Data processing method and related equipment

Abstract

The embodiment of the invention provides a data processing method and related equipment. The method comprises the steps of determining a voucher application list in response to a triggering operation of a target object on a voucher application interface; acquiring behavior sequence data of the instant messaging object and bound first voucher information; acquiring second voucher information held by the object; fusing and processing the behavior sequence data of the instant messaging object, the first voucher information and the second voucher information to obtain voucher association feature information of the instant messaging object; determining similar objects of the target object according to the voucher association feature information of the instant messaging object and the target object; according to the voucher association feature information of the similar object, obtaining a simulation checking index of the similar object to the target voucher; and obtaining an approval index of the target object to the target voucher according to the simulated approval index of the similar object to the target voucher.

Description

Technical Field

[0001] The present disclosure relates to the field of computer and communication technology, and in particular to a data processing method, a data processing device, a computer equipment, a computer-readable storage medium, and a computer program product. Background Art

[0002] With the development of mobile Internet technology, more and more credentials can be applied for online through applications or web pages installed on the terminal, or through mini-programs in the application. Users no longer need to go offline to a distant physical store or counter to apply, which simplifies the difficulty and process of users applying for credentials and provides great convenience for users.

[0003] In the related art, in one case, through an application, webpage or applet, users can only apply for a single certificate, or a certificate issued by a single institution, such as a certificate of a certain institution or a joint certificate of a certain institution. Due to the lack of diverse choices in this method, users can only choose whether to apply, and there is only one approval result. When the risk control strategy of the institution is slightly tightened, the approval rate of the institution will be greatly reduced, and the user experience is poor at this time. In another case, through an application, webpage or applet, users can choose from a variety of certificates. For platforms operating multiple certificates, the disadvantage is that the scenario is single, and there is only the data of the platform itself, which cannot accurately identify whether the user is the target recommendation object of each certificate. Therefore, the approval rate is not high, and the approval accuracy is low. At the same time, in reality, the approval rate of various institutions is not high. Users often find it difficult to choose the corresponding certificate to apply for due to the uncertainty of the institution's approval, which causes users to waste a lot of time comparing multiple certificates when choosing. After the user is rejected by the institution's approval, the user needs to spend time again to select and apply for the certificate. Summary of the invention

[0004] The embodiments of the present disclosure provide a data processing method, a data processing apparatus, a computer device, a computer-readable storage medium, and a computer program product, which can be used to improve the accuracy and efficiency of credential application.

[0005] The present disclosure provides a data processing method, which includes: in response to a trigger operation of a target object on a credential application interface, determining a credential application list for the target object, the credential application list including a target credential issued by a target organization; acquiring, from an instant messaging service end, behavior sequence data generated by an instant messaging object during use of an instant messaging client and first credential information bound in the instant messaging client, the instant messaging object including the target object; acquiring, from a credential application service end connected to the credential application interface, second credential information held by an object, the object including the target object; fusing and processing the behavior sequence data, the first credential information and the second credential information of the instant messaging object to obtain credential-related feature information of the instant messaging object; obtaining a matching degree between the instant messaging object and the target object according to the credential-related feature information of the instant messaging object and the credential-related feature information of the target object, and determining a similar object of the target object from the instant messaging object according to the matching degree; obtaining a simulated approval index of the similar object for the target credential according to the credential-related feature information of the similar object; and obtaining an approval pass index of the target object for the target credential according to the simulated approval index of the similar object for the target credential.

[0006] The disclosed embodiment provides a data processing method, including: in response to a trigger operation of a target object on a credential application interface, displaying a credential application list of the target object and an approval index of the target object on the credential application interface, wherein the credential application list includes target credentials issued by a target institution; in response to a trigger operation of the target object on the target credential, displaying an application page of the target credential. The approval index is determined based on a simulated approval index of a similar object of the target object on the target credential.

[0007] The disclosed embodiment provides a data processing device, which includes: a processing unit, which is used to determine a credential application list for the target object in response to a trigger operation of a target object on a credential application interface, wherein the credential application list includes a target credential issued by a target institution; a first receiving unit, which is used to obtain, from an instant messaging service end, behavior sequence data generated by an instant messaging object during use of an instant messaging client and first credential information bound to the instant messaging client, wherein the instant messaging object includes the target object; a second receiving unit, which is used to obtain second credential information held by an object from a credential application service end connected to the credential application interface, wherein the object includes the target object; and the processing unit is further used to merge and process the instant messaging object. The processing unit is further used to obtain the credential-associated characteristic information of the instant messaging object based on the behavior sequence data of the object, the first credential information and the second credential information; the processing unit is further used to obtain the matching degree between the instant messaging object and the target object according to the credential-associated characteristic information of the instant messaging object and the credential-associated characteristic information of the target object, and determine the same type of object as the target object from the instant messaging object according to the matching degree; the processing unit is further used to obtain the simulation approval index of the same type of object for the target credential according to the credential-associated characteristic information of the same type of object; the processing unit is further used to obtain the approval pass index of the target object for the target credential according to the simulation approval index of the same type of object for the target credential.

[0008] The disclosed embodiment provides a data processing device, including: a response unit, for displaying the target object's credential application list and the target object's approval index for the target credential on the credential application interface in response to the target object's triggering operation on the credential application interface, wherein the credential application list includes the target credential issued by the target institution; the response unit is also used to display the target credential application page in response to the target object's triggering operation on the target credential. The approval index is determined based on the simulated approval index of the target credential by the target object of the same type.

[0009] An embodiment of the present disclosure provides a computer device, including a processor, a memory, and an input / output interface; the processor is connected to the memory and the input / output interface, respectively, wherein the input / output interface is used to receive and output data, the memory is used to store a computer program, and the processor is used to call the computer program so that the computer device including the processor executes the method in any embodiment of the present disclosure.

[0010] An embodiment of the present disclosure provides a computer-readable storage medium storing a computer program. The computer program is suitable for being loaded and executed by a processor so that a computer device having the processor executes a method in any embodiment of the present disclosure.

[0011] The embodiments of the present disclosure provide a computer program product or a computer program, which includes a computer instruction stored in a computer-readable storage medium. A processor of a computer device reads the computer instruction from the computer-readable storage medium, and the processor executes the computer instruction, so that the computer device executes the method provided in various optional modes in any embodiment of the present disclosure.

[0012] The disclosed embodiments provide a data processing method, a data processing device, a computer device, a computer-readable storage medium, and a computer program product. On the one hand, by responding to the trigger operation of the target object on the credential application interface, a credential application list including a target credential issued by a target organization is determined for the target object; then, the behavior sequence data generated by the instant messaging object during the use of the instant messaging client and the first credential information bound in the instant messaging client are obtained from the instant messaging server, and the second credential information held by the object is obtained from the credential application server. By integrating and processing the behavior sequence data, the first credential information, and the second credential information of the instant messaging object, the credential-related feature information of the instant messaging object obtained is enriched, and no longer solely depends on the second credential information of the credential application server. Therefore, when the credential-related feature information of the instant messaging object and the target object is used to obtain the matching degree between the instant messaging object and the target object, the same type of object of the target object can be more accurately determined from the instant messaging object according to the matching degree. On the other hand, based on the voucher-associated characteristic information of the same type of objects, the simulated approval index of the same type of objects for the target voucher is obtained; and then, based on the simulated approval index of the same type of objects for the target voucher, the approval pass index of the target object for the target voucher is obtained, which can assist the target object to select the voucher to apply for more accurately and efficiently from the voucher application list, reduce the selection difficulty of the target object, and improve the approval pass rate of the voucher after application, reduce the number of repeated operations of the target object to re-select the voucher to apply, and can more efficiently meet the target object's demand for vouchers. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 It is a flow chart of a data processing method provided by an embodiment of the present disclosure.

[0014] Figure 2 It is a flowchart of another data processing method provided by an embodiment of the present disclosure.

[0015] Figure 3It is a schematic diagram of generating a first feature extraction model and a second feature extraction model provided by an embodiment of the present disclosure.

[0016] Figure 4 The performance diagram of the second feature extraction model adopted in the embodiment of the present disclosure is exemplified.

[0017] Figure 5 It is a flowchart of another data processing method provided by an embodiment of the present disclosure.

[0018] Figure 6 The figure shows, by way of example, a label diagram output by the second feature extraction model used in the embodiment of the present disclosure.

[0019] Figure 7 It is a flowchart of another data processing method provided by an embodiment of the present disclosure.

[0020] Figure 8 The schematic diagram of the architecture for implementing the data processing method provided by the embodiment of the present disclosure is exemplified.

[0021] Fig. 9 It is a flowchart of another data processing method provided by an embodiment of the present disclosure.

[0022] Fig.10 It is a flowchart of another data processing method provided by an embodiment of the present disclosure.

[0023] Fig.11 It is an interactive schematic diagram of a data processing method provided by an embodiment of the present disclosure.

[0024] Fig.12 It is a schematic diagram of an application scenario of a data processing method provided by an embodiment of the present disclosure.

[0025] Fig.13 It is a flowchart of another data processing method provided by an embodiment of the present disclosure.

[0026] Fig.14 It is a schematic diagram of a data processing method provided by an embodiment of the present disclosure.

[0027] Fig.15 It is a schematic diagram of a credential application interface provided by an embodiment of the present disclosure.

[0028] Fig.16 It is a schematic diagram of another credential application interface provided by an embodiment of the present disclosure.

[0029] Fig.17 It is a schematic diagram of another credential application interface provided by an embodiment of the present disclosure.

[0030] Fig.18It is a schematic diagram of another credential application interface provided by an embodiment of the present disclosure.

[0031] Fig.19 It is a schematic diagram of another credential application interface provided by an embodiment of the present disclosure.

[0032] Fig. 20 It is a flowchart of another data processing method provided by an embodiment of the present disclosure.

[0033] Fig.21 It is a schematic diagram of a data processing device provided by an embodiment of the present disclosure.

[0034] Fig. 22 It is a schematic diagram of a data processing device provided by an embodiment of the present disclosure.

[0035] Fig.23 It is a structural diagram of a computer device provided in an embodiment of the present disclosure. DETAILED DESCRIPTION

[0036] The following will be combined with the drawings in the embodiments of the present disclosure to clearly and completely describe the technical solutions in the embodiments of the present disclosure. Obviously, the described embodiments are only part of the embodiments of the present disclosure, not all of the embodiments. Based on the embodiments in the present disclosure, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present disclosure.

[0037] In the embodiments of the present disclosure, the term "module" or "unit" refers to a computer program or a part of a computer program with a predetermined function, and works together with other related parts to achieve a predetermined goal, and can be implemented in whole or in part by using software, hardware (such as processing circuits or memories) or a combination thereof. Similarly, a processor (or multiple processors or memories) can be used to implement one or more modules or units. In addition, each module or unit can be part of an overall module or unit that includes the function of the module or unit.

[0038] The relevant data collection and processing in this disclosure should be strictly in accordance with the requirements of relevant national laws and regulations when applied in examples, and the informed consent or separate consent of the personal information subject should be obtained, and subsequent data use and processing should be carried out within the scope of authorization of laws and regulations and the personal information subject.

[0039] Each step in the method provided in any embodiment of the present disclosure may be performed by any computer device / electronic device, for example, any server and / or terminal, the server may include one or more of an instant messaging server, a credential application server, etc., and the terminal may be used to display a credential application interface and / or install an instant messaging client. However, the present disclosure is not limited thereto.

[0040] like Figure 1 As shown, the method provided by the embodiment of the present disclosure includes the following steps.

[0041] In S1, in response to a triggering operation of a target object on a credential application interface, a credential application list is determined for the target object, the credential application list including target credentials issued by a target institution.

