,000 to 12,000 yuan; and the housing provident fund contribution base is high (standard for public institutions). In terms of credit assessment, this user's occupational stability score is 92 out of 100, which is considered high. The system offers extremely high stability, with an income credibility score of 88 out of 100. As it is certified by a public institution, its credibility is high. A credit limit of 20,000 yuan is recommended for high-credit purchases. Recommended products include 0 down payment installment plans with high postpaid limits and exclusive teacher discount packages. These packages are automatically matched to civil servants, teachers, and doctors, providing higher credit limits. Employees of large state-owned enterprises will be recommended to receive group customer packages with shared family benefits. Freelancers who cannot provide fixed proof may be required to provide supplementary materials or have their prepaid limits restricted. Next, a profile of special care needs based on disability certificates is created. The AI ​​disability certificate analysis process involves inputting a picture of the disability certificate, extracting the name, ID number, disability category, disability level, and issuing authority via OCR, then verifying the authenticity of the official seal of the Disabled Persons' Federation, and finally using a classification model to categorize the disability type into visual, auditory, physical, intellectual, and mental disabilities, ultimately outputting a care profile. For example, the output result might be: the disability category is hearing impairment, the disability level is level two, special communication needs include video calls requiring subtitles, text customer service requiring vibration alerts, policy matching allowing for communication fee reductions, dedicated customer service channels, and accessibility terminal discounts, and a reminder that the user needs to apply for a special disability package certification. The service suggestion is automatic allocation of sign language video customer service seats, corresponding to the application scenario of automatically applying for communication fee discounts that meet policy requirements, and the customer service system automatically identifying and prioritizing the allocation of seats with special skills such as sign language customer service, while also recommending accessibility terminal devices such as large-screen mobile phones and hearing aid compatible devices. Next, we'll look at family structure and lifecycle profiles based on relationship proof. Taking the construction of a family profile for a minor opening an account as an example, the resulting family profile might show the account holder as 16 years old, the guardian relationship as father and son, the family structure as not being a single-parent family with the father as guardian, the possibility of an only child based on age and regional policies, and the high level of family attention due to the father personally handling the application rather than entrusting someone else. The predicted consumption characteristics are primarily learning resources, parental control, and moderate entertainment. Risk characteristics include game addiction, live-streaming rewards, and nighttime use. Payment relies entirely on the guardian. The recommended package strategy prioritizes student packages with learning resources (mandatory), parental control features (optional), and anti-addiction reminders. Corresponding application scenarios include automatically adding green internet learning resource packages for student users, providing family network discounts for single-parent families, focusing on psychological support services, and recommending family-sharing packages for families with multiple children to maximize family consumption.Next, we create a profile of the living environment and consumption level based on the background of the on-site photos. Using computer vision, we perform scene analysis on the uploaded photos and videos. If background elements include mahogany furniture, pianos, high-end appliances, a high greening rate in the community with separate pedestrian and vehicle access and children's playground facilities, and clothing with brand names, smartwatches, and designer glasses, the consumption level is determined to be high-end, and the recommendation strategy is to recommend gigabit FTTR smart home packages and international services. If background elements include simple decoration and basic furniture, characteristics of an urban village (dense buildings and narrow alleys), and ordinary clothing without obvious brands, the consumption level is determined to be value-for-money, and the recommendation strategy is to recommend affordable data packages, data allowances, and discounted second-hand devices. If the background shows a shop front desk with a visible business license, as well as work uniforms and name tags, the user type is determined to be a small business owner, and the recommendation strategy is to recommend enterprise broadband cloud storage and shop promotion services. Finally, behavioral risk and fraud identification profiling is performed. Based on multi-material consistency verification, a verification matrix is ​​obtained through AI multimodal cross-validation. Consistency checks include: comparing the face on the ID card with the live face at the scene, achieving a 99.2% facial similarity; comparing the ID card age with the visually estimated age, showing an age difference of +1 year, which is within a reasonable range; comparing the ID card address with the scene's photographed environment, indicating a low environment match (not at home); comparing the employer's certificate with the scene's photographed environment, indicating a matching office environment; and comparing the declared purpose with the user's historical behavior patterns, indicating high pattern consistency. One risk signal is a mismatch between the environment address and the employer's certificate, but a matching employer's certificate. The interpretation is that it is reasonable for users to handle the matter at the office location, and the risk level is low. Another signal is that there are signs of multiple people guiding the process during the liveness detection, which is interpreted as possibly indicating non-independent decision-making. It is necessary to pay attention to whether it is being sold to, and the risk level is medium. The final risk score is 23 out of 100, which is low risk. At the same time, high-risk patterns can be identified, including: overly perfect materials, that is, all certificates are brand new and have no signs of use, which may be specially prepared fraudulent materials; abnormal emotional state, that is, appearing nervous and frequently looking at others during the process, which may be manipulating the account opening; and multiple repeated patterns, that is, different users using the same background and the same guide, which may be an abnormal card application team.Different materials correspond to different information extraction dimensions, core technologies, and profiling value. For on-site photos or live videos, extractable information dimensions include facial features (age, gender, emotional state), environmental background (shooting location, store, outdoor decoration style), clothing and accessories (clothing brand, jewelry, type of glasses), and behavioral characteristics (cooperation and operational proficiency). The core technologies are computer vision scene recognition and micro-expression analysis. Profiling value lies in obtaining real-time status, living environment, and consumption level. For the front and back of ID cards, structured information such as name, gender, ethnicity, date of birth, and address can be extracted. Anti-counterfeiting features (printing quality, chip information) can be used to verify the authenticity of the document. Deep address analysis can be performed to obtain the issuing authority's address level and regional characteristics. The core technologies are OCR anti-counterfeiting point detection and address NLP analysis. Profiling value lies in obtaining basic identity information, geographical origin, and document risk. For occupational certificates, occupational types such as civil servant, teacher, doctor, and corporate employee can be extracted, along with the type of employer (government, state-owned enterprise, private enterprise, etc.). For enterprises, this system provides information on job levels, from general employees to management, inferring income ranges based on occupation and region. Its core technologies include document classification, key information extraction, and industry knowledge extraction. The profile value lies in assessing job stability, income level, and creditworthiness. For disability certificates, it can extract disability categories (visual, hearing, physical, intellectual disabilities, etc.), disability levels (one to four), and communication assistance needs. Its core technologies include seal recognition, table parsing, and classification models. The profile value lies in clarifying special care needs and eligibility for preferential policies. For income certificates for minors, it can extract the guardian's occupation, family income level, and the authority of the proof unit. Its core technologies include handwritten and printed identification, digital extraction, and unit credibility verification. The profile value lies in assessing the family's economic situation and ability to fulfill obligations. For relationship certificates for minors, it can extract guardian relationships (parents or other relatives, family structure, single-parent, double-parent, other household registration information). Its core technologies include relationship graph construction and document verification. The profile value lies in clarifying family structure and guardianship stability.

