Recommendation processing method and device

Abstract

The embodiment of the invention provides a recommendation processing method and device.The recommendation processing method comprises the steps that in the recommendation processing process of resource services, after attribute data of a user, first behavior data of the user in the resource services and credit data of the user are obtained, the first behavior data of the user in the resource services are obtained; and predicting second behavior data of the user in the third-party resource service according to the credit data, identifying a service state of the user in the resource service according to the attribute data, the first behavior data and the second behavior data, inputting the service state and the attribute data into a large language model to generate a recommendation strategy, obtaining the recommendation strategy, and performing recommendation processing.

Description

Technical Field

[0001] This document relates to the field of data processing technology, and in particular to a recommendation processing method and device. Background Art

[0002] With the continuous development and promotion of Internet technology, the application scope of recommendation processing based on Internet technology is becoming wider and wider, covering many fields, such as resource service fields. Users' demand for recommendation processing is becoming more and more diversified. Based on this, the application of large language models provides a feasible means for user recommendation processing. However, as the application scope of large language models continues to expand and the application scenarios of large language models become increasingly complex, the requirements for the performance and functions of large language models are also increasing, which brings certain challenges to the application of large language models. Summary of the invention

[0003] One or more embodiments of the present specification provide a recommendation processing method, including: obtaining attribute data of a user, first behavior data of the user in a resource service, and credit data of the user. Predicting the behavior of the user in a third-party resource service based on the credit data to obtain second behavior data. Based on the attribute data, the first behavior data, and the second behavior data, identifying the service status of the user to obtain the service status of the user in the resource service. Inputting the service status and the attribute data into a large language model to generate a recommendation strategy to obtain a recommendation strategy, and performing recommendation processing.

[0004] One or more embodiments of the present specification provide a recommendation processing device, including: a data acquisition module, configured to acquire attribute data of a user, first behavior data of the user in a resource service, and credit data of the user. A behavior prediction module, configured to predict the behavior of the user in a third-party resource service based on the credit data to obtain second behavior data. A state identification module, configured to identify the service state of the user based on the attribute data, the first behavior data, and the second behavior data to obtain the service state of the user in the resource service. A strategy generation module, configured to input the service state and the attribute data into a large language model to generate a recommendation strategy to obtain a recommendation strategy, and perform recommendation processing.

[0005] One or more embodiments of the present specification provide a recommendation processing device, including: a processor; and a memory configured to store computer executable instructions, wherein when the computer executable instructions are executed, the processor: obtains attribute data of a user, first behavior data of the user in a resource service, and credit data of the user. Predict the behavior of the user in a third-party resource service based on the credit data to obtain second behavior data. Based on the attribute data, the first behavior data, and the second behavior data, identify the service status of the user to obtain the service status of the user in the resource service. Input the service status and the attribute data into a large language model to generate a recommendation strategy to obtain a recommendation strategy, and perform recommendation processing.

[0006] One or more embodiments of the present specification provide a computer-readable storage medium for storing computer-executable instructions, which implement the following steps when executed: obtaining attribute data of a user, first behavior data of the user in a resource service, and credit data of the user. Predicting the behavior of the user in a third-party resource service based on the credit data to obtain second behavior data. Based on the attribute data, the first behavior data, and the second behavior data, identifying the service status of the user to obtain the service status of the user in the resource service. Inputting the service status and the attribute data into a large language model to generate a recommendation strategy to obtain a recommendation strategy, and performing recommendation processing. BRIEF DESCRIPTION OF THE DRAWINGS

[0007] In order to more clearly illustrate one or more embodiments of the present specification or the technical solutions in the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present specification. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative labor. Figure 1 A schematic diagram of an implementation environment of a recommended processing method provided for one or more embodiments of this specification; Figure 2 A processing flow chart of a recommended processing method provided for one or more embodiments of this specification; Figure 3 A processing flow chart of a recommended processing method applied to a resource service scenario provided in one or more embodiments of this specification; Figure 4 A schematic diagram of a recommended processing device embodiment provided for one or more embodiments of this specification; Figure 5 A schematic diagram of the structure of a recommended processing device provided for one or more embodiments of this specification. DETAILED DESCRIPTION

[0008] In order to enable those skilled in the art to better understand the technical solutions in one or more embodiments of this specification, the technical solutions in one or more embodiments of this specification will be clearly and completely described below in conjunction with the drawings in one or more embodiments of this specification. Obviously, the described embodiments are only part of the embodiments of this specification, not all of them. Based on one or more embodiments of this specification, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of this document.

[0009] The recommended processing method provided in one or more embodiments of this specification can be applied to the implementation environment of recommended processing. Figure 1 , the implementation environment at least includes: A server 101 for performing recommendation processing; in addition, the implementation environment may also include a user terminal 102.

[0010] Among them, server 101 can be one or more servers, a server cluster composed of several servers, or a cloud server of a cloud computing platform; server 101 can be deployed with a behavior prediction module 103, a service status identification module 104 and a large language model 105; the behavior prediction module 103 can predict the user's second behavior data in a third-party resource service based on the user's credit data, the service status identification module 104 can identify the user's service status in the resource service based on the user's attribute data, the user's first behavior data in the resource service and the second behavior data, and the large language model 105 can generate a recommendation strategy based on the service status and attribute data and perform recommendation processing.