[0042] In the embodiments of the present disclosure, the target object may be any object or user who applies for a voucher. The voucher in the embodiments of the present disclosure refers to an electronic voucher and / or physical voucher that can be used by an institution, an object applies to the institution, and the institution approves the application and then issues the voucher to the corresponding object. The function of the voucher may be set by the institution that issues the voucher, and the present disclosure does not limit this. For example, the voucher may be a savings card, a credit card, a transportation card, etc.

[0043] The voucher in the disclosed embodiment is a credit certificate issued by a commercial organization, voucher company or voucher platform to a credit-qualified object (consumer or user). No cash is required when using the voucher. Overdrafts can be made within the specified limit and repayments can be made on the billing date. The voucher is an electronic payment tool and a credit certificate.

[0044] The credential application interface in the embodiment of the present disclosure is a visual user interface displayed to an object (including a target object) for the object to apply for a credential on it. The credential application interface can be displayed or opened by an application installed on the terminal, or can be opened by a small program of an instant messaging client installed on the terminal, which is not limited by the present disclosure. The credential application interface can be a front-end interface of a credential platform, which can interact with a user.

[0045] In the disclosed embodiment, the voucher platform refers to a platform that cooperates with multiple institutions or organizations, allows users to freely apply for vouchers from different institutions, and provides services such as viewing application progress, voucher binding, voucher repayment, and voucher installment. It can also be called a voucher application platform. The voucher platform establishes an independent account for each user, recording which vouchers of which institutions the user has applied for, the approval results of each application (passed or not), credit limit, used limit, repayment record and other information. The voucher platform supports access methods through multiple channels, such as web pages, mobile applications, third-party payment platforms, etc.

[0046] In the embodiment of the present disclosure, the trigger operation of the target object on the credential application interface can be that the target object opens the credential application interface on the terminal, or performs a login operation on the credential application interface, or triggers the setting control on the credential application interface, which is not limited in the present disclosure. After the target object performs a trigger operation on the credential application interface, the trigger operation can be sent to the credential application server (such as the server of the credential platform) connected to the credential application interface. The trigger operation carries the target object information, such as any one or more of the user name, mobile phone number, ID number, login password, etc. entered by the target object in the credential application interface, as long as the identity information of the target object can be uniquely identified. After receiving the trigger operation, the credential application server responds to the trigger operation and determines a credential application list for the target object. In some embodiments, the credential application list can include all the credentials that the target object can apply for on the credential application platform, that is, the credential application server can perform a risk investigation on all the credentials provided by the credential application platform according to the target object information, and exclude the credentials that are determined to be unsuitable for the target object, and do not include them in the credential application list of the target object. In other embodiments, the credential application list includes all the credentials provided by the credential application platform.

[0047] In the disclosed embodiment, the target credential may be any credential in the credential application list. The target institution may be any company, organization or platform that has the authority to issue the target credential, such as a credential platform.

[0048] In S2, the behavior sequence data generated by the instant messaging object during the use of the instant messaging client and the first credential information bound in the instant messaging client are obtained from the instant messaging server, and the instant messaging object includes the target object.

[0049] Instant Messaging in the disclosed embodiment is a real-time communication system that allows two or more people to use the network to transmit text messages, files, voice and video in real time for communication. Users can also display personal information, life status, hobbies, etc. on it to share and communicate with friends. The real-time communication system includes an instant messaging client and an instant messaging server. The instant messaging client is the front end, which mainly provides an interface and interactive experience for users, supports the embedding of applets and web pages, and allows users to directly access various applications and services within the instant messaging client without installing additional applications, such as directly opening the credential application interface. The instant messaging server is the back end, which is responsible for processing a large amount of user data, generating behavioral sequence data of each instant messaging object, and efficiently storing and retrieving these data so that the required information can be quickly obtained. The instant messaging client and the instant messaging server can achieve efficient and real-time communication through instant messaging technology, including IP (Internet Protocol, Internet Interconnection Protocol) / TCP (Transmission Control Protocol, Transmission Control Protocol) / UDP (User Datagram Protocol, User Datagram Protocol) / Sockets (Network Programming Interface) and other technologies.

[0050] The instant messaging object in the disclosed embodiment refers to a user who uses an instant messaging client. Different instant messaging clients can implement instant messaging through an instant messaging server. In the process of using and operating an instant messaging client, an instant messaging object will generate a large amount of behavior data. These behavior data are sorted according to the generation time or storage time, and a large amount of behavior sequence data will be formed, such as the user's basic information (age, gender, region, etc.), the videos clicked, the comments made, the communication content with other users, the items purchased, the music listened to, the payment consumption behavior, the taxi-hailing behavior, etc.

[0051] The instant messaging client in the disclosed embodiment also supports or integrates an instant messaging wallet, which relies on the social payment platform of instant messaging, so that users can recharge cards issued by institutions or obtain electronic currency in other ways to make electronic payments. The instant messaging wallet can bind the credentials of the instant messaging object, and the first credential information refers to the relevant information of the credentials bound to the instant messaging object in the instant messaging wallet, such as which credentials issued by which institutions are bound. For example, the bound credential can be a card.

[0052] In the disclosed embodiment, the instant messaging object includes a target object, that is, the target object is a user of both the credential application platform and the instant messaging client.

[0053] In S3, second credential information held by an object including the target object is obtained from a credential application server connected to the credential application interface.

[0054] The object in the embodiment of the present disclosure refers to the user of the credential application platform. The credential application server stores which credentials each object has applied for on the credential application platform and which credentials have been approved by the corresponding institution, that is, which credentials the object holds, as the second credential information.

[0055] In S4, the behavior sequence data of the instant messaging object, the first credential information and the second credential information are integrated and processed to obtain the credential-related feature information of the instant messaging object.

[0056] In the disclosed embodiments, the credential-associated characteristic information of the instant messaging object refers to any information related to the credential application of the instant messaging object, for example, the information can be used to assist in determining the probability or possibility that the target object passes the application for the target credential, and / or the information is used to assist in determining whether the instant messaging object is a similar object to the target object when the target object applies for the target credential. The credential-associated characteristic information of the instant messaging object is a combination of the behavior sequence data of the instant messaging object, the first credential information bound in the instant messaging client, and the second credential information held by the instant messaging object on the credential application server. That is, after obtaining the second credential information held by the object, it is determined whether the object is also an instant messaging object. If it is an instant messaging object, the second credential information held by the instant messaging object on the credential application server is obtained.

[0057] In S5, the matching degree between the instant messaging object and the target object is obtained according to the credential-associated characteristic information of the instant messaging object and the credential-associated characteristic information of the target object, and the same type of object as the target object is determined from the instant messaging objects according to the matching degree.

[0058] In the disclosed embodiment, the matching degree of the target object and each instant messaging object is calculated based on the credential-associated feature information of the target object and each instant messaging object, respectively. The matching degree reflects the similarity of the credential-associated feature information of the target object and the instant messaging object. If the credential-associated feature information of the target object and a certain instant messaging object is more similar, the matching degree of the target object and the instant messaging object is greater; otherwise, the matching degree is smaller. Therefore, the same type of objects (referred to as the same type of users in the following embodiments) of the target object can be screened out from the instant messaging object by the size of the matching degree. For example, an instant messaging object whose matching degree is greater than a set matching degree threshold can be considered to be a same type of user of the target object, or, the matching degrees are arranged in descending order, and the instant messaging objects ranked in the front predetermined percentage can be considered to be the same type of users of the target object.

[0059] In S6, according to the voucher-associated characteristic information of the same type of object, the simulated approval index of the same type of object for the target voucher is obtained.

[0060] In an exemplary embodiment, based on the credential-associated characteristic information of the similar objects, a simulated approval index of the similar objects for the target credential is obtained, including: constructing the approval probability of the target institution for the credential-associated characteristic information; based on the approval probability of the target institution for the credential-associated characteristic information and the credential-associated characteristic information of the similar objects, a simulated approval index of the similar objects for the target credential is obtained.

[0061] In the embodiment of the present disclosure, constructing the approval probability of the target institution for the characteristic information associated with the credential means setting, for each target institution, the approval probability corresponding to the target credential when the characteristic information associated with the credential takes different values. For example, assuming that the target credential is the credential of Institution A, and the characteristic information associated with the credential includes age, it is set that when the age is in the first interval, the approval probability of the credential of Institution A is 0.1; when the age is in the second interval, the approval probability of the credential of Institution A is 0.3; when the age is in the third interval, the approval probability of the credential of Institution A is 0.2, ... Different target institutions can set different or the same approval probabilities for the same value of the characteristic information associated with the same credential, and the present disclosure does not limit this. The specific setting can be set according to the risk control strategy of each institution.

[0062] In the disclosed embodiment, after determining the same type of objects, the voucher-related characteristic information of the same type of objects is obtained, and based on the voucher-related characteristic information of the same type of objects and the approval probability of the target institution for the voucher-related characteristic information, the simulated approval index of each same type of object for the target voucher can be obtained. Among them, the simulated approval index reflects the estimated probability of the same type of object passing the approval of the target voucher based on the approval probability set by the target institution for the voucher-related characteristic information. Since the approval probability is in line with the risk control strategy set by the target institution, the simulated approval index is also in line with the risk control strategy of the target institution.

[0063] In S7, the approval index of the target object for the target voucher is obtained according to the simulated approval index of the similar object for the target voucher.

[0064] In an exemplary embodiment, the method provided by the embodiment of the present disclosure also includes: sending the credential application list and the target object's approval index for the target credential, so as to display the credential application list and the target object's approval index for the target credential for the target object on the credential application interface.

[0065] In the disclosed embodiment, since there is a high degree of similarity between the voucher-associated feature information of the same object and the target object, the target object's approval index for the target voucher can be predicted based on the simulated approval index of the same object for the target voucher, that is, the probability of the target object passing the approval of the target institution after applying for the target voucher. When the voucher application service returns the voucher application list to the voucher application interface, the approval index of each voucher (including the target voucher) in the voucher application list can also be returned to the voucher application interface, thereby assisting the target object in selecting which voucher to apply for from multiple vouchers. When the target object selects a voucher to apply for, the application information of the voucher is submitted to the corresponding institution for review. The application information of the voucher may include the target object information, such as the identity information, consumption information, credit information, overdue information, historical application information, historical approval results, etc. of the target object information.

[0066] The data processing method provided by the embodiment of the present disclosure, on the one hand, by responding to the trigger operation of the target object on the credential application interface, determines a credential application list including the target credential issued by the target organization for the target object; then obtains the behavior sequence data generated by the instant messaging object during the use of the instant messaging client and the first credential information bound in the instant messaging client from the instant messaging server, and obtains the second credential information held by the object from the credential application server. By integrating and processing the behavior sequence data, the first credential information and the second credential information of the instant messaging object, the credential-related feature information of the instant messaging object obtained is enriched, and no longer solely depends on the second credential information of the credential application server. Therefore, when the credential-related feature information of the instant messaging object and the target object is used to obtain the matching degree between the instant messaging object and the target object, the same type of object of the target object can be more accurately determined from the instant messaging object according to the matching degree. On the other hand, according to the voucher-related characteristic information of the same object, the simulated approval index of the same object for the target voucher is obtained; then, according to the simulated approval index of the same object for the target voucher, the approval index of the target object for the target voucher is obtained, which can assist the target object to more accurately and efficiently select the voucher to apply for from the voucher application list, reduce the selection difficulty of the target object, and improve the approval rate of the voucher after application, reduce the number of repeated operations of the target object to reselect the voucher for application, and more efficiently meet the target object's voucher application needs. As the number of applications initiated by users decreases, the number of network communications between the voucher application interface and the voucher application server can be reduced, thereby reducing the data processing pressure on the voucher application server and improving data processing efficiency.