[0138] This application implements the above-mentioned operator user profile generation method through five layers: application layer, profile reasoning layer, multimodal fusion layer, feature extraction layer, and data access layer. The following is a detailed description of the multiple layers involved in this application with reference to the accompanying drawings.

[0139] For example, such as Figure 6 As shown, in the application layer, data is passed from bottom to top. At the bottom, profile tags and decision suggestions are passed in and sent upwards to the unified profile API service gateway. Then, the unified profile API service gateway provides support for five business scenarios in the application layer: the business module responsible for order review and risk decision-making, the business module responsible for intelligent order dispatch and resource scheduling, the business module responsible for real-time marketing and accurate recommendation, the business module responsible for customer service assistant and agent empowerment, and the business module responsible for business analysis and report insight.

[0140] For example, such as Figure 7As shown, this hierarchy is a bottom-up progressive structure: the bottom layer input is the fused feature vector, which is passed up to the "Portrait Interpretability Engine"; the Portrait Interpretability Engine has the ability to generate evidence chains, and provides two types of evidence chain examples: the evidence chain of high consumption potential, which is formed by splicing together the address of high-end communities, credit card payments, and professional information of technical personnel; and the evidence chain of special care needs, which is formed by splicing together the certificate of second-level hearing disability and the address of barrier-free facilities; after processing, the Portrait Interpretability Engine outputs the results to the upstream multi-task portrait reasoning engine, supporting the four types of portrait models under the multi-task portrait reasoning engine, which are, from left to right, the credit risk portrait model, the consumption capacity portrait model, the family structure portrait model, and the special needs portrait model.