[0011] The user terminal 102 may be a mobile phone, a personal computer, a tablet computer, an e-book reader, a wearable device, a device for information interaction based on AR (Augmented Reality) / VR (Virtual Reality), a laptop computer, etc., and may also be installed with an application program, and subprograms may be run within the application program, and recommendation processing may be performed through the application program or the subprogram.

[0012] In this implementation environment, server 101 performs behavior prediction based on the user's credit data, obtains the user's second behavior data in a third-party resource service, identifies the service status based on the user's attribute data, the user's first behavior data in the resource service, and the second behavior data, obtains the user's service status in the resource service, and then generates a recommendation strategy based on the service status and attribute data to obtain the recommendation strategy and perform recommendation processing.

[0013] It should be pointed out that, considering that user-related data belongs to the user's privacy to a certain extent, if you want to obtain user-related data such as user attribute data, user's first behavior data in resource services, and user's credit data, you can obtain user authorization before obtaining it. The authorization method can be to send a data acquisition reminder to the user during the recommendation process. After the user confirms the reminder through an instruction, he can obtain authorization to obtain user-related data. In addition, user authorization can also be obtained through other methods, such as obtaining authorization to obtain user-related data through the signing of an agreement.

[0014] One or more embodiments of a recommended processing method provided in this specification are as follows: Reference Figure 2 The recommended processing method provided in this embodiment specifically includes steps S202 to S208.

[0015] Step S202, obtaining the user's attribute data, the user's first behavior data in the resource service, and the user's credit data.

[0016] In this embodiment, the user's attribute data may include the user's basic attribute data, such as the user's date of birth, occupation and / or education level. In addition, the user's attribute data may also be other types of attribute data, such as the user's salary amount. Resource services refer to services related to resources, such as resource services may be resource application services, resource services may be applications, subprograms within applications (such as mini-programs), or resource service units within applications, which may be used for recommendation processing, and applications may be payment applications.

[0017] The first behavior data of a user in a resource service may include the behavior data of the user in the resource service, and specifically may include the resource behavior data of the user in the resource service. For example, the first behavior data may be a resource application record of the user in the resource service, and the resource application record may include the resource application time, the resource application amount and / or whether the resource return is overdue; in addition, the first behavior data may also include other types of behavior data, such as the browsing data of the user in the resource service.

[0018] The user's credit data may include the user's credit data on a third-party platform. The third-party platform may be a third-party platform corresponding to a financial management institution. For example, the credit data may be the user's credit data or credit report on a third-party platform. Optionally, credit data can be obtained by: Obtain the user's credit data on the third-party platform; and / or obtain the application materials submitted by the user during the resource application process of the resource service, and extract the user's credit data from the application materials.

[0019] The application materials may include resource application materials, such as a user's credit report, a user's asset data and / or a user's identity data, and may also include other resource application materials.

[0020] Specifically, on the one hand, the user's credit data can be obtained from a third-party platform. Due to the professionalism of the third-party platform and the users' recognition of the third-party platform, the user's credit data obtained from the third-party platform is more comprehensive. On the other hand, the application materials submitted by the user in the process of applying for resources for resource services can be obtained, and the user's credit data can be extracted from the application materials. In addition, in order to further improve the comprehensiveness of the credit data, the user's first credit data can be obtained from the third-party platform, and then the application materials submitted by the user in the process of applying for resources for resource services can be obtained, and the user's second credit data can be extracted from the application materials, and the first credit data and the second credit data can be used together as the user's credit data.

[0021] It should be noted that after obtaining the user's credit data as mentioned above, the user's credit data in a third-party resource service can also be identified from the user's credit data, such as the user's credit data in a third-party resource service; the third-party resource service here can be a third-party resource service that has an associated relationship with the resource service, such as a third-party resource service that has a competitive relationship with the resource service, that is, other resource services.

[0022] It should be supplemented that the above-mentioned step S202 can be replaced by obtaining the user's attribute data, the user's first behavior data in the resource service and / or the user's credit data, and forming a new implementation method with the other processing steps provided in this embodiment; or, step S202 can also be replaced by obtaining the user's attribute data and obtaining the user's first behavior data in the resource service, and when the acquisition result of the first behavior data is empty, obtaining the user's credit data, and forming a new implementation method with the other processing steps provided in this embodiment; step S202 can also be replaced by obtaining the user's attribute data, the user's first behavior data in the resource service and the user's second behavior data in the third-party resource service, and forming a new implementation method with the other processing steps provided in this embodiment.

[0023] It should also be added that the user's attribute data and the user's first behavior data in the resource service can be extracted from the user's portrait data; in addition to the user's date of birth and occupation, the user's attribute data can also include the user's interest preferences, consumption capacity and / or the user's preferred shopping time period. In addition, the user's attribute data can also be other types of attribute data.

[0024] Step S204: predicting the user's behavior of third-party resource services based on the credit data to obtain second behavior data.

[0025] After obtaining the user's attribute data, the user's first behavior data in the resource service, and the user's credit data, in this step, starting from the user's credit data, the user's second behavior data in the third-party resource service is identified to assist in the subsequent identification of the user's service status in the resource service.

[0026] The third-party resource service in this embodiment may be a third-party resource service that is associated with the resource service, such as a third-party resource service that is in competition with the resource service, that is, other resource services. The second behavior data may include the user's behavior data in the third-party resource service, and specifically may include the user's resource behavior data in the third-party resource service. For example, the second behavior data may be the user's resource application record in the third-party resource service. The resource application record may include the resource application time, the resource application amount, and / or whether the resource return is overdue, that is, it may be one or more resource application times, resource application amounts, and / or whether the resource return is overdue. In addition, the second behavior data may also include other types of behavior data, such as the user's browsing data in the third-party resource service.