[0067] exist Figure 2 In the embodiment, the above Figure 1Step S4 in the embodiment may further include the following steps.

[0068] In S41, the trained first feature extraction model is called to process the behavior sequence data of the instant messaging object to obtain the behavior sequence features of the instant messaging object.

[0069] The first feature extraction model in the disclosed embodiment refers to a machine learning model or a deep learning model for extracting features from the behavior sequence data of the instant messaging object obtained from the instant messaging service end, and the extracted behavior sequence features of the instant messaging object can be used to characterize the portrait of the instant messaging object. That is, by collecting and analyzing various behavior data of the WeChat instant messaging object in the process of using the instant messaging client, including the user's basic information, interests and hobbies, social relationships, behavioral habits, etc., a personalized portrait of the user is constructed to better meet user needs and provide high-quality services. For example, user data is presented in a labeled form, such as "urban white-collar women", "technology enthusiasts", "travel experts", etc. The portrait can include multiple aspects such as the user's basic information, consumption habits, interests and hobbies.

[0070] In some embodiments, in an advertising system (a system that can implement advertising distribution), a behavioral data source or a portrait data source of an instant messaging object can be imported from an instant messaging service end, that is, the behavioral sequence data of the instant messaging object is selected, and a data granularity of a long sequence of ten thousand is selected, that is, the length of the behavioral sequence data of an instant messaging object is ten thousand or more. By calling the above-mentioned first feature extraction model, the behavioral label data (i.e., the behavioral sequence features of the instant messaging object) within the preset time closest to the current time (the following examples are all within the past two years, but the present disclosure is not limited to this) can be generated. The behavioral label data can more completely and accurately draw out the attributes, behavioral patterns, preferences, and trends of the instant messaging object based on the behavioral sequence data of the instant messaging object, which helps to better understand user needs and scenario analysis.

[0071] In S42, an instant messaging object identifier is generated for the instant messaging object according to the real-name authentication information of the instant messaging object in the instant messaging client.

[0072] In the disclosed embodiment, the real-name authentication information (e.g., the user's real-name ID number) of the instant messaging object in the instant messaging client is obtained to generate an instant messaging object identity (ID) of the instant messaging object. On the one hand, since the user's real-name ID number of each instant messaging object is unique, the generated instant messaging object identity can be used to uniquely identify each instant messaging object, and the object that uniquely corresponds to the instant messaging object in the credential application platform can be determined through the instant messaging object identity to facilitate subsequent data association. On the other hand, by replacing the user's real-name ID number with the instant messaging object identity, data desensitization can be achieved to avoid leakage of the user's real-name ID number.

[0073] In S43, the behavior sequence feature, the first credential information, and the second credential information belonging to the instant messaging object are associated through the instant messaging object identifier.

[0074] In the disclosed embodiment, the behavior sequence data from the instant messaging server, the bound first credential information, and the second credential information from the credential application server are associated with multiple data sources. The multiple data sources involved are all based on the real-name authentication information of the instant messaging client. The system associates the multiple data sources through the user's unique instant messaging object identifier to ensure the accuracy of the data association.

[0075] In S44, the trained second feature extraction model is called to process the associated behavior sequence features, the first credential information and the second credential information to obtain the credential-associated feature information of the instant messaging object.

[0076] The second feature extraction model in the disclosed embodiment refers to a machine learning model or a deep learning model used to extract the characteristics of the associated behavior sequence, the first credential information, and the second credential information. The extracted characteristics are referred to as credential association characteristic information. The second feature extraction model can also be referred to as a credential feature model. The behavior labels (i.e., behavior sequence characteristics) output by the first feature extraction model are more inclined to the characteristic description of the operation of the instant messaging object in the instant messaging client.

[0077] Figure 3 It is a schematic diagram of generating a first feature extraction model and a second feature extraction model provided by an embodiment of the present disclosure. The first feature extraction model and the second feature extraction model in the embodiment of the present disclosure both have the ability to process tens of thousands of long sequence data, that is, they can extract features from the input tens of thousands of long sequences. The advertising system in the embodiment of the present disclosure has a big data model, that is, it is possible to train models of various business scenarios in the advertising system, such as training the first feature extraction model and the second feature extraction model, so that the behavior sequence data of tens of thousands of long sequences of instant messaging objects can be integrated to establish a credential feature model. Figure 3 As shown, the method provided by the embodiment of the present disclosure may include the following steps.

[0078] In S31, the advertising system sends a request to the big data model to read the user's behavior tags for the past two years.

[0079] In the embodiment of the present disclosure, the advertising system requests to obtain the behavior sequence characteristics of the instant messaging object in the past two years (this time period is only an example, and the present disclosure is not limited to this).

[0080] In S32, the big data model imports behavioral data from the instant messaging server.

[0081] In the disclosed embodiment, after receiving the request in S31 sent by the advertising system, the big data model responds to the request and imports the behavioral sequence data of each instant messaging object in the past two years from the instant messaging server.

[0082] In S33, the big data model establishes a long sequence behavior data model of tens of thousands levels.

[0083] In the disclosed embodiment, a ten thousand level long sequence behavior data model is pre-established by utilizing the big data model of the advertising system, i.e., the above-mentioned first feature extraction model. The behavior sequence data of each instant messaging object in the past two years imported from the instant messaging server in S32 is input into the established ten thousand level long sequence behavior data model, and the behavior sequence characteristics (behavior labels) of each instant messaging object in the past two years are predicted and output.

[0084] When establishing a ten-thousand-level long-sequence behavior data model, the ten-thousand-level long-sequence historical behavior sequence data of each instant messaging object can be collected and input into the ten-thousand-level long-sequence behavior data, and then the ten-thousand-level long-sequence behavior data model can be trained according to the real behavior sequence characteristics of the historical behavior sequence data until the model converges or reaches the set number of iterations.

[0085] In S34, the big data model outputs the user's behavior tags (i.e., behavior sequence features) for the past two years to the advertising system.

[0086] In the disclosed embodiment, the big data model returns the predicted behavior labels of each instant messaging object in the past two years to the advertising system.

[0087] In S35, the advertising system sends a request to the credential platform to read the data source of the user's application credential.

[0088] In the embodiment of the present disclosure, in the advertising system, an external data source is selected for import, that is, a request is sent to the credential platform to request the user (ie, object) of the credential platform to apply for a credential data source, that is, a user credential data source.

[0089] In S36, the credential platform returns the user credential application history record and approval result (ie, the second credential information) to the advertising system.

[0090] In the disclosed embodiment, after the credential platform receives the request sent by the advertising system, it imports the credential approval data source, i.e., the credentials held by the credential platform users and the user's credential application history records, i.e., which credentials have been applied for in the past, how many times they have been applied for, and whether the approval result of each credential application is passed or not, into the advertising system.

[0091] In S37, the advertising system reads the credentials (ie, first credential information) bound by the user in the instant messaging wallet from the instant messaging server.

[0092] In the disclosed embodiment, the advertising system also requests the instant messaging server to read the credentials bound to the instant messaging object in the instant messaging wallet, that is, to import the credential data held by the instant messaging object in the instant messaging wallet. When the instant messaging object holds an institutional credential, it means that there is an association between the institution's approval credential and the behavior tag of the instant messaging object. This type of data can help the credential feature model to more accurately identify the institution's preference when approving credentials.

[0093] It should be noted that the execution order of the above S31-S34, S35-S36 and S37 is not limited, and they can be executed in parallel or one after another, and the present disclosure does not limit this.

[0094] In S38, the advertising system desensitizes the data: the user's real-name information is identified and a user ID is generated.

[0095] In the disclosed embodiment, after the advertising system obtains the above-mentioned third-party data sources, it first makes the data transparent by identifying the user's real-name information (real-name authentication information of the instant messaging object), generating a unique user ID (instant messaging object identifier), and removing duplicate, erroneous or incomplete data to ensure data quality.

[0096] In some embodiments, a unique user ID can be generated by combining incremental information (such as a timestamp), fixed information (machine information of the instant messaging client, such as machine brand, type, operating system, etc.), user information (real-name ID number of the instant messaging object, used to prevent duplication), and random information (global serial number).

[0097] In S39, the advertising system performs data association: combining the user ID and associating multiple data sources.

[0098] In the disclosed embodiment, the advertising system associates the three-party data sources obtained above with the generated user ID, that is, associates the behavior sequence characteristics, the first credential information, and the second credential information of the instant messaging object.

[0099] In S310 , the advertising system imports model training data into the voucher feature model.

[0100] In S311 , the credential feature model reads the configured expert experience.

[0101] The expert experience configured in the embodiments of the present disclosure refers to the development of an artificially configured interface, which can intervene in the training results of the credential feature model through artificial experience, such as setting hyperparameters. Another example is data standardization: for different scenarios, the data output is analyzed through expert experience, the applicability in the scenario is analyzed, and standardized processing is performed to form the output of the credential feature model. Figure 5 The various labels shown.

[0102] In S312, the credential feature model is trained.

[0103] In the disclosed embodiment, the advertising system uses the above-mentioned associated third-party data sources as training data of the credential feature model (i.e., the real credential-associated feature information of the training data is known) and imports them into the credential feature model. The credential feature model outputs the predicted credential-associated feature information of the training data, constructs a loss function based on the real credential-associated feature information and predicted credential-associated feature information of the training data, trains the credential feature model, and obtains the trained credential feature model.

[0104] In S313, the voucher feature model is output to the advertising system.

[0105] In the disclosed embodiment, the trained credential feature model is stored in the advertising system.

[0106] The method provided by the embodiment of the present disclosure imports the user's behavioral sequence data of the past two years, the card binding data in the instant messaging wallet (the first credential information), and the credential platform approval data (i.e., the second credential data), and establishes a credential feature model with a long sequence level of tens of thousands through the big data model capability of the advertising system, comprehensive expert experience and parameter training. However, the models in the related technologies often have only hundreds of sequences, and the short sequences can only cover the hourly behavioral history for highly active users, which cannot solve the problem and will cause serious short-term interest chasing problems.

[0107] Figure 4 It is used to illustrate the impact of different sequence lengths on the model effect. Here, the behavior sequence data of instant messaging subjects browsing short videos and live broadcasts in the instant messaging client is taken as an example. Figure 4The horizontal axis is the length of the behavior sequence (i.e. the length of the behavior sequence data of the instant messaging object), and the vertical axis is the model AUC (Area Under Curve, the area under the ROC (receiver operating characteristic curve) curve and the coordinate axis). Figure 4 It can be seen that the model accuracy increases significantly as the sequence length increases.

[0108] In various content scenarios of instant messaging clients, users have a browsing history of over 10,000 and it continues to increase rapidly. Ordinary model capabilities only support short-term sequences of hundreds of lengths, which has great limitations and cannot explicitly characterize users' multi-peak interests (multi-peak interests refer to users' interests showing different peaks in multiple aspects, that is, users' interest distribution has multiple obvious points of interest, such as users will show different points of interest in browsing, clicking, purchasing and other behaviors). Information loss is serious, and users' long-tail interests (referring to people's more specific and less common interests and hobbies in a certain field. Unlike popular culture, followers of long-tail interests are usually niche) are lost. After cooperating with the big data model capabilities of the advertising system, the voucher platform can support long-sequence analysis of tens of thousands of behaviors, characterize features at the finest granularity, assist in the accuracy and diversity of data models, and realize the modeling of tens of thousands of long sequences for voucher recommendations.

[0109] In an exemplary embodiment, the credential-associated characteristic information of the instant messaging object includes at least one of the following: demographic characteristics (also referred to as demographic tags), behavioral characteristics, and interest characteristics (also referred to as interest tags).

[0110] Figure 5 FIG. 1 is a flowchart of another data processing method provided by an embodiment of the present disclosure. Figure 5 As shown, in Figure 5 In the embodiment, the above Figure 1 Step S5 in the embodiment may further include the following steps.