[0141] For example, such as Figure 8 As shown, this layer is divided into two parts. The upper layer uses a cross-modal attention fusion network for processing. First, it inputs features from four different modalities: text features, image features, video features, and behavioral features. After input, attention weights are assigned and features are aligned in sequence to obtain a fusion vector. After feature fusion, the feature vector is output and passed to the multi-dimensional consistency verification module below for verification. This module contains four verification items: address consistency verification (verifying the consistency between the place of residence, delivery location, and workplace); identity verification (verifying the consistency between the ID card and the live person); behavioral logic verification (verifying the logical fit between the package price, occupational characteristics, and age); and relationship authenticity verification (verifying the authenticity of the relationship between the guardian and the minor).

[0142] For example, such as Figure 9 As shown, this hierarchy divides feature extraction into four different extraction modules, arranged from left to right: CV visual feature extraction, OCR text feature extraction, NLP semantic feature extraction, and behavioral feature extraction. CV visual feature extraction includes facial features, liveness detection, background scene recognition, clothing analysis, and environmental attribute inference. OCR text feature extraction includes ID card name and number, document address, issuing authority, disability category and level recognition, and seal authenticity detection. NLP semantic feature extraction includes address semantic parsing and standardization, occupation and employer entity extraction, and relationship proof kinship analysis. Behavioral feature extraction includes payment time series features, payment method preference features, decision hesitation index, and payment failure retry mode. Below these four modules, an external knowledge graph association module is also provided. This module contains three parts: address value profile, covering average neighborhood price, business district level, school district attributes, and building location; occupational stability score, covering employer type, industry prosperity, and job level; and network resource coverage, covering gigabit connectivity, FTTR support, and port capacity.

[0143] For example, such as Figure 10 As shown, this layer includes a multi-source heterogeneous data adapter, which can adapt to and process six major categories of business data: orders, real-name authentication, payment, customer service, logistics, and resources. Below, arranged horizontally, are data streams from five source business systems, from left to right: the order system data stream, containing package information, product attributes, order amount, delivery address, acceptance channel, and appointment time; the real-name authentication data stream, containing photos of the front and back of an ID card, on-site liveness video, proof of occupation, proof of disability, and proof of relationship; the payment system data stream, containing payment method, payment timestamp, reason for payment failure, installment information, and account binding information; the customer service system data stream, containing follow-up dialogue records, consultation intent, complaint records, satisfaction feedback, and order follow-up records; and the logistics dispatch system data stream, containing arrival time, engineer schedule, material inventory, and coverage area. Below these five data streams is a real-time data pipeline. The processing flow of this pipeline starts from order creation or status change, sequentially undergoing real-time capture, data cleaning, and finally format unification, completing the access processing of multi-source heterogeneous data.

[0144] It should be noted that the above-described method embodiments, or the various possible implementations of the method embodiments, can be executed individually, or, provided there is no conflict, they can be combined with each other. The specific implementation can be determined according to actual usage requirements, and this application embodiment does not impose any restrictions on this.

[0145] As can be seen, the above mainly describes the solutions provided by the embodiments of this application from a methodological perspective. To achieve the above functions, the embodiments of this application provide corresponding hardware structures and / or software modules for executing each function. Those skilled in the art should readily recognize that, in conjunction with the modules and algorithm steps of the various examples described in the embodiments disclosed herein, the embodiments of this application can be implemented in hardware or a combination of hardware and computer software. Whether a function is executed in hardware or by computer software driving hardware depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0146] This application embodiment can divide the comprehensive user profile construction device into functional modules according to the above method example. For example, each function can be divided into its own functional module, or two or more functions can be integrated into one processing module. The integrated module can be implemented in hardware or as a software functional module. Optionally, the module division in this application embodiment is illustrative and only represents one logical functional division; other division methods may be used in actual implementation.

[0147] In some embodiments, this application also provides a comprehensive user profile construction apparatus. This comprehensive user profile construction apparatus may include one or more functional modules for implementing the comprehensive user profile construction method of the above method embodiments.

[0148] For example, Figure 11 This is a schematic diagram of a comprehensive user profile building device provided in an embodiment of this application. Figure 11 As shown, the comprehensive user profile building device 900 includes an acquisition module 901 and a processing module 902.