[0027] In actual applications, since resource services may compete with third-party resource services, it may be difficult to directly obtain the user's second behavior data in third-party resource services. In this regard, in order to ensure the rationality of obtaining the user's second behavior data in third-party resource services and ensure the accuracy of the second behavior data; this embodiment can predict the user's behavior in third-party resource services based on the user's credit data to obtain the second behavior data. Specifically, the user's second behavior data in third-party resource services can be predicted based on the user's credit data.

[0028] In a specific implementation, in the process of predicting the user's behavior in a third-party resource service based on the credit data and obtaining the second behavior data, the user's credit data in the third-party resource service can be extracted from the user's credit data, and the second behavior data of the user in the third-party resource service can be determined based on the extracted credit data; In addition, a prediction model can also be introduced to input the user's credit data into the prediction model for behavior prediction, and obtain the user's second behavior data in third-party resource services, so as to improve the accuracy and efficiency of behavior prediction.

[0029] It should be added that, based on the above-mentioned failure to obtain the user's credit data, step S204 may not be executed, and the following step S206 may be directly executed or an alternative execution method of step S206 may be executed according to actual needs.

[0030] Step S206: Based on the attribute data, the first behavior data and the second behavior data, the service status of the user is identified to obtain the service status of the user in the resource service.

[0031] The above-mentioned method starts from the user's credit data, predicts the user's behavior in the third-party resource service, and obtains the user's second behavior data in the third-party resource service. In this step, the user's service status in the resource service is identified based on the user's attribute data, the first behavior data and the second behavior data.

[0032] The service status in this embodiment may include service progress or service nodes. For example, the service status of a user in a resource service may be a customer acquisition status, a growth status, a mature status, a decline status, or a loss status. For another example, the service status of a user in a resource service may be a cognitive status, a consideration status, a purchase status, or a trustworthy status.

[0033] In actual applications, users may have different usage habits for resource services and third-party resource services. If the user's service status in the resource service is directly identified based on the user's first behavior data in the resource service, it may not be possible to accurately identify the user's service status based on the user's demand for services of this type. For example, based on the user's first behavior data in resource service a, it is identified that the user's service status in resource service a is in a declining state, but the user's second behavior data in resource service b shows that the user is more active in resource service b. In this case, if the user's service status in resource service a is directly classified as a declining state, since the user has a high demand for services of this type, it may cause the user to leave resource service a. In view of this, in order to improve the flexibility and comprehensiveness of service status identification, the first optional implementation mode provided in this embodiment is to identify the service status of the user based on the attribute data, the first behavior data and the second behavior data, and obtain the service status of the user in the resource service. The first resource application data of the user in the resource service can be calculated according to the first behavior data, and the second resource application data of the user in the third-party resource service can be calculated according to the second behavior data. The service status of the user in the resource service is identified based on the user's attribute data, the first resource application data and the second resource application data. Specifically, the following operations can be performed: Calculating first resource application data of the user in the resource service according to the first behavior data, and calculating second resource application data of the user in the third-party resource service according to the second behavior data; Determining a candidate service state in a service state sequence of resource services based on the attribute data and the first resource application data; If the second resource application data meets the threshold condition of the adjacent service state of the candidate service state, the adjacent service state is used as the service state.

[0034] Among them, the first resource application data of the user in the resource service may include the total resource amount, total number of applications, application frequency and / or data on whether the resource return is overdue for the user's resource application in the resource service; the second resource application data of the user in the third-party resource service may include the total resource amount, total number of applications, application frequency and / or number of overdue times for the user's resource application in the third-party resource service. The adjacent service state of the candidate service state may be the previous or previous n service states of the candidate service state in the service state sequence of the resource service, or the next or next m service states of the candidate service state in the service state sequence; n and m may be any positive integers, and n and m may be the same or different. The threshold conditions of the adjacent service states may include the resource application thresholds of the adjacent service states, such as the total resource amount threshold, the total number of applications threshold, the application frequency threshold and / or the overdue times threshold.

[0035] Specifically, in the process of determining the candidate service state in the service state sequence of the resource service based on the attribute data and the first resource application data, the attribute data and the first resource application data can be matched with each service state in the service state sequence to obtain the candidate service state; or the service state sequence of the resource service can be updated according to the attribute data to obtain an updated service state sequence, and the first resource application data can be matched with the threshold conditions of each service state in the updated service state sequence to obtain the candidate service state.

[0036] For example, the service status sequence of a resource service is customer acquisition status, growth status, mature status, decline status, and churn status. If the user category is identified as an elderly user category based on the attribute data, a new service status is added to the service status sequence, and the updated service status sequence is obtained as customer acquisition status, initial growth status, growth status, mature status, decline status, and churn status. It is determined whether the total resource amount, total number of applications, application frequency, and / or number of overdue applications in the first resource application data trigger the threshold conditions for each service status in the updated service status sequence, and the service status corresponding to the threshold conditions triggered by the first resource application data is used as the service status of the user in the resource service.