[0111] In S51, a first similarity between the demographic characteristics of the instant messaging object and the demographic characteristics of the target object is calculated, and a first weight of the first similarity is obtained.

[0112] In the disclosed embodiment, the demographic features of the instant messaging object and the target object may be vectorized respectively, and then the cosine distance, Euclidean distance, etc. between the two may be obtained as the first similarity between the target object and the corresponding instant messaging object.

[0113] In S52, a second similarity between the behavior characteristics of the instant messaging object and the behavior characteristics of the target object is calculated, and a second weight of the second similarity is obtained.

[0114] In the disclosed embodiment, the behavior features of the instant messaging object and the target object may be vectorized respectively, and then the cosine distance, Euclidean distance, etc. between the two may be obtained as the second similarity between the target object and the corresponding instant messaging object.

[0115] In S53, a third similarity between the interest-related feature of the instant messaging object and the interest-related feature of the target object is calculated, and a third weight of the third similarity is obtained.

[0116] In the disclosed embodiment, the interest-type features of the instant messaging object and the target object may be vectorized respectively, and then the cosine distance, Euclidean distance, etc. between the two may be obtained as the third similarity between the target object and the corresponding instant messaging object.

[0117] In S54, the matching degree between the instant messaging object and the target object is obtained according to the first similarity and its first weight, the second similarity and its second weight, and the third similarity and its third weight.

[0118] In the disclosed embodiment, the sum of the first weight, the second weight, and the third weight is equal to a preset constant, such as 1. The sizes of the three can be initially set to a size respectively, for example, the first weight of the demographic tag is 40%, the second weight of the behavioral tag is 35%, and the third weight of the interest tag is 25%, then the matching degree between the target object and the instant messaging object = 40% × first similarity + 35% × second similarity + 25% × third similarity. It can be understood that the above ratio parameters can only be used as system initialization parameters, and will be automatically adjusted according to the data recovery situation later.

[0119] In S55, according to the matching degree between the instant messaging object and the target object, objects of the same type as the target object are determined from the instant messaging objects.

[0120] In the embodiment of the present disclosure, a matching threshold can be set. When the matching degree between an instant messaging object and the target object is greater than the matching threshold, the instant messaging object and the target object are considered to be the same type of object. Otherwise, they are not the same type of object as the target object. The size of the matching threshold can be set according to actual needs, such as 70%, but the present disclosure is not limited thereto.

[0121] The credential feature model in the embodiment of the present disclosure describes the multi-dimensional information of the instant messaging object in the credential field, and is modeled through three categories of tags (demographic tags, behavioral tags, and interest tags) to improve the accuracy of identifying similar objects or similar users.

[0122] like Figure 6As shown, the labels output by the credential feature model can be vertically subdivided into demographic labels, behavioral labels, and interest-based labels.

[0123] In the disclosed embodiment, demographic labels refer to labels that are classified according to the demographic characteristics of the instant messaging object. For example, demographic labels may include, but are not limited to: age (referring to the age stage of the user, such as children, teenagers, middle-aged, elderly, etc.), gender (such as male or female), region (for example, the geographical area where the instant messaging object is located is determined based on the location of the instant messaging client), education level (or education level, such as elementary school, junior high school, high school, university, etc.), car, real estate, marriage (referring to the user's marital status, such as married, unmarried, divorced, etc.), social activity level (which can be determined by information such as the frequency of use of the instant messaging client), mobile phone system (such as the operating system of the instant messaging client), business, etc.

[0124] In the disclosed embodiment, age is a demographic label, which is segmented according to the age range of the real-name authentication information of the instant messaging object. Since the age range of the credential application is generally 18-55 years old, for example, the age label of the instant messaging object in the range of 18 to 24 years old can be set to T1, the age label of the instant messaging object in the range of 24 to 30 years old can be set to T2, the age label of the instant messaging object in the range of 30 to 35 years old can be set to T3, the age label of the instant messaging object in the range of 35 to 45 years old can be set to T4, and the age label of the instant messaging object in the range of 45 to 55 years old can be set to T5. The differences in consumption levels, assets, interests, etc. of different age groups are relatively obvious.

[0125] In the disclosed embodiment, region is also a demographic label, which is divided according to the city where the instant messaging object is located when using the instant messaging client to make payment. The institution has different voucher issuance and approval strategies for cities of different levels.

[0126] In the disclosed embodiments, a behavior tag refers to a tag that classifies and describes the behavior characteristics of an instant messaging object. These tags can be classified according to different dimensions, such as transaction behavior, investment preference, credit behavior, risk tolerance, etc. Each tag represents the behavior characteristics of an instant messaging object in a certain aspect. For example, a behavior tag may include, but is not limited to: consumption preferences, consumption habits, consumption frequency, consumption level, payment habits (payment by voucher), financial asset level, card-holding institution, credit, number of credits, overdue information, risk rating, etc.

[0127] In the disclosed embodiment, credit investigation is a behavior label that can be provided by a personal consumer credit platform bound to an instant messaging client, and is used to evaluate whether the instant messaging object is creditworthy and the reference credit rating.

[0128] In the disclosed embodiment, consumption frequency is a behavior label, for example, it can be divided according to the consumption frequency of instant messaging objects in the past three months. Instant messaging objects with high consumption frequency are helpful for organizations to find active credential users.

[0129] In the disclosed embodiment, the consumption level is a behavioral label. For example, it can be divided according to the average monthly consumption amount of the instant messaging object in the past six months. The instant messaging object with large consumption is helpful for the institution's installment business.

[0130] In the disclosed embodiment, the credit limit is a behavior label, which can be divided by the credit limit of the instant messaging object provided by the personal consumer credit platform bound to the instant messaging client. Users with a certain credit limit are more likely to be approved. Since the personal consumer credit platform can check the credit of the institution, it is the same as the institutional process, which is essentially an objective reflection of the user's credit.

[0131] In the disclosed embodiment, interest tags are an overview and description of the interests and hobbies of the instant messaging object, and these tags can be classified according to different interests and hobbies. For example, interest tags may include but are not limited to: audio-visual, games, cooking, news reading, art, technology, sports and outdoor, travel, etc. Interest tags can be divided according to the portrait data of the instant messaging object, and institutional credentials have more obvious group preferences. For example, car owners, business travelers, outbound travel, new house decoration, etc., behind them correspond to the needs of different funding scenarios.

[0132] The data processing method provided in the embodiment of the present disclosure provides a credential application recommendation method based on the instant messaging system and credential platform data, integrates the portrait data of the credential platform users and the instant messaging objects, and realizes the modeling of ten thousand-level long sequences through the credential feature model. By utilizing the ten thousand-level long sequence capability, the time range of the model data processing is expanded, and the accuracy of identifying similar users is improved (mainly for behavioral tags, which can evaluate the user's net worth in multiple aspects). By utilizing the ten thousand-level long sequence credential feature model, the diverse characteristics of users are extracted to assist users in selecting institutions or credential companies that are easier to approve, so as to meet the needs of users for applying for credentials, while meeting the demands of users for applying for credentials and the preferences of institutional customers.

[0133] Figure 7 FIG. 1 is a flowchart of another data processing method provided by an embodiment of the present disclosure. Figure 7 As shown, in Figure 7 In the embodiment, the above Figure 1 Step S6 in the embodiment may further include the following steps.

[0134] In S61, a single simulation approval index of the target voucher by the same object is obtained according to the approval probability of the target institution for the voucher-related characteristic information and the voucher-related characteristic information of the same object.

[0135] In the disclosed embodiment, a single simulation approval index refers to a simulation approval index of a single object of the same type for the target voucher. A simulation approval index of a same type for the target voucher refers to a simulation approval index obtained by combining the single simulation approval indexes of all objects of the same type for the target voucher.

[0136] In S62, according to the matching degree between the similar object and the target object, the simulation approval weight of the single simulation approval indicator of the target voucher of the similar object is determined.

[0137] In some embodiments, the simulation approval weight of the single simulation approval indicator of each similar object can be set to be the same. In other embodiments, the simulation approval weight of the single simulation approval indicator of the target voucher of the similar object with a greater matching degree can be set to be greater, and vice versa. The present disclosure does not limit the specific value of the simulation approval weight.

[0138] In S63, the simulation approval index of the same object for the target voucher is obtained according to the single simulation approval index of the same object for the target voucher and its simulation approval weight.

[0139] In the disclosed embodiment, the single simulation approval index of each similar object for the target voucher and its simulation approval weight may be weighted summed to obtain the simulation approval index of the target voucher of the similar object of the target object.

[0140] The data processing method provided by the embodiment of the present disclosure sets the simulation approval weight of a single simulation approval indicator according to the matching degree between each similar object and the target object, so that the simulation approval indicator of the obtained similar object for the target voucher can reflect the matching degree between different similar objects and the target object, thereby further improving the accuracy of the obtained simulation approval indicator, thereby more accurately assisting the target object to select the applied voucher.

[0141] Continue to refer Figure 7 In an exemplary embodiment, the above Figure 7 Step S62 in the embodiment may further include the following steps.

[0142] In S621, social data of the target object is obtained from the instant messaging server, the social data including the social objects of the target object and communication data between the target object and the social objects, the communication data including communication content and communication interaction frequency, and the instant messaging objects include the social objects.

[0143] In the embodiment of the present disclosure, the social data of the target object can also be obtained from the instant messaging server, that is, other instant messaging objects that have had contact with the target object through the instant messaging client, followed each other, viewed each other's information, and are in the target object's address book are considered to be the social objects of the target object, and the communication content between the social object and the target object, such as chat information, and the frequency of communication interaction, etc., are the communication data between the social object and the target object.

[0144] In S622, based on the communication data between the target object and the social object, a social strength index between the target object and the social object is determined.

[0145] In some embodiments, the social strength index may be set according to the communication interaction frequency between the target object and the social object. For example, the more frequent the communication interaction frequency between the two, the larger the corresponding social strength index; otherwise, the smaller the corresponding social strength index.

[0146] In other embodiments, the social strength index is calculated not only based on the communication interaction frequency between the social object and the target object, but also based on the communication content between the social object and the target object. For example, if the communication interaction frequency between a social object and a target object is more frequent, and the communication content between the two shows that the work content of the two is more similar, then the social strength index of the two is greater; otherwise, it is smaller. By comprehensively considering the communication content and the communication interaction frequency, the social strength index between the target object and the social object can be more accurately evaluated.

[0147] In S623, a simulation approval weight of a single simulation approval index of the target credential by the similar object is determined according to the matching degree between the similar object and the target object and the social strength index between the target object and the social object.

[0148] In the disclosed embodiment, after determining the simulated approval weight of the single simulated approval index of the same type of object according to the matching degree, it is further possible to determine whether the same type of object is the social object of the target object based on the social data from the instant messaging server. If it is the social object of the target object, the simulated approval weight of the single simulated approval index of the same type of object can be increased accordingly, for example, by 10%. In other embodiments, after determining that the same type of object is the social object of the target object, the simulated approval weight of the same type of object can also be increased according to the social strength index between the target object and the same type of object. For example, the larger the social strength index, the greater the proportion of the increase in the corresponding simulated approval weight.

[0149] The data processing method provided by the embodiment of the present disclosure, when determining the simulation approval index of the target credential for the same object, not only considers the matching degree between the same object and the target object, but also comprehensively considers whether the same object is the social object of the target object, and the social strength index when the same object is the social object of the target object, thereby further improving the accuracy of the simulation approval index of the target credential for the same object.