[0149] The acquisition module 901 is used to acquire the first user's communication service order information, payment behavior information, and identity verification information from the first operator; the processing module 902 is used to construct a user order profile based on the communication service order information acquired by the acquisition module 901, which represents the first user's communication needs preferences and the service attributes of the services already processed; and to construct a user payment profile based on the payment behavior information, which represents the first user's spending power and willingness to spend; and to construct a user identity profile based on the identity verification information, which represents the first user's identity attributes, accessibility service needs, and family structure; and to construct a comprehensive user profile based on the user order profile, the user payment profile, and the user identity profile.

[0150] One possible implementation is that the aforementioned communication service order information includes: user address information; the aforementioned processing module 902 is specifically used to: parse the user address information using a natural language processing algorithm to obtain an address parsing result, wherein the address parsing result represents the actual geographical location corresponding to the user address information; determine an address value tag based on the address parsing result and a preset knowledge graph, wherein the address value tag represents the consumption potential corresponding to the actual geographical location; and construct the user order profile based on the address value tag.

[0151] Another possible implementation is that the aforementioned payment behavior information includes: payment time information, which includes: service selection time information and payment completion time information; the aforementioned processing module 902 is specifically used to: determine a user payment behavior tag based on the aforementioned service selection time information and the aforementioned payment completion time information, the user payment behavior tag representing the behavioral characteristics of the aforementioned first user when paying for a communication service order; and construct the aforementioned user payment profile based on the aforementioned user payment behavior tag.

[0152] Another possible implementation is that the processing module is specifically used to: identify the aforementioned identity verification image using an optical character recognition algorithm to obtain consumption level information; determine a user identity tag based on the aforementioned consumption level information, the user identity tag representing the social identity of the aforementioned first user; and construct the aforementioned user identity profile based on the aforementioned user identity tag.

[0153] Another possible implementation is that the aforementioned communication service order information also includes at least one of the following: service package information, user age information, and user gender information; the aforementioned payment behavior information also includes at least one of the following: payment channel information, payment status information, and payment device and network environment information; the aforementioned identity verification information also includes at least one of the following: ID photo, occupation information, accessibility needs information, and family structure information.

[0154] Another possible implementation is that the aforementioned comprehensive user profile is used to generate the target strategy corresponding to the first user, and the target strategy includes at least one of the following: security protection strategy and business recommendation strategy.

[0155] The comprehensive user profile building device provided in this application acquires new users' communication service order information, payment behavior information, and identity verification information respectively to build user order profiles, user payment profiles, and user identity profiles from multiple dimensions. The multiple profiles are then merged to obtain a comprehensive user profile that reflects user characteristics from multiple perspectives. Based on this comprehensive user profile, a target strategy corresponding to the user is generated. This avoids the problem of low accuracy in user profiles caused by building user profiles based solely on information actively entered by new users when registering an account, and improves the accuracy of the constructed user profiles.

[0156] It should be noted that the comprehensive user profile building device can implement all the processes implemented in the above method embodiments and achieve the same beneficial effects. To avoid repetition, it will not be described again here.

[0157] In the case where the functions of the integrated modules described above are implemented in hardware, this application provides a possible structural schematic diagram of the electronic device involved in the above embodiments. For example... Figure 12 As shown, the electronic device 90 includes: a processor 92, a communication interface 93, and a bus 94. Optionally, the electronic device 90 may also include a memory 91.

[0158] Processor 92 may implement or execute various exemplary logic blocks, modules, and circuits described in conjunction with the disclosure of this application. Processor 92 may be a central processing unit, a general-purpose processor, a digital signal processor, an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. It may implement or execute various exemplary logic blocks, modules, and circuits described in conjunction with the disclosure of this application. Processor 92 may also be a combination that implements computational functions, such as including one or more microprocessor combinations, a combination of a DSP and a microprocessor, etc.

[0159] Communication interface 93 is used to connect with other devices via a communication network. This communication network can be Ethernet, wireless access network, wireless local area network (WLAN), etc.

[0160] The memory 91 may be a read-only memory (ROM) or other type of static storage device capable of storing static information and instructions, random access memory (RAM) or other type of dynamic storage device capable of storing information and instructions, or electrically erasable programmable read-only memory (EEPROM), disk storage media or other magnetic storage devices, or any other medium capable of carrying or storing desired program code in the form of instructions or data structures and accessible by a computer, but is not limited thereto.