[0037] On this basis, in order to further improve the user experience, the threshold conditions can be dynamically adjusted for users with higher resource application requirements, so that the threshold conditions of the service status can be more adapted to each user; in an optional implementation manner provided by this embodiment, after the adjacent service status is executed as the service status, the following operations are further performed: If the second resource application data satisfies the second threshold condition of the adjacent service state, the threshold conditions of the remaining service states after the adjacent service state in the service state sequence of the resource service are updated for the user.

[0038] Among them, the total resource amount threshold in the second threshold condition may be greater than the total resource amount threshold in the threshold condition of the adjacent service state, the total application number threshold may be greater than the total application number threshold in the threshold condition of the adjacent service state, the application frequency threshold may be greater than the application frequency threshold in the threshold condition of the adjacent service state, and the overdue number threshold may be less than the overdue number threshold in the threshold condition of the adjacent service state.

[0039] Specifically, the total resource amount threshold in the threshold conditions of the remaining service status may be lowered, the total application number threshold may be lowered, the application frequency threshold may be lowered, and / or the overdue number threshold may be raised; thereby, when the second resource application data meets the second threshold condition, it represents that the user has a high demand for resource application, so the threshold conditions of the remaining service status may be dynamically updated to make the threshold conditions of the remaining service status more adapted to the user's demand for resource application.

[0040] In actual application scenarios, users belong to different user groups, and their demands for services such as resource services may be different. In order to meet the diverse demands of different users for services such as resource services, the accuracy of the service status of the identified user in the resource service is improved. In the second optional implementation mode provided in this embodiment, in the process of identifying the service status of the user based on the attribute data, the first behavior data, and the second behavior data, and obtaining the service status of the user in the resource service, the following operations are performed: Determine the user group to which the user belongs based on the attribute data, and read the service status sequence of the resource service corresponding to the user group; The service state is filtered out in the service state sequence based on the first behavior information and the second behavior information.

[0041] Among them, the user group to which the user belongs may include the user group to which the user belongs, for example, the user group may be an elderly user group or a young user group, for example, the user group may be a college user group, a white-collar user group or a blue-collar user group, for example, the user group may be a female user group or a male user group, the user group may be a user group characterized by multiple dimensions, and this embodiment does not specifically limit the number of multiple dimensions, for example, the multiple dimensions may be a user group characterized by a date of birth dimension, an occupation dimension and / or a male or female dimension - an elderly, female, white-collar user group or a youth, male, blue-collar user group.

[0042] Specifically, in the process of filtering out service states in a service state sequence based on the first behavior information and the second behavior information, the first resource application data of the user in the resource service can be calculated according to the first behavior data, and the second resource application data of the user in the third-party resource service can be calculated according to the second behavior data, and a candidate service state is determined in the service state sequence of the resource service according to the first resource application data; if the second resource application data meets the threshold condition of an adjacent service state of the candidate service state, the adjacent service state is used as the service state; in the process of determining the candidate service state in the service state sequence of the resource service according to the first resource application data here, the service state corresponding to the threshold condition satisfied by the first resource application data can be filtered out from the service state sequence of the resource service as a candidate service state.

[0043] On this basis, since different user groups have different requirements for resource services, the service status sequence of the resource service corresponding to the user can be flexibly adjusted; in an optional implementation provided by this embodiment, in the process of reading the service status sequence of the resource service corresponding to the user group, the following operations are performed: If the user group is the first user group, the service states in the preset service state sequence of the resource service are eliminated according to the number of intervals to obtain a service state sequence; If the user group is the second user group, the preset service state sequence is used as the service state sequence.

[0044] Among them, the second user group may include an elderly user group; it should be noted that the elderly user group that the second user group may include here may be a user group containing elderly user characteristics, such as an elderly user group, and another example is an elderly, female, and white-collar user group, and another example is an elderly, male user group. These user groups containing elderly characteristics can all represent elderly user groups. The number of intervals may include the number of intervals for elimination, such as eliminating service states at intervals of x in a preset service state sequence. The first user group may be a user group that does not contain elderly user characteristics, that is, a non-elderly user group, such as a female and white-collar user group, and another example is a youth and male user group, and another example is a youth and blue-collar user group. These can all be non-elderly user groups that do not contain elderly user characteristics, among which youth and blue-collar can represent user characteristics.

[0045] Specifically, in the process of eliminating service states in a preset service state sequence of a resource service according to the number of intervals to obtain a target service state sequence as a service state sequence, the number of intervals can be calculated based on the number of features and / or feature types of user features included in the first user group to which the user belongs, and the service states in the preset service state sequence of the resource service are eliminated according to the number of intervals to obtain a service state sequence; specifically, the number of features and / or feature types can be input into a number calculation model to calculate the number of intervals to obtain the number of intervals.

[0046] For example, the first user group is a female, white-collar, and youth user group. The first user group includes the three user characteristics of female, white-collar, and youth, so the number of characteristics is 3. The characteristic types include whether the user is male or female, occupation type, and birth time type. Assuming the number of intervals is 1, then in the preset service status sequence of the resource service, the service status is eliminated every interval. In the preset service status sequence of "customer acquisition status, growth status, maturity status, decline status, and churn status", the growth status and decline status are eliminated, and the service status sequence obtained is "customer acquisition status, maturity status, churn status".