[0150] Figure 8 The schematic diagram of the architecture for implementing the data processing method provided by the embodiment of the present disclosure is exemplified. In the embodiment of the present disclosure, the voucher approval recommendation model 802 simultaneously integrates the voucher feature model obtained by the above training, the calculation of the matching degree to screen the same type of objects of the target object, and the determination function of obtaining the target object's approval pass index for the target voucher based on the simulated approval index of the same type of objects for the target voucher.

[0151] The overall framework of the data processing method provided by the embodiment of the present disclosure is as follows: Figure 8 As shown, when a user (e.g., a target object) logs into the credential platform 801 (i.e., logs into the credential application interface of the credential platform 801), the credential platform 801 recommends credentials to the user, i.e., displays a credential application list, which includes various credentials launched by various institutions or agencies, and also displays the approval index of each credential. Here, it is assumed that "X% of similar users have been approved", and the "X%" here refers to the simulated approval index of the target credential by the same objects, that is, the simulated approval index of the same objects or users for a certain credential can be used as the approval index of the target object for the credential. When the user clicks on a certain credential among the recommended credentials, the credential platform 801 jumps to the institution to apply for the credential.

[0152] In the process of the credential platform 801 recommending credentials to the user, the credential platform 801 uses the credential approval recommendation model 802 to predict and output the probability that similar users of the user have approved each recommended credential, for example, displaying each credential on the interface of the credential platform 801 and displaying "X% of similar users have approved" in the associated position of each credential.

[0153] The disclosed embodiment utilizes the portrait data source 803 (including user's ten thousand level long sequence behavior data) and the user credential data source 804 (e.g., the first credential information from the instant messaging wallet 8071 of the instant messaging server and the second credential information from the credential platform 801), integrates the capabilities of the advertising model 805 and the big data computing engine 806, generates the credential approval recommendation model 802, and finally outputs the information that X% of similar users have been reviewed and approved for the user. The advertising model 805 refers to the credential feature model obtained by training using the big data model capability of the above-mentioned advertising system. The big data computing engine 806 can call the credential feature model to obtain the credential-related feature information of the instant messaging object, calculate the matching degree based on the credential-related feature information of the instant messaging object, to screen the similar objects of the target object, and calculate the simulated approval index of the similar objects for the target credential, and determine the approval index of the target object for the target credential according to the simulated approval index of the similar objects for the target credential.

[0154] The data processing method provided in the embodiment of the present disclosure combines the behavioral sequence data of the instant messaging system and the card binding data of the instant messaging wallet and the credential platform to establish a credential approval recommendation model, obtain the approval pass index of the target object for the target credential, and apply it to the credential application recommendation scenario, so as to assist users in obtaining more information about the institution's review strategy when applying for credentials, and help users choose credentials of different institutions according to their needs, thereby improving user experience and efficiency. For the credential platform, it improves product competitiveness and helps to improve the overall review pass rate of the platform. For institutions, it helps institutions match users of different customer groups that the institutions need more, and match the two-way needs for credentials between institutions and users. That is, the method provided in the embodiment of the present disclosure can bring significant benefits to users, credential platforms, and institutions.

[0155] Fig. 9 FIG. 1 is a flowchart of another data processing method provided by an embodiment of the present disclosure. Fig. 9 As shown, in Fig. 9 In the embodiment, the above Figure 1 Step S7 in the embodiment may further include the following steps.

[0156] In S71, social data of the target object is obtained from the instant messaging service end, the social data includes social objects of the target object, and the instant messaging object includes the social objects.

[0157] In S72, the application frequency and the number of the target credential held by the social object are obtained from the credential application server.

[0158] In the disclosed embodiment, the credential platform can obtain information on how many times the social object of the target object has applied for the target credential on the credential platform (ie, application frequency) and how many target credentials the social object currently holds (ie, the number of certificates held).

[0159] In S73, according to the application frequency and the number of the target credentials held by the social object, the approval index of the target credentials by the social object is obtained.

[0160] In the disclosed embodiment, the ratio of the number of certificates held to the frequency of application can be used as the social object's approval index for the target certificate. For example, if the social object has applied for a certificate of an institution three times on the certificate platform, but currently only holds one certificate of the institution, then the social object's approval index for the certificate of the institution is 1 / 3.

[0161] In S74, a first configuration weight of a simulated approval indicator of the target credential by the same object and a second configuration weight of an approval pass indicator of the target credential by the social object are obtained.

[0162] In the embodiment of the present disclosure, a first configuration weight of the target object's similar objects for the simulation approval index of the target credential, and a second configuration weight of the target object's social objects for the approval pass index of the target credential can be configured. The sum of the first configuration weight and the second configuration weight can be equal to 1. The present disclosure does not limit the specific values ​​of the first configuration weight and the second configuration weight.

[0163] In S75, the approval index of the target object for the target credential is obtained according to the simulated approval index of the target object for the target credential and its first configuration weight, and the approval index of the target object for the target credential and its second configuration weight.

[0164] In the disclosed embodiment, the approval index of the target object for the target credential can be obtained by weighted summing the simulated approval index of the target object's similar objects and its first configuration weight, and the approval index of the target object's social objects and its second configuration weight for the target credential.

[0165] The data processing method provided by the embodiment of the present disclosure, when obtaining the approval pass index of the target object for the target credential, takes into account the simulated approval index of the target object's similar objects and the approval pass index of the target credential for the target credential by the target object's social objects. The target object's similar objects have similar credential-associated feature information as the target object, while the target object's social objects may have different credential-associated feature information from the target object, that is, the approval pass index of the target object for the target credential is comprehensively considered from different dimensions, thereby further improving the accuracy of the approval pass index of the target object for the target credential.

[0166] Continue to refer Fig. 9 In an exemplary embodiment, the social data also includes communication data between the target object and the social object, and the communication data includes communication content and communication interaction frequency.

[0167] Above Fig. 9 Step S73 in the embodiment may further include the following steps.

[0168] In S731, a single approval index of the social object for the target credential is obtained according to the application frequency and the number of the social object's holdings for the target credential.

[0169] In S732, based on the communication data between the target object and the social object, a social strength index between the target object and the social object is determined.

[0170] In S733, based on the social strength index between the target object and the social object, a weight of an approval index of a single approval index of the target credential by the social object is determined.

[0171] In some embodiments, different social objects may be set to have the same weight of the approval indicator. In other embodiments, when the social strength indicator between a certain social object and the target object is greater, the weight of the approval indicator of the social object is greater; otherwise, it is smaller.

[0172] In S734, the approval index of the social object for the target credential is obtained according to the single approval index of the social object for the target credential and its approval index weight.

[0173] In the disclosed embodiment, a weighted sum of single approval indicators and approval indicator weights of different social objects may be performed to serve as the approval indicator of the target object's social object for the target credential.

[0174] The data processing method provided in the embodiment of the present disclosure can set the weight of the approval index between the target object and the social object based on the communication data between the two, thereby achieving that the social object with a larger social strength index has a greater influence on the approval index of the target object, which can further improve the accuracy of the approval index of the target object for the target credential.

[0175] Fig.10 FIG. 1 is a flowchart of another data processing method provided by an embodiment of the present disclosure. Fig.10 As shown, in Fig.10 In the embodiment, the above Fig. 9 Step S75 in the embodiment may further include the following steps.

[0176] In S751, the application quantity and the approved quantity of the target voucher are obtained from the voucher application server.

[0177] In S752, based on the application quantity and approval quantity of the target voucher, a benchmark approval index of the target voucher is obtained.

[0178] In the disclosed embodiment, all application times (i.e., the number of applications) and the number of approvals for the target credential that have been applied for in the past are obtained from the credential application server, and the ratio of the number of approvals for the target credential to the number of applications is used as the benchmark approval indicator for the target credential.

[0179] In S753, the approval index of the target credential by the target object is obtained according to the benchmark approval index of the target credential, the simulated approval index of the target credential by the similar object and its first configuration weight, and the approval index of the target credential by the social object and its second configuration weight.

[0180] The data processing method provided in the embodiment of the present disclosure, when obtaining the approval index of the target object for the target credential, not only takes into account the simulated approval index of the target object's similar objects and the simulated approval index of the target object's social objects for the target credential, but also takes into account the benchmark approval index of the target credential on the credential application platform, thereby further improving the accuracy of the obtained approval index of the target object for the target credential.

[0181] Fig.11 is an interactive schematic diagram of a data processing method provided by an embodiment of the present disclosure. Fig.11 In the embodiment, a certificate is taken as an example, and the benchmark approval index of the target object, the social object and the certificate is taken as an example. Fig.11 As shown, the method provided by the embodiment of the present disclosure may include the following steps.

[0182] In S1101, the user (here, the target object) enters the credential homepage (ie, the credential application interface of the credential platform).

[0183] In S1102, the credential platform inputs a list of credentials that the user can apply for to the data platform.

[0184] In the disclosed embodiment, when a user logs in or opens a credential application interface, the credential application interface sends a request to the credential application server of the credential platform, and the credential application server of the credential platform determines a list of credentials that the user can apply for as a credential application list, and the credential list includes credentials launched by various institutions. The credential application server of the credential platform sends the list of credentials that the user can apply for to the data platform.

[0185] In S1103, the data platform inputs the full amount of behavior data from the instant messaging server.

[0186] In the disclosed embodiment, after the data platform receives the list of credentials that the user can apply for sent by the credential platform, it imports the full amount of behavior data from the instant messaging server, that is, the behavior sequence data of all instant messaging objects (including the target object) using the instant messaging client stored on the instant messaging server and the credential data bound to the instant messaging wallet (that is, the first credential information of each instant messaging object).

[0187] In S1104, the data platform will input the full amount of behavioral data into the credential feature model.

[0188] In the disclosed embodiment, the trained credential feature model is imported into the data platform. The data platform inputs the full amount of behavior data obtained from the instant messaging service end into the trained credential feature model to output the credential feature label of each instant messaging object, that is, the credential associated feature information.

[0189] In the embodiment of the present disclosure, the data platform also obtains the second credential information of each instant messaging object on the credential platform from the credential platform ( Fig.11 (not shown), that is, which credentials are held, which credentials have been applied for before, how many times have they been applied for, etc. After the data platform associates the full amount of behavior data of each instant messaging object with the second credential information, it is input into the credential feature model to output the credential association feature information of each instant messaging object.

[0190] In S1105, the credential feature model returns the credential feature label to the data platform.

[0191] In the disclosed embodiment, the credential feature model returns the output credential feature label of each instant messaging object to the data platform, for example Figure 5 The three major categories of labels are shown.

[0192] In S1106, the data platform calculates the user credential approval similarity (ie, matching degree).

[0193] In the disclosed embodiment, after obtaining the credential feature label of each instant messaging object, the data platform can use methods such as cosine similarity and Pearson correlation coefficient to calculate the matching degree between different instant messaging objects and the target object in the credential approval scenario to identify similar users of the target object.

[0194] In S1107, the data platform constructs a user credential approval matrix.

[0195] In the disclosed embodiment, each element in the user credential approval matrix describes the approval probability of each institution or each institution (including the target institution) for different credential feature tags.

[0196] In some embodiments, each institution sets its own approval probability for different credential feature tags, and then sends it to the credential platform. The credential platform aggregates the approval probabilities of each institution or organization for different credential feature tags to the data platform, and the data platform constructs a user credential approval matrix. In other embodiments, the data platform can set the user credential approval matrix based on historical cooperation experience with various institutions.

[0197] For example, taking age in the demographic label as an example, Institution A sets the approval probability of 0.1 for those aged 18 to 24; the approval probability set for those aged 24 to 30 is 0.2; the approval probability set for those aged 30 to 35 is 0.3; the approval probability set for those aged 35 to 45 is 0.3; and the approval probability set for those aged 45 to 55 is 0.1.

[0198] In S1108, the data platform inputs the user's social relationship from the instant messaging service.