[0161] As one possible implementation, the memory 91 can exist independently of the processor 92. The memory 91 can be connected to the processor 92 via a bus 94 and is used to store instructions or program code. When the processor 92 calls and executes the instructions or program code stored in the memory 91, it can implement the comprehensive user profile construction method provided in this application embodiment.

[0162] In another possible implementation, memory 91 can also be integrated with processor 92.

[0163] Bus 94 can be an Extended Industry Standard Architecture (EISA) bus, etc. Bus 94 can be divided into address bus, data bus, control bus, etc. For ease of representation, Figure 12 The bus is represented by a single thick line, but this does not mean that there is only one bus or one type of bus.

[0164] Through the above description of the implementation methods, those skilled in the art can clearly understand that, for the sake of convenience and brevity, only the division of the above functional modules is used as an example. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the service calling device can be divided into different functional modules to complete all or part of the functions described above.

[0165] This application embodiment also provides a chip, which includes a processor and a communication interface. The communication interface is coupled to the processor. The processor is used to run programs or instructions to implement the various processes of the above-described comprehensive user profile construction method embodiment, and can achieve the same technical effect. To avoid repetition, it will not be described again here.

[0166] It should be understood that the chip mentioned in the embodiments of this application may also be referred to as a system-on-a-chip, system chip, chip system, or system-on-a-chip, etc.

[0167] This application also provides a readable storage medium storing a program or instructions. When executed by a computer, the program or instructions implement the comprehensive user profile construction method provided in the above embodiments. It is understood that all or part of the processes in the above method embodiments can be executed by computer instructions instructing related hardware. The readable storage medium can be any of the foregoing embodiments or memory. The readable storage medium can also be an external storage device of the service invocation device, such as a pluggable hard drive, SmartMedia Card (SMC), Secure Digital (SD) card, flash card, etc., equipped on the service invocation device. Further, the readable storage medium can include both internal storage units of the service invocation device and external storage devices. The readable storage medium is used to store the computer program and other programs and data required by the service invocation device. The readable storage medium can also be used to temporarily store data that has been output or will be output.

[0168] This application also provides a computer program product, which is stored in a storage medium and, when executed by a computer, implements the comprehensive user profile construction method provided in the above embodiments.

[0169] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element. Furthermore, it should be noted that the scope of the methods and apparatuses in the embodiments of this application is not limited to performing functions in the order shown or discussed, but may also include performing functions substantially simultaneously or in the reverse order, depending on the functions involved. For example, the described methods may be performed in a different order than described, and various steps may be added, omitted, or combined. Additionally, features described with reference to certain examples may be combined in other examples.

[0170] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a computer software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes several instructions to cause a terminal (which may be a mobile phone, computer, server, or network device, etc.) to execute the methods described in the various embodiments of this application.

[0171] The embodiments of this application have been described above with reference to the accompanying drawings. However, this application is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of this application without departing from the spirit and scope of the claims, and all of these forms are within the protection scope of this application.

Claims

1. A comprehensive user profile construction method, characterized in that, include: Obtain the first user's communication service order information, payment behavior information, and identity verification information from the first operator; Based on the communication service order information, a user order profile is constructed, which represents the communication needs preferences and service attributes of the first user. Based on the payment behavior information, a user payment profile is constructed, which represents the first user's spending power and willingness to spend. Based on the authentication information, a user identity profile is constructed, which represents the first user's identity attributes, accessibility service needs, and family structure. A comprehensive user profile is constructed based on the user order profile, the user payment profile, and the user identity profile.

2. The comprehensive user profile construction method according to claim 1, characterized in that, The communication service order information includes: user address information; the step of constructing a user order profile based on the communication service order information includes: The user address information is parsed using a natural language processing algorithm to obtain an address parsing result, which represents the actual geographical location corresponding to the user address information. Based on the address parsing results and the preset knowledge graph, an address value tag is determined, which represents the consumption potential corresponding to the actual geographical location. Based on the address value tags, the user order profile is constructed.

3. The comprehensive user profile construction method according to claim 1, characterized in that, The payment behavior information includes: payment time information, which includes: service selection time information and payment completion time information; the step of constructing a user payment profile based on the payment behavior information includes: Based on the service selection time information and the payment completion time information, a user payment behavior tag is determined, and the user payment behavior tag characterizes the behavioral characteristics of the first user when paying for a communication service order. Based on the user's payment behavior tags, the user payment profile is constructed.