[0047] In addition, in actual applications, in the process of obtaining the first behavior data, the result of obtaining the first behavior data may be empty, that is, the user does not have behavior data in the resource service. In this case, in order to include users who do not have behavior data in the resource service into the resource service and make them registered users of the resource service, thereby increasing the user usage of the resource service, and at the same time being able to recommend processing to users in the future based on the user's demand for services of this type; in the third optional implementation manner provided by this embodiment, in the process of identifying the service status of the user based on the attribute data, the first behavior data and the second behavior data, and obtaining the service status of the user in the resource service, the following operations are performed: If the data of the first behavior is empty, the resource application data of the user in the third-party resource service is calculated based on the data of the second behavior; A service state is determined in a sequence of service states for a resource service based on the attribute data and the resource request data.

[0048] Specifically, in the process of determining the service status in the service status sequence of the resource service based on the attribute data and the resource application data, the user group to which the user belongs can be determined based on the attribute data, and the service status sequence of the resource service corresponding to the user group can be read, and the service status corresponding to the threshold condition satisfied by the resource application data can be screened out from the service status sequence of the resource service as the service status of the user in the resource service; reading the service status sequence of the resource service corresponding to the user group here is similar to the execution process mentioned above, and you can refer to it for reference and will not repeat it here.

[0049] In order to improve the efficiency of service status identification, optionally, based on the attribute data, the first behavior data and the second behavior data, the service status of the user is identified, and the service status operation of the user in the resource service is obtained and executed based on the identification engine.

[0050] It should be supplemented that step S206 can be replaced by identifying the service status of the user based on attribute data, first behavior data and / or second behavior data, obtaining the service status of the user in the resource service, and forming a new implementation method with other processing steps provided in this embodiment.

[0051] Step S208: Input the service status and the attribute data into a large language model to generate a recommendation strategy to obtain a recommendation strategy, and perform recommendation processing.

[0052] The above identifies the user's service status in resource services from three aspects: the user's attribute data, the first behavior data and the second behavior data. In this step, a recommendation strategy is generated based on the service status and attribute data through a large language model to obtain a recommendation strategy, and the recommendation is processed to the user.

[0053] The large language model in this embodiment may be a large language model with an LLM (Language Model) structure; the recommendation strategy may include a recommendation strategy for recommending resource services to users, for example, the recommendation strategy includes recommendation channels and / or service recommendation data.

[0054] In a first optional implementation manner provided by this embodiment, during the process of generating a recommendation strategy using a large language model, the following operations are performed: Determine the user group the user belongs to based on attribute data, and determine the recommendation channel based on the user group and service status; A service recommendation element is determined based on the service status, and service recommendation data corresponding to the service recommendation element is generated.

[0055] Among them, the recommendation channel may include a channel for recommendation processing, such as a recommendation channel may be a text message, a push message, a voice call and / or an email, and in addition, the recommendation channel may also be other types of recommendation channels. The service recommendation element may include an element for recommendation processing, such as a service recommendation element may be a recommendation copy and / or a recommendation benefit, the recommendation copy may be a recommendation copy that provides a basic introduction to the resource service, and the recommended benefit may be benefit information provided for the resource service, such as red envelopes and discount information; the service recommendation data corresponding to the service recommendation element may include recommendation data generated for the service recommendation element, such as a service recommendation copy for the resource service generated for the recommendation copy, and another example is the benefit recommendation data for the resource service generated for the recommendation benefit.

[0056] Specifically, in the process of determining recommendation channels based on user groups and service statuses, the user groups and service statuses may be matched with preset recommendation channels to obtain recommendation channels that match the user groups and service statuses; in the process of determining service recommendation elements based on the service statuses, if the service status is a first service status, the service recommendation elements may be determined to be basic copy recommendations; if the service status is a second service status, the service recommendation elements may be determined to be basic copy recommendations and equity recommendations; the first service status here may be a customer acquisition status, and the second service status may be a growth status, a mature status, a decline status, or a loss status.

[0057] In the process of generating service recommendation data corresponding to the service recommendation element, if the service recommendation element includes recommended rights, the rights recommendation parameters can be calculated according to the service status and user group, and the rights recommendation data can be generated according to the rights recommendation parameters; the rights recommendation parameters can be the red envelope amount or the full reduction ratio.

[0058] In addition, in actual applications, different user locations may lead to different network conditions of the user's terminal device. For example, when the user is indoors, he may be accustomed to using a broadband network, and when the user is outdoors, the terminal device can use data traffic. In the case of outdoor, voice calls may consume more traffic and are not suitable as a recommendation channel. In view of this, in order to avoid disturbing the user while recommending the service, and also to avoid the user spending extra fees for the service recommendation; in the second optional implementation mode provided by this embodiment, in the process of generating the recommendation strategy, the following operations are performed: Determine the recommendation channel for recommendation processing based on the user's location; A recommendation topic is determined based on the attribute data, and service recommendation data under the recommendation topic is generated according to the service status.

[0059] On this basis, in an optional implementation manner provided by this embodiment, in the process of determining a recommendation channel for recommendation processing according to the user location, the following operations are performed: If the location type of the user location is the first type, using the first recommendation channel as the recommendation channel; optionally, the first type includes an indoor type; If the location type is the second type, the second recommended channel is used as the recommended channel; optionally, the second type includes an outdoor type.

[0060] In specific implementation, on the basis that the above step S208 can be performed based on the recognition engine, in an optional implementation provided by this embodiment, the following operations are further performed: Compare the response results of the user group to the test recommendation strategy with the response results of the user group to the benchmark recommendation strategy; optionally, the test recommendation strategy is generated based on a large language model; Adjust the parameters of the recognition engine according to the comparison results.