[0199] In the disclosed embodiment, the data platform also obtains the user's social relationship (ie, social data) from the instant messaging server, namely, the social object of the target object, and the communication data between the target object and the social object, wherein the communication data includes the communication content and the communication interaction frequency.

[0200] In S1109, the data platform calculates the social relationship weight of the user credential.

[0201] In the disclosed embodiment, in the data platform, the social data source stored in the instant messaging server is imported to construct the user's social relationship network, which is based on the real friends of the target object. For example, the social identity relationship and social relationship strength index (i.e., social strength index) can be constructed based on the intimacy and interaction frequency (i.e., communication interaction rating) between the target object and its social object. For example, if the average daily interaction frequency exceeds 10 conversations, the social strength index = 1; if the average daily interaction frequency exceeds 100 conversations, the social strength index = 2, and so on, it can be modified according to the actual situation.

[0202] In the disclosed embodiment, the data platform calculates the social strength index between the target object and the social object based on the user social relationship of the target object obtained, and calculates the user social relationship weight (such as the simulated approval weight and / or the approval pass index weight) based on the social strength index. For example, when the same type of object is also the social object of the target object, the simulated approval weight of the same type of object is increased by 10%. When friends with similar behavioral tags, interest tags, and demographic tags hold the corresponding institutional credentials, it can be considered that the user (i.e., the target object) is more likely to obtain approval from the institution. That is, through the relationship between friends, the recognition or characteristic value of the user's identity is improved.

[0203] In S1110, the data platform calculates the approval rate of similar user institutions.

[0204] In the disclosed embodiment, the data platform can calculate the approval rate of institutions for similar users based on the social relationship weights of user credentials, the user credential approval matrix and the credential feature labels, that is, the simulated approval index of the credentials of each institution in the credential list for similar objects.

[0205] In the disclosed embodiment, the benchmark approval index of each institution's voucher in the voucher list can be calculated by the following formula = the number of approved vouchers of the institution / the number of applications for the institution's vouchers*100%, that is, the voucher application records of the user (here refers to all the applicants of the institution's vouchers) on the voucher platform are used as the denominator, and the approved records are used as the numerator to calculate the benchmark approval rate (i.e., the benchmark approval index). For example, if there are 1,000 people applying for the vouchers of Institution A, and 200 people pass the approval of the vouchers of Institution A, then the benchmark approval rate of the vouchers of Institution A is 20%.

[0206] In the disclosed embodiment, the approval results of similar users are evaluated: based on the credential feature tags of the matched similar users of the target object, the above-mentioned user credential approval matrix is ​​used to calculate the simulated institutional approval results (i.e., simulated approval indicators) of similar users applying for the credentials of each institution in the credential list in real time. After calculating and obtaining the single approval pass indicators of similar users for the credentials of each institution, if the similar users are the social objects of the target object, the simulated approval weight of the single simulated approval indicator of the similar users can be increased. For example, the net worth of the person (i.e., the target object) can be proved through friendships, such as friends with high-frequency communication with multiple executives, and the content is related to work, then the simulated approval weight of the similar users can be increased.

[0207] In the disclosed embodiment, according to the above steps, the friend approval rate (i.e., the social object's approval approval index for the target credential) and the similar user approval rate (i.e., the simulated approval index for the target credential by the similar object) are calculated, and combined with the benchmark approval approval rate of the credential, the target object's approval approval rate for the credential, i.e., the approval approval index, is calculated. For example, the approval approval rate of a certain institution's credential in the recommended list of credentials that the user can apply for = the benchmark approval approval rate of the institution's credential + (same user approval weight * same user approval rate) + (friend approval weight * friend approval rate). Here, the same user approval weight defaults to 40%, and the friend approval weight defaults to 60%, which will be automatically adjusted later. The above-mentioned similar user approval weight is the first configuration weight, and the friend approval weight is the second configuration weight. The overall confidence level is adjusted by adjusting the first configuration weight and the second configuration weight.

[0208] In an exemplary embodiment, the approval rate of similar users is equal to the weighted sum of the single simulated approval indexes of multiple similar users of the target object. If a similar user is also a friend of the target object, the simulated approval weight of this similar user is increased. The friend approval rate can be equal to the weighted sum of the approval indexes of multiple friends of the target object. The greater the social strength index of a friend, the greater the corresponding approval index weight.

[0209] In the disclosed embodiment, friends and similar users are judged from different dimensions. Friends are just in the address book of the target object, and they may not be similar users. Similar users are users who are selected from all instant messaging objects and belong to the same category as the target object.

[0210] In S1111, the data platform returns a credential recommendation list (including approval rate) to the credential platform.

[0211] In the disclosed embodiment, the target credential can be any credential in the credential recommendation list. The data platform can output recommendation data to the credential platform, that is, not only displaying the credential recommendation list (i.e., the credential application list) on the credential application interface, but also displaying the approval rate of each credential on the credential recommendation list (i.e., the approval index of each target credential by the target object).

[0212] In S1112, the credential platform displays a credential application list to the user.

[0213] In the disclosed embodiment, the credential application interface at the front end of the credential platform displays a credential application list, i.e., a credential recommendation list (including approval pass rate), to assist users in selecting credentials that are easier to approve. When a user enters the credential platform, the platform outputs the final recommended credential list and displays the estimated institutional approval pass rate (i.e., the target object's approval pass index for the target credential) of each credential in the credential list, to guide users to select institutional credentials that are easier to approve.

[0214] The method provided by the embodiment of the present disclosure integrates the big data computing capability of the data platform, uses the comprehensive relationship chain collaborative filtering algorithm, and calculates the credential recommendation results by analyzing the social relationship chain between users, which can improve the accuracy of credential application recommendation. Not only the credential recommendation list is displayed, but also the approval rate is displayed, which can reduce the difficulty of users selecting credentials.

[0215] Fig.12 Schematic diagram of an application scenario of a data processing method provided by an embodiment of the present disclosure. Fig.12 As shown, user 1201 (i.e., target object) uses terminal 1202 to open credential application interface 12021 of credential platform, and sends a trigger request to credential application server 1203 of credential platform through credential application interface 12021. Credential application server 1203 of credential platform is connected to the back-end servers of each institution with which it cooperates. For example, it is assumed that credential application server 1203 of credential platform is connected to back-end server 12041 of institution A and back-end server 12042 of institution B respectively. Credential application server 1203 of credential platform is also connected to instant messaging server 1209, advertising system 1210 and data platform 1211 respectively.

[0216] There is no limit on the number of users who can use the instant messaging client. Fig.12Two users are shown for illustration only. User 1205 uses terminal 1206 to log in to the instant messaging client, and user 1207 uses terminal 1208 to log in to the instant messaging client. Each instant messaging client is connected to the instant messaging server 1209. When users 1205 and 1207 operate their respective instant messaging clients, their respective behavior sequence data are generated and stored in the instant messaging server 1209. The instant messaging client is also bound to the instant messaging wallet, and the instant messaging server 1209 can also store the first credential information of each of users 1205 and 1207, that is, information related to the credential bound to the instant messaging wallet.

[0217] After receiving the trigger request, the credential application server 1203 sends a request to the data platform 1211. The data platform 1211 calls the trained credential feature model stored in the advertising system 1210, obtains the behavior sequence data and the first credential information of each instant messaging object from the instant messaging server 1209, obtains the second credential information from the credential application server 1203, desensitizes and associates the behavior sequence data, the first credential information and the second credential information of each instant messaging object, and inputs them into the credential feature model to obtain the credential-related feature information of each instant messaging object. Then, the big data computing engine in the data platform 1211 is used to obtain the matching degree between the instant messaging object and the target object according to the credential-related feature information of each instant messaging object, and the same type of the target object is determined from the instant messaging object according to the matching degree. According to the approval probability of the target organization for the credential-related feature information and the credential-related feature information of the same type of object, the simulated approval index of the same type of object for the target credential is obtained; according to the simulated approval index of the same type of object for the target credential, the approval pass index of the target object for the target credential is obtained. The data platform 1211 returns the target object's approval indicator for the target certificate to the certificate application server 1203, and the certificate application server 1203 returns the target object's approval indicator for the target certificate to the certificate application interface 12021.

[0218] When the user 1201 selects a certificate to apply for according to the displayed approval index in the certificate application interface 12021, the application request for the certificate will be sent to the certificate application server 1203. In response to the application request, the certificate application server 1203 will send an approval request to the back-end server of the corresponding institution. The back-end server of the corresponding institution will respond to the approval request, approve whether the application of the certificate by the user 1201 is approved, and return the approval result to the certificate application interface 12021.

[0219] Fig.13 Provides the certificate application process. Fig.13 As shown, the method provided by the embodiment of the present disclosure may include the following steps.

[0220] In S131 , the user accesses the credential platform.

[0221] In S132, the voucher platform calls the voucher approval recommendation model and recommends vouchers to the user through the voucher approval recommendation model. The specific content of the voucher approval recommendation model can refer to the above embodiment.

[0222] In S133, the jump institution applies for a certificate.

[0223] In S134, agency approval.

[0224] In S135, it is determined whether the approval is passed; if the approval is passed, the user can use the card (ie, the approved credential); if the approval is not passed, the process returns to the above step S131.

[0225] In S136, if the approval is passed, the user can use the card.

[0226] For users, when new users apply for vouchers, institutions often have difficulty or slow approval due to lack of credit records or low consumption levels, which ultimately affects the efficiency of such users' use of funds. The voucher recommendation method provided in the disclosed embodiment utilizes the capabilities of the voucher approval recommendation model, combines the portrait data of massive instant messaging systems, and the user voucher application result information held by the voucher platform itself, and provides users with a voucher application recommendation method based on instant messaging system and voucher platform data, helping users to efficiently meet their funding needs.

[0227] like Fig.14 As shown, the credential application recommendation method based on the instant messaging system and the credential platform data provided by the embodiment of the present disclosure performs data preheating in the advertising model 805 and the credential platform 801, which includes: after the full amount of features (i.e., behavior sequence data) and the institutional review data are imported into the advertising model 805 from the instant messaging server 807, the credential feature modeling is performed, that is, the credential feature model is generated. Among them, the imported full amount of features and institutional review data can be updated regularly. Among them, the institutional review data imported into the advertising model 805 includes the second credential information imported from the credential platform 201, that is, the identification of which institutions of the users of the credential platform have been reviewed and approved. The institutional review data imported into the advertising model 805 also includes the credential that has been bound to the instant messaging object identified from the instant messaging wallet 8071, that is, the first credential information. After the advertising model 805 performs data cleaning and data desensitization on the imported behavior sequence data and institutional review data, it performs 10,000-level sequence modeling for credential feature modeling, that is, generates a credential feature model. The full amount of features, the generated credential feature model and institutional review data are input into the data platform 141. The data platform 141 may include the above-mentioned big data computing engine 806.

[0228] When a user (target object) accesses the credential platform 801, if the user opens the credential platform 801 to apply for a credential, the big data computing engine 806 is used to match the user's tags, that is, the full amount of features and institutional review data are input into the credential feature model to identify the user's similar users, and the institutional review data of similar users is counted to obtain recommendation results, such as the simulated approval indicators of similar users of each credential. The big data computing engine 806 outputs the recommendation results to the credential platform 801, displays the credential recommendation list, and displays "X% of similar users have been reviewed and approved".

[0229] Fig.15 is a schematic diagram of a credential application interface provided by an embodiment of the present disclosure. Fig.15 As shown, a voucher application interface 1500 is displayed, and a voucher list is displayed on the voucher application interface 1500, for example, including a first area 1510 displaying the pink card of Institution A, a second area 1520 displaying the gold card of Institution A, and a third area 1530 displaying the platinum card of Institution B. Taking the first area 1510 as an example, the card appearance 1511 of the voucher of the pink card of Institution A, the icon 1512 of Institution A, the estimated approval rate 1513 (i.e., the approval index of the target object for the pink card of Institution A, assumed to be 78.6%), and the trigger control 1514 are displayed.