4. The comprehensive user profile construction method according to claim 1, characterized in that, The authentication information includes: an authentication image; the step of constructing a user identity profile based on the authentication information includes: The consumption level information is obtained by recognizing the authentication image using an optical character recognition algorithm. Based on the consumption level information, a user identity tag is determined, and the user identity tag represents the social identity of the first user; Based on the user identity tags, the user identity profile is constructed.

5. The comprehensive user profile construction method according to claims 1 to 4, wherein the communication service order information further includes at least one of the following: service package information, user age information, and user gender information; the payment behavior information further includes at least one of the following: payment channel information, payment status information, and payment device and network environment information; and the identity verification information further includes at least one of the following: ID photo, occupation information, accessibility requirements information, and family structure information.

6. The method according to claim 1, characterized in that, The comprehensive user profile is used to generate the target strategy corresponding to the first user, and the target strategy includes at least one of the following: security protection strategy and business recommendation strategy.

7. A comprehensive user profile construction device, characterized in that, include: Acquisition module and processing module; The acquisition module is used to acquire the first user's communication service order information, payment behavior information, and identity verification information from the first operator; The processing module is used to construct a user order profile based on the communication service order information obtained by the acquisition module. The user order profile represents the communication demand preferences and service attributes of the first user. as well as, Based on the payment behavior information, a user payment profile is constructed, which represents the first user's spending power and willingness to spend. as well as, Based on the authentication information, a user identity profile is constructed, which represents the first user's identity attributes, accessibility service needs, and family structure. as well as, Based on the user order profile, the user payment profile, and the user identity profile, a comprehensive user profile is constructed. The comprehensive user profile is used to generate a target strategy corresponding to the first user. The target strategy includes at least one of the following: a security protection strategy and a business recommendation strategy.

8. The comprehensive user profile construction device according to claim 7, characterized in that, The communication service order information includes: user address information; the processing module is specifically used for: The user address information is parsed using a natural language processing algorithm to obtain an address parsing result, which represents the actual geographical location corresponding to the user address information. Based on the address parsing results and the preset knowledge graph, an address value tag is determined, which represents the consumption potential corresponding to the actual geographical location. Based on the address value tags, the user order profile is constructed.

9. The comprehensive user profile construction device according to claim 7, characterized in that, The payment behavior information includes: payment time information, which includes: service selection time information and payment completion time information; the processing module is specifically used for: Based on the service selection time information and the payment completion time information, a user payment behavior tag is determined, and the user payment behavior tag characterizes the behavioral characteristics of the first user when paying for a communication service order. Based on the user's payment behavior tags, the user payment profile is constructed.

10. The comprehensive user profile construction device according to claim 7, characterized in that, The processing module is specifically used for: The consumption level information is obtained by recognizing the authentication image using an optical character recognition algorithm. Based on the consumption level information, a user identity tag is determined, and the user identity tag represents the social identity of the first user; Based on the user identity tags, the user identity profile is constructed.

11. The comprehensive user profile construction device according to claims 7 to 10, wherein the communication service order information further includes at least one of the following: service package information, user age information, and user gender information; the payment behavior information further includes at least one of the following: payment channel information, payment status information, and payment device and network environment information; and the identity verification information further includes at least one of the following: ID photo, occupation information, accessibility needs information, and family structure information.

12. The comprehensive user profile construction device according to claim 7, characterized in that, The comprehensive user profile is used to generate the target strategy corresponding to the first user, and the target strategy includes at least one of the following: security protection strategy and business recommendation strategy.

13. An electronic device, characterized in that, It includes a processor and a memory, the memory storing programs or instructions that can run on the processor, the programs or instructions being executed by the processor to implement the comprehensive user profile construction method as described in any one of claims 1 to 6.

14. A readable storage medium, characterized in that, The readable storage medium stores a program or instructions, which, when executed by a computer, implement the comprehensive user profile construction method as described in any one of claims 1 to 6.

15. A computer program product, characterized in that, The computer program product is stored in a storage medium, and when executed by a computer, the computer program product implements the comprehensive user profile construction method as described in any one of claims 1 to 6.

Images