[0061] The test recommendation strategy may be a recommendation strategy for testing. The benchmark recommendation strategy may be a recommendation strategy set manually. The test recommendation strategy and the benchmark recommendation strategy may be recommendation strategies generated for the same user or the same group of users.

[0062] It should be noted that the above-mentioned step S208 can be replaced by inputting the service status or attribute data into the large language model to generate a recommendation strategy to obtain a recommendation strategy, and perform recommendation processing, and form a new implementation method with other processing steps provided in this embodiment; or, step S208 can also be replaced by generating a recommendation strategy according to the service status and / or attribute data to obtain a recommendation strategy and perform recommendation processing, and form a new implementation method with other processing steps provided in this embodiment; or, it can also be replaced by generating a recommendation strategy according to the service status and / or attribute data to obtain a recommendation strategy; it should be noted that during the execution of this embodiment, each step can dynamically adjust the data required by the previous step as needed. For example, if step S208 generates a recommendation strategy according to the service status, that is, only the service status is used, and the service status identification in step S206 only uses the first behavior data and the second behavior data, then step S202 does not need to obtain the user's attribute data, that is, the data obtained in each step can be adjusted according to the needs of other steps.

[0063] To summarize, the one or more recommendation processing methods provided in this embodiment firstly obtain the user's attribute data, the user's first behavior data in the resource service and the user's credit data, then predict the user's behavior in the third-party resource service based on the credit data, and obtain the user's second behavior data in the third-party resource service, so as to achieve the security and rationality of obtaining the user's second behavior data in the third-party resource service; secondly, calculate the user's first resource application data in the resource service based on the first behavior data, and calculate the user's second resource application data in the third-party resource service based on the second behavior data, and determine the candidate service state in the service state sequence of the resource service based on the attribute data and the first resource application data. If the second resource application data meets the threshold condition of the adjacent service state of the candidate service state, the adjacent service state is used as the service state, thereby identifying the user's service state in the resource service by combining the user's first resource application data in the resource service and the user's second resource application data in the third-party resource service, thereby improving the comprehensiveness and accuracy of the identified user's service state; finally, the service state and attribute data are input into the large language model for recommendation strategy generation to obtain the recommendation strategy, and the recommendation processing is performed, so that the recommendation strategy generated by the service state with higher accuracy can be more suitable for the user, avoiding large deviations in the service state, resulting in incompatibility of the recommendation strategy, thereby causing user loss, and improving user retention rate.

[0064] It should be supplemented that, each optional implementation method and each feasible execution method in steps S202 to S208 provided in this embodiment can be executed independently as needed, or can be combined and referenced to each other. At the same time, each specific execution step in each optional implementation method or each feasible execution method can also be executed independently or combined as needed. This embodiment does not make specific limitations on this.

[0065] The following uses the application of a recommendation processing method provided in this embodiment in a resource service scenario as an example to further illustrate the recommendation processing method provided in this embodiment. Figure 3 ,The recommended processing method applied to the resource service scenario specifically includes the following steps.

[0066] Step S302, obtaining the user's attribute data, the user's first behavior data in the resource service, and the user's credit data.

[0067] Step S304, predicting the user's behavior in the third-party resource service based on the credit data, and obtaining the user's second behavior data in the third-party resource service.

[0068] Step S306: calculating the first resource application data of the user in the resource service according to the first behavior data, and calculating the second resource application data of the user in the third-party resource service according to the second behavior data.

[0069] Step S308: determining a candidate service state in a service state sequence of the resource service based on the attribute data and the first resource application data.

[0070] Step S310: If the second resource application data meets the threshold condition of the adjacent service state of the candidate service state, the adjacent service state is used as the service state of the user in the resource service.

[0071] Step S312, input the service status and attribute data into the large language model to determine the recommendation channel for recommendation processing according to the user location, determine the recommendation topic based on the attribute data, and generate service recommendation data under the recommendation topic according to the service status.

[0072] Step S314: recommending service recommendation data to users according to the recommendation channel.

[0073] It should be noted that any one of steps S302 to S314 or any combination of multiple steps can be replaced by the corresponding technical means provided by the above steps S202 to S208 according to the needs of implementation and deployment, and any one of steps S302 to S314 or any combination of multiple steps can be combined into a new implementation method according to the needs of implementation and deployment; and any one of steps S302 to S314 or any combination of multiple steps can also form a new implementation method with one or more steps provided by the above steps S202 to S208 according to the needs of actual deployment, or form a new implementation method with one or more optional implementation methods provided by steps S202 to S208, which will not be repeated here.

[0074] An example of a recommended processing device provided in this specification is as follows: In the above-mentioned embodiment, a recommendation processing method is provided, and correspondingly, a recommendation processing device is also provided, which will be described below with reference to the accompanying drawings.

[0075] Reference Figure 4 , which shows a schematic diagram of an embodiment of a recommendation processing device provided in this embodiment.

[0076] Since the device embodiment corresponds to the method embodiment, the description is relatively simple, and the relevant parts can refer to the corresponding description of the method embodiment provided above. The device embodiment described below is only illustrative.