[0230] Fig.16 is a schematic diagram of another credential application interface provided by an embodiment of the present disclosure. In response to a trigger operation of a target object on the credential application interface, the credential application interface displays a credential application list of the target object and an approval index of the target object for the target credential, wherein the credential application list includes the target credential issued by the target institution. For example, when a user logs in to the credential application interface, the following is displayed: Fig.16 The credential application interface 1600 shown in the figure displays a credential list, for example, including a first area 1610 displaying the pink card of Institution A, a second area 1620 displaying the gold card of Institution A, and a third area 1630 displaying the platinum card of Institution B. Taking the first area 1610 as an example, the card appearance 1511 of the credential of the pink card of Institution A, the icon 1512 of Institution A, 78.8% of the same users have been approved (i.e., the simulated approval index of the gold card of Institution A by the same objects) 1613, and the trigger control 1514 are displayed.

[0231] Fig.17 is a schematic diagram of another credential application interface provided by an embodiment of the present disclosure. In response to the triggering operation of the target object on the target credential, the application page of the target credential is displayed. For example, assuming that the user clicks the above Fig.15 or Fig.16 If the trigger control of the certificate issued by any institution in Fig.17The application page 1700 of the voucher is shown. The application page 1700 displays some relevant information of the voucher, such as the type of voucher, the rights and interests enjoyed, etc. The application page 1700 may also include one or more check controls 1701 and an operation control 1702.

[0232] In response to the target object's triggering operation on the target credential in the application page of the target credential, the applicant information page is displayed; in response to the target object's operation instruction on the applicant information page, the identity information filling page is displayed; in response to the target object's operation instruction on the identity information filling page, the applicant information is obtained. For example, when the user checks the check control 1701 and clicks the operation control 1702, the following is displayed: Fig.18 The page 1800 for confirming the applicant's information can display one or more input controls 1810 and confirmation controls 1820. The user can operate the input controls 1810 to input the applicant's name, pinyin, ID card and other information. When the user clicks the confirmation control 1820, the page 1800 will jump to the following page: Fig.19 The identity information filling page 1900 is shown to further complete the card applicant information on the identity information filling page 1900.

[0233] Fig. 20 It is a flowchart of another data processing method provided by an embodiment of the present disclosure. Fig. 20 The method provided in the embodiment may be executed by any electronic device or computer device, for example, by a terminal device, but the present disclosure is not limited thereto. Fig. 20 As shown, the method provided by the embodiment of the present disclosure includes the following steps.

[0234] In S2001, in response to the target object's triggering operation on the credential application interface, the credential application list of the target object and the target object's approval index for the target credential are displayed on the credential application interface, and the credential application list includes the target credential issued by the target institution.

[0235] In S2002, in response to the target object's triggering operation on the target credential, an application page for the target credential is displayed.

[0236] The approval index is determined based on the simulated approval index of the target voucher by the same type of objects as the target object.

[0237] Fig. 20 For other contents of the embodiment, reference may be made to other embodiments of the present disclosure.

[0238] Fig.21 Schematic diagram of a data processing device provided by an embodiment of the present disclosure. Fig.21As shown, the data processing device 200 provided in the embodiment of the present disclosure may include a processing unit 201, a first receiving unit 202 and a second receiving unit 203. The processing unit 201 is used to determine a credential application list for the target object in response to a trigger operation of the target object on the credential application interface, and the credential application list includes a target credential issued by a target institution. The first receiving unit 202 is used to obtain the behavior sequence data generated by the instant messaging object during the use of the instant messaging client and the first credential information bound in the instant messaging client from the instant messaging service end, and the instant messaging object includes the target object. The second receiving unit 203 is used to obtain the second credential information held by the object from the credential application service end connected to the credential application interface, and the object includes the target object. The processing unit 201 is also used to merge and process the behavior sequence data, the first credential information and the second credential information of the instant messaging object to obtain the credential-related feature information of the instant messaging object. The processing unit 201 is also used to obtain the matching degree between the instant messaging object and the target object according to the credential-related feature information of the instant messaging object and the credential-related feature information of the target object, and determine the same type of object of the target object from the instant messaging object according to the matching degree. The processing unit 201 is also used to obtain the simulated approval index of the target voucher by the target object according to the voucher-related characteristic information of the target object. The processing unit 201 is also used to obtain the approval index of the target object for the target voucher according to the simulated approval index of the target voucher by the target object, so as to display the voucher application list and the approval index of the target object for the target voucher for the target object on the voucher application interface.

[0239] In an exemplary embodiment, the processing unit 201 is also used to: call a trained first feature extraction model, process the behavior sequence data of the instant messaging object, and obtain the behavior sequence features of the instant messaging object; generate an instant messaging object identifier for the instant messaging object through the real-name authentication information of the instant messaging object in the instant messaging client; associate the behavior sequence features, the first credential information, and the second credential information belonging to the instant messaging object through the instant messaging object identifier; call a trained second feature extraction model, process the associated behavior sequence features, the first credential information, and the second credential information, and obtain the credential association feature information of the instant messaging object.

[0240] In an exemplary embodiment, the credential-associated feature information of the instant messaging object includes at least one of the following: demographic features, behavioral features, and interest-based features. The processing unit 201 is also used to: calculate a first similarity between the demographic features of the instant messaging object and the demographic features of the target object, and obtain a first weight for the first similarity; calculate a second similarity between the behavioral features of the instant messaging object and the behavioral features of the target object, and obtain a second weight for the second similarity; calculate a third similarity between the interest-based features of the instant messaging object and the interest-based features of the target object, and obtain a third weight for the third similarity; obtain a matching degree between the instant messaging object and the target object based on the first similarity and its first weight, the second similarity and its second weight, and the third similarity and its third weight; determine a similar object of the target object from the instant messaging object based on the matching degree between the instant messaging object and the target object.

[0241] In an exemplary embodiment, the processing unit 201 is also used to: construct the target institution's approval probability for the voucher-associated characteristic information; and obtain the simulated approval index of the similar object for the target voucher based on the target institution's approval probability for the voucher-associated characteristic information and the voucher-associated characteristic information of the similar object.

[0242] In an exemplary embodiment, the processing unit 201 is also used to: obtain a single simulation approval index of the same object for the target voucher based on the approval probability of the target institution for the voucher-associated characteristic information and the voucher-associated characteristic information of the same object; determine the simulation approval weight of the single simulation approval index of the same object for the target voucher based on the matching degree between the same object and the target object; obtain the simulation approval index of the same object for the target voucher based on the single simulation approval index of the same object for the target voucher and its simulation approval weight.

[0243] In an exemplary embodiment, the first receiving unit 202 is also used to: obtain the social data of the target object from the instant messaging service end, the social data including the social object of the target object and the communication data between the target object and the social object, the communication data including the communication content and the communication interaction frequency, and the instant messaging object includes the social object. The processing unit 201 is also used to: determine the social strength index between the target object and the social object based on the communication data between the target object and the social object; determine the simulation approval weight of the single simulation approval index of the target credential by the similar object according to the matching degree between the similar object and the target object and the social strength index between the target object and the social object.

[0244] In an exemplary embodiment, the first receiving unit 202 is further used to: obtain the social data of the target object from the instant messaging service end, the social data includes the social object of the target object, and the instant messaging object includes the social object. The second receiving unit 203 is further used to: obtain the application frequency and the number of the target credential held by the social object from the credential application service end. The processing unit 201 is further used to: obtain the approval index of the target credential by the social object according to the application frequency and the number of the target credential held by the social object; obtain the first configuration weight of the simulated approval index of the target credential by the same object and the second configuration weight of the approval index of the target credential by the social object; obtain the approval index of the target credential by the target object according to the simulated approval index of the target credential by the same object and its first configuration weight, and the approval index of the target credential by the social object and its second configuration weight.

[0245] In an exemplary embodiment, the social data also includes communication data between the target object and the social object, and the communication data includes communication content and communication interaction frequency. The processing unit 201 is also used to: obtain the single approval index of the social object for the target credential according to the application frequency and the number of the social object for the target credential; determine the social strength index between the target object and the social object based on the communication data between the target object and the social object; determine the approval index weight of the single approval index of the social object for the target credential based on the social strength index between the target object and the social object; obtain the approval index of the social object for the target credential according to the single approval index of the social object for the target credential and its approval index weight.

[0246] In an exemplary embodiment, the second receiving unit 203 is further used to: obtain the number of applications and the number of approvals of the target credential from the credential application service end. The processing unit 201 is further used to: obtain the benchmark approval index of the target credential based on the number of applications and the number of approvals of the target credential; obtain the approval index of the target credential by the target object according to the benchmark approval index of the target credential, the simulated approval index of the target credential by the same object and its first configuration weight, and the approval index of the target credential by the social object and its second configuration weight.

[0247] In an exemplary embodiment, the data processing device 200 also includes a first sending unit, which is used to send the credential application list and the target object's approval indicator for the target credential, so as to display the credential application list and the target object's approval indicator for the target credential for the target object on the credential application interface.

[0248] Fig.21 For other contents of the embodiment, reference can be made to other embodiments mentioned above and will not be repeated here.

[0249] Fig. 22 Schematic diagram of a data processing device provided by an embodiment of the present disclosure. Fig. 22 As shown, the data processing device 220 provided in the embodiment of the present disclosure includes a response unit 221.

[0250] The response unit 221 is used to respond to the trigger operation of the target object on the credential application interface, and display the credential application list of the target object and the approval index of the target object for the target credential on the credential application interface. The credential application list includes the target credential issued by the target institution.

[0251] The response unit 221 is further configured to display an application page for the target credential in response to a triggering operation of the target object on the target credential.

[0252] The approval index is determined based on the simulated approval index of the target voucher by the same type of objects as the target object.

[0253] Fig. 22 For other contents of the embodiment, reference can be made to other embodiments mentioned above and will not be repeated here.

[0254] Furthermore, an embodiment of the present disclosure also provides a computer device, including a processor, a memory, and an input-output interface; the processor is connected to the memory and the input-output interface, respectively, wherein the input-output interface is used to receive data and output data, the memory is used to store a computer program, and the processor is used to call the computer program so that the computer device executes the method provided by any embodiment of the present disclosure.

[0255] Furthermore, an embodiment of the present disclosure also provides a computer-readable storage medium, which stores a computer program, and the computer program is suitable for being loaded and executed by a processor so that a computer device having the processor executes the method provided by any embodiment of the present disclosure.

[0256] Furthermore, an embodiment of the present disclosure also provides a computer program product, including a computer program, which implements the method provided by any embodiment of the present disclosure when executed by a processor.

[0257] It is understandable that the terminal mentioned in the embodiments of the present disclosure may be a computer device, and the computer device in the embodiments of the present disclosure includes but is not limited to a terminal or a server. In other words, the computer device may be a server or a terminal, or a system consisting of a server and a terminal. Among them, the terminal mentioned above may be an electronic device, including but not limited to a mobile phone, a tablet computer, a desktop computer, a laptop computer, a PDA, a vehicle-mounted device, an augmented reality / virtual reality (AR / VR) device, a helmet display, a smart TV, a wearable device, a smart speaker, a digital camera, a camera, and other mobile Internet devices (mobile internet device, MID) with network access capabilities, or terminals in scenarios such as trains, ships, and flights.

[0258] Among them, the servers mentioned above can be independent physical servers, or they can be server clusters or distributed systems composed of multiple physical servers. They can also be cloud servers that provide basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, vehicle-road collaboration, content delivery networks (Content Delivery Network, CDN), as well as big data and artificial intelligence platforms.