[0077] This embodiment provides a recommendation processing device, including: The data acquisition module 402 is configured to acquire the attribute data of the user, the first behavior data of the user in the resource service and the credit data of the user; The behavior prediction module 404 is configured to predict the user's behavior of a third-party resource service according to the credit data to obtain second behavior data; A state identification module 406 is configured to identify the service state of the user based on the attribute data, the first behavior data and the second behavior data, and obtain the service state of the user in the resource service; The strategy generation module 408 is configured to input the service status and the attribute data into a large language model to generate a recommendation strategy to obtain a recommendation strategy, and perform recommendation processing.

[0078] An example of a recommended processing device provided in this specification is as follows: Corresponding to the recommendation processing method described above, based on the same technical concept, one or more embodiments of this specification further provide a recommendation processing device, which is used to execute the recommendation processing method provided above. Figure 5 A schematic diagram of the structure of a recommended processing device provided for one or more embodiments of this specification.

[0079] This embodiment provides a recommendation processing device, including: like Figure 5 As shown, the recommendation processing device may have relatively large differences due to different configurations or performances, and may include one or more processors 501 and memory 502, and the memory 502 may store one or more storage applications or data. Among them, the memory 502 can be a short-term storage or a persistent storage. The application stored in the memory 502 may include one or more modules (not shown in the figure), and each module may include a series of computer executable instructions in the recommendation processing device. Furthermore, the processor 501 can be configured to communicate with the memory 502 and execute a series of computer executable instructions in the memory 502 on the recommendation processing device. The recommendation processing device may also include one or more power supplies 503, one or more wired or wireless network interfaces 504, one or more input / output interfaces 505, one or more keyboards 506, etc.

[0080] In a specific embodiment, the recommendation processing device includes a memory and one or more programs, wherein the one or more programs are stored in the memory, and the one or more programs may include one or more modules, and each module may include a series of computer executable instructions for the recommendation processing device, and the one or more programs are configured to be executed by one or more processors, including computer executable instructions for performing the following: Acquire attribute data of a user, first behavior data of the user in a resource service, and credit data of the user; Predicting the user's behavior of providing a third-party resource service based on the credit data to obtain second behavior data; Based on the attribute data, the first behavior data and the second behavior data, the service status of the user is identified to obtain the service status of the user in the resource service; The service status and the attribute data are input into a large language model to generate a recommendation strategy to obtain a recommendation strategy, and then a recommendation process is performed.

[0081] An embodiment of a computer-readable storage medium provided in this specification is as follows: Corresponding to the recommended processing method described above, based on the same technical concept, one or more embodiments of this specification also provide a computer-readable storage medium.

[0082] The computer-readable storage medium provided in this embodiment is used to store computer-executable instructions, and the computer-executable instructions implement the following steps when executed: Acquire attribute data of a user, first behavior data of the user in a resource service, and credit data of the user; Predicting the user's behavior of providing a third-party resource service based on the credit data to obtain second behavior data; Based on the attribute data, the first behavior data and the second behavior data, the service status of the user is identified to obtain the service status of the user in the resource service; The service status and the attribute data are input into a large language model to generate a recommendation strategy to obtain a recommendation strategy, and then a recommendation process is performed.

[0083] It should be noted that the embodiment of a computer-readable storage medium in this specification and the embodiment of a recommended processing method in this specification are based on the same inventive concept, so the specific implementation of this embodiment can refer to the implementation of the aforementioned corresponding method, and the repeated parts will not be repeated.

[0084] An embodiment of a computer program product provided in this specification is as follows: Corresponding to the recommendation processing method described above, based on the same technical concept, one or more embodiments of this specification also provide a computer program product.

[0085] A computer program product, comprising a computer program / instruction, which, when executed by a processor, implements the following steps: Acquire attribute data of a user, first behavior data of the user in a resource service, and credit data of the user; Predicting the user's behavior of providing a third-party resource service based on the credit data to obtain second behavior data; Based on the attribute data, the first behavior data and the second behavior data, the service status of the user is identified to obtain the service status of the user in the resource service; The service status and the attribute data are input into a large language model to generate a recommendation strategy to obtain a recommendation strategy, and then a recommendation process is performed.

[0086] It should be noted that an embodiment of a computer program product in this specification and an embodiment of a recommended processing method in this specification are based on the same inventive concept, so the specific implementation of this embodiment can refer to the implementation of the aforementioned corresponding method, and the repeated parts will not be repeated.

[0087] Each embodiment in this specification is described in a progressive manner, and the same or similar parts between the embodiments can be referenced to each other. Each embodiment focuses on the differences from other embodiments. For example, the device embodiment, equipment embodiment and computer-readable storage medium embodiment are similar to the method embodiment, so the description is relatively simple. To read the relevant contents in the device embodiment, equipment embodiment and computer-readable storage medium embodiment, please refer to the partial description of the method embodiment.