[0259] Optionally, the data involved in the embodiments of the present disclosure may be stored in a computer device, or the data may be stored based on cloud storage technology, which is not limited here.

[0260] See also Fig.23 , Fig.23 Schematic diagram of a computer device provided by an embodiment of the present disclosure. Fig.23 As shown, the computer device in the embodiment of the present disclosure may include: one or more processors 2301, a memory 2302, and an input-output interface 2303. The processor 2301, the memory 2302, and the input-output interface 2303 are connected via a bus 2304. The memory 2302 is used to store a computer program, which includes program instructions. The input-output interface 2303 is used to receive data and output data, such as for data interaction between a host machine and a computer device, or for data interaction between various virtual machines in a host machine; the processor 2301 is used to execute the program instructions stored in the memory 2302.

[0261] The processor 2301 may execute the method described in any embodiment of the present disclosure.

[0262] In some feasible implementations, the processor 2301 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or the processor may be any conventional processor, etc.

[0263] The memory 2302 may include a read-only memory and a random access memory, and provide instructions and data to the processor 2301 and the input / output interface 2303. A portion of the memory 2302 may also include a non-volatile random access memory. For example, the memory 2302 may also store information about the device type.

[0264] The terms "first", "second", etc. in the specification, claims, and drawings of the embodiments of the present disclosure are used to distinguish different objects, rather than to describe a specific order. In addition, the terms "including" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, device, product, or equipment that includes a series of steps or units is not limited to the listed steps or modules, but may optionally include steps or modules that are not listed, or may optionally include other step units inherent to these processes, methods, devices, products, or equipment.

[0265] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in this description according to function. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this disclosure.

[0266] The method and related apparatus provided by the embodiments of the present disclosure are described with reference to the method flow chart and / or structural diagram provided by the embodiments of the present disclosure. Specifically, each process and / or box in the method flow chart and / or structural diagram, as well as the combination of the processes and / or boxes in the flow chart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable application display device to generate a machine, so that the instructions executed by the processor of the computer or other programmable application display device generate instructions for implementing the process in the process. Figure 1 A process or multiple processes and / or structures Figure 1 These computer program instructions can also be stored in a computer-readable memory that can guide a computer or other programmable application display device to work in a specific manner, so that the instructions stored in the computer-readable memory produce a product including an instruction device, which implements the functions specified in the process. Figure 1 A process or multiple processes and / or structures Figure 1 These computer program instructions can also be loaded onto a computer or other programmable application display device, so that a series of operation steps are executed on the computer or other programmable device to produce a computer-implemented process, so that the instructions executed on the computer or other programmable device provide for implementing the process in the process. Figure 1 A flow or multiple flows and / or structures illustrate the steps of the functions specified in one block or multiple blocks.

Claims

1. A data processing method, characterized in that: include: In response to a trigger operation of a target object on a credential application interface, determining a credential application list for the target object, the credential application list including a target credential issued by a target institution; Acquire, from an instant messaging service end, behavior sequence data generated by an instant messaging object during use of an instant messaging client and first credential information bound to the instant messaging client, wherein the instant messaging object includes the target object; Acquire second credential information held by an object from a credential application server connected to the credential application interface, the object including the target object; fusing and processing the behavior sequence data of the instant messaging object, the first credential information, and the second credential information to obtain credential-related feature information of the instant messaging object; Obtaining a degree of match between the instant messaging object and the target object according to the credential-associated feature information of the instant messaging object and the credential-associated feature information of the target object, and determining a similar object of the target object from the instant messaging object according to the degree of match; According to the voucher-related characteristic information of the same object, obtaining the simulated approval index of the same object for the target voucher; According to the simulated approval index of the target voucher by the similar object, the approval pass index of the target voucher by the target object is obtained.

2. The method according to claim 1, characterized in that The step of fusing and processing the behavior sequence data of the instant messaging object, the first credential information, and the second credential information to obtain credential-related feature information of the instant messaging object includes: Calling the trained first feature extraction model to process the behavior sequence data of the instant messaging object to obtain the behavior sequence features of the instant messaging object; Generating an instant messaging object identifier for the instant messaging object through the real-name authentication information of the instant messaging object in the instant messaging client; Associating the behavior sequence feature, the first credential information, and the second credential information belonging to the instant messaging object through the instant messaging object identifier; The trained second feature extraction model is called to process the associated behavior sequence features, the first credential information, and the second credential information to obtain credential-associated feature information of the instant messaging object.

3. The method according to claim 1 or 2, characterized in that The credential-related characteristic information of the instant messaging object includes at least one of the following: demographic characteristics, behavioral characteristics, and interest characteristics; The method of obtaining a matching degree between the instant messaging object and the target object according to the credential-associated characteristic information of the instant messaging object and the credential-associated characteristic information of the target object, and determining a similar object of the target object from the instant messaging object according to the matching degree includes: Calculating a first similarity between the demographic characteristics of the instant messaging object and the demographic characteristics of the target object, and obtaining a first weight of the first similarity; Calculating a second similarity between the behavior characteristics of the instant messaging object and the behavior characteristics of the target object, and obtaining a second weight of the second similarity; Calculating a third similarity between the interest-related feature of the instant messaging object and the interest-related feature of the target object, and obtaining a third weight of the third similarity; Obtaining a matching degree between the instant messaging object and the target object according to the first similarity and its first weight, the second similarity and its second weight, and the third similarity and its third weight; According to the matching degree between the instant messaging object and the target object, a similar object of the target object is determined from the instant messaging objects.

4. The method according to claim 1, characterized in that According to the voucher-related characteristic information of the same object, obtaining the simulated approval index of the same object for the target voucher includes: Constructing the probability of the target institution approving the characteristic information associated with the credential; According to the approval probability of the target institution for the voucher-related characteristic information and the voucher-related characteristic information of the similar object, a simulated approval index of the similar object for the target voucher is obtained.

5. The method according to claim 4, characterized in that According to the approval probability of the target institution for the voucher-related characteristic information and the voucher-related characteristic information of the similar object, obtaining the simulated approval index of the similar object for the target voucher, including: According to the approval probability of the target institution for the voucher-related characteristic information and the voucher-related characteristic information of the similar object, a single simulated approval index of the similar object for the target voucher is obtained; Determining a simulation approval weight of a single simulation approval indicator of the target voucher by the similar object according to a matching degree between the similar object and the target object; According to the single simulation approval index of the same object for the target voucher and its simulation approval weight, the simulation approval index of the same object for the target voucher is obtained.

6. The method according to claim 5, characterized in that Determining the simulation approval weight of the single simulation approval indicator of the target voucher by the similar object according to the matching degree between the similar object and the target object includes: Acquire the social data of the target object from the instant messaging server, the social data including the social objects of the target object and the communication data between the target object and the social objects, the communication data including the communication content and the communication interaction frequency, and the instant messaging objects including the social objects; Determining a social strength index between the target object and the social object based on the communication data between the target object and the social object; According to the matching degree between the similar object and the target object, and the social strength index between the target object and the social object, a simulation approval weight of a single simulation approval index of the similar object for the target credential is determined.

7. The method according to claim 1, characterized in that Obtaining the approval index of the target object for the target credential according to the simulated approval index of the target object for the target credential includes: Acquire the social data of the target object from the instant messaging server, wherein the social data includes the social objects of the target object, and the instant messaging objects include the social objects; Obtaining the application frequency and the number of the target credential held by the social object from the credential application server; According to the application frequency and the number of the target credentials held by the social object, obtaining the approval index of the target credentials by the social object; Obtaining a first configuration weight of a simulated approval indicator of the target credential by the same object and a second configuration weight of an approval pass indicator of the target credential by the social object; The approval index of the target object for the target credential is obtained according to the simulated approval index of the similar object for the target credential and its first configuration weight, and the approval index of the social object for the target credential and its second configuration weight.

8. The method according to claim 7, characterized in that The social data also includes communication data between the target object and the social object, and the communication data includes communication content and communication interaction frequency; Wherein, obtaining the approval index of the target credential by the social object according to the application frequency and the number of the target credential held by the social object includes: According to the application frequency and the number of the target credentials held by the social object, obtaining a single approval index of the target credentials by the social object; Determining a social strength index between the target object and the social object based on the communication data between the target object and the social object; Determining, based on the social strength index between the target object and the social object, an approval index weight of a single approval index of the target credential by the social object; The approval index of the target credential by the social object is obtained according to the single approval index of the target credential by the social object and its approval index weight.

9. The method according to claim 7, characterized in that According to the simulated approval index of the target credential by the similar object and its first configuration weight, and the approval pass index of the target credential by the social object and its second configuration weight, obtaining the approval pass index of the target credential by the target object includes: Obtaining the number of applications and the number of approved applications for the target credential from the credential application server; Based on the number of applications and the number of approvals for the target voucher, a benchmark approval index for the target voucher is obtained; The approval index of the target credential by the target object is obtained according to the benchmark approval index of the target credential, the simulated approval index of the target credential by the similar object and its first configuration weight, and the approval index of the target credential by the social object and its second configuration weight.

10. The method according to claim 1, characterized in that Also includes: The credential application list and the target object's approval indicator for the target credential are sent to display the credential application list and the target object's approval indicator for the target credential on the credential application interface for the target object.

11. A data processing method, characterized in that: include: In response to a trigger operation of the target object on the credential application interface, displaying the target object's credential application list and the target object's approval index for the target credential on the credential application interface, wherein the credential application list includes the target credential issued by the target institution; In response to the target object triggering the target credential, displaying an application page for the target credential; The approval index is determined based on the simulated approval index of the target voucher by the same type of objects as the target object.

12. A data processing device, characterized in that: include: A processing unit, configured to determine a credential application list for the target object in response to a trigger operation of the target object on the credential application interface, wherein the credential application list includes a target credential issued by a target institution; A first receiving unit is used to obtain, from an instant messaging server, behavior sequence data generated by an instant messaging object during use of an instant messaging client and first credential information bound to the instant messaging client, wherein the instant messaging object includes the target object; A second receiving unit, configured to obtain second credential information held by an object from a credential application server connected to the credential application interface, the object including the target object; The processing unit is further used to merge and process the behavior sequence data of the instant messaging object, the first credential information and the second credential information to obtain the credential-related feature information of the instant messaging object; The processing unit is further configured to obtain a matching degree between the instant messaging object and the target object based on the credential-associated feature information of the instant messaging object and the credential-associated feature information of the target object, and determine a similar object of the target object from the instant messaging object based on the matching degree; The processing unit is further configured to obtain a simulation approval index of the target voucher by the object of the same type according to the voucher-related characteristic information of the object of the same type; The processing unit is further configured to obtain an approval pass index of the target object for the target voucher based on a simulated approval index of the target object for the target voucher.

13. A data processing device, characterized in that: include: A response unit, configured to respond to a trigger operation of a target object on a credential application interface, and display a credential application list of the target object and an approval pass index of the target credential by the target object on the credential application interface, wherein the credential application list includes target credentials issued by a target institution; The response unit is further configured to display an application page for the target credential in response to a trigger operation of the target object on the target credential; The approval index is determined based on the simulated approval index of the target voucher by the same type of objects as the target object.

14. A computer device, characterized in that: Includes processor, memory, input and output interfaces; The processor is connected to the memory and the input / output interface respectively, wherein the input / output interface is used to receive and output data, the memory is used to store a computer program, and the processor is used to call the computer program so that the computer device executes the method described in any one of claims 1 to 10, or executes the method described in claim 11.

15. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and the computer program is suitable for being loaded and executed by a processor, so that a computer device having the processor executes the method according to any one of claims 1 to 10, or executes the method according to claim 11.

16. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, it implements the method described in any one of claims 1 to 10, or executes the method described in claim 11.

Images