[0088] The above is a description of a specific embodiment of the specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims can be performed in an order different from that in the embodiments and still achieve the desired results. In addition, the processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0089] In the 1930s, improvements to a technology could be clearly distinguished as hardware improvements (for example, improvements to circuit structures such as diodes, transistors, switches, etc.) or software improvements (improvements to method flows). However, with the development of technology, many of today's method flow improvements can be regarded as direct improvements to hardware circuit structures. Designers almost always obtain the corresponding hardware circuit structure by programming the improved method flow into the hardware circuit. Therefore, it cannot be said that an improvement in a method flow cannot be implemented using hardware entity modules. For example, a programmable logic device (PLD) (such as a field programmable gate array (FPGA)) is such an integrated circuit whose logical function is determined by the user's programming of the device. Designers can "integrate" a digital system on a PLD by programming it themselves, without having to ask chip manufacturers to design and produce dedicated integrated circuit chips. Moreover, nowadays, instead of manually making integrated circuit chips, this kind of programming is mostly implemented by "logic compiler" software, which is similar to the software compiler used when developing and writing programs, and the original code before compilation must also be written in a specific programming language, which is called hardware description language (HDL). There is not only one kind of HDL, but many kinds, such as ABEL (Advanced Boolean Expression Language), AHDL (Altera Hardware Description Language), Confluence, CUPL (Cornell University Programming Language), HDCal, JHDL (Java Hardware Description Language), Lava, Lola, MyHDL, PALASM, RHDL (Ruby Hardware Description Language), etc. The most commonly used ones are VHDL (Very-High-Speed ​​Integrated Circuit Hardware Description Language) and Verilog. Those skilled in the art should also know that it is only necessary to program the method flow slightly in the above-mentioned hardware description languages ​​and program it into the integrated circuit, and then it is easy to obtain the hardware circuit that implements the logic method flow.

[0090] The controller may be implemented in any suitable manner, for example, the controller may take the form of a microprocessor or processor and a computer-readable medium storing a computer-readable program code (e.g., software or firmware) executable by the (micro)processor, a logic gate, a switch, an application-specific integrated circuit (ASIC), a programmable logic controller, and an embedded microcontroller, examples of which include but are not limited to the following microcontrollers: ARC 625D, Atmel AT91SAM, Microchip PIC18F26K20, and Silicone Labs C8051F320, and the memory controller may also be implemented as part of the control logic of the memory. It is also known to those skilled in the art that, in addition to implementing the controller in a purely computer-readable program code manner, the controller may be implemented in the form of a logic gate, a switch, an application-specific integrated circuit, a programmable logic controller, and an embedded microcontroller by logically programming the method steps. Therefore, such a controller may be considered as a hardware component, and the devices for implementing various functions included therein may also be considered as structures within the hardware component. Or even, the devices for implementing various functions may be considered as both software modules for implementing the method and structures within the hardware component.

[0091] The systems, devices, modules or units described in the above embodiments may be implemented by computer chips or entities, or by products with certain functions. A typical implementation device is a computer. Specifically, the computer may be, for example, a personal computer, a laptop computer, a cellular phone, a camera phone, a smart phone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or a combination of any of these devices.

[0092] For the convenience of description, the above devices are described in terms of functions and are divided into various units. Of course, when implementing the embodiments of this specification, the functions of each unit can be implemented in the same or multiple software and / or hardware.

[0093] It should be understood by those skilled in the art that one or more embodiments of the present specification may be provided as a method, system or computer program product. Therefore, one or more embodiments of the present specification may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present specification may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0094] This specification is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of this specification. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable test processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable test processing device generate instructions for implementing the process in the flowchart. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0095] These computer program instructions may also be stored in a computer readable memory capable of directing a computer or other programmable test processing device to operate in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0096] These computer program instructions may also be loaded onto a computer or other programmable test processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, whereby the instructions executed on the computer or other programmable device provide for implementing the process described in the flow. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0097] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0098] Memory may include non-permanent storage in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. Memory is an example of a computer-readable medium.

[0099] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined in this article, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0100] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of features includes not only those features, but also other features not explicitly listed, or also includes features inherent to such process, method, commodity or device. In the absence of further restrictions, features defined by the phrase "including a ..." do not exclude the existence of other identical features in the process, method, commodity or device including the features.

[0101] One or more embodiments of the present specification may be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. One or more embodiments of the present specification may also be practiced in distributed computing environments where tasks are performed by remote processing devices connected through a communication network. In a distributed computing environment, program modules may be located in local and remote computer storage media, including storage devices.

[0102] Each embodiment in this specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0103] The above description is only an embodiment of this document and is not intended to limit this document. For those skilled in the art, this document may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of this document should be included in the scope of the claims of this document.

Claims

1. A recommended processing method, comprising: Acquire attribute data of a user, first behavior data of the user in a resource service, and credit data of the user; Predicting the user's behavior of providing a third-party resource service based on the credit data to obtain second behavior data; Based on the attribute data, the first behavior data and the second behavior data, the service status of the user is identified to obtain the service status of the user in the resource service; The service status and the attribute data are input into a large language model to generate a recommendation strategy to obtain a recommendation strategy, and then a recommendation process is performed.

2. The recommendation processing method according to claim 1, wherein the identifying the service status of the user based on the attribute data, the first behavior data and the second behavior data to obtain the service status of the user in the resource service comprises: Calculate the first resource application data of the user in the resource service according to the first behavior data, and calculate the second resource application data of the user in the third-party resource service according to the second behavior data; determining a candidate service state in a service state sequence of the resource service based on the attribute data and the first resource application data; If the second resource application data meets the threshold condition of the adjacent service state of the candidate service state, the adjacent service state is used as the service state.

3. The recommendation processing method according to claim 2, after performing the operation of using the adjacent service status as the service status, further comprising: If the second resource application data satisfies the second threshold condition of the adjacent service state, the threshold conditions of the remaining service states after the adjacent service state in the service state sequence of the resource service are updated for the user.

4. The recommendation processing method according to claim 1, wherein generating a recommendation strategy comprises: Determine the user group to which the user belongs according to the attribute data, and determine a recommendation channel based on the user group and the service status; A service recommendation element is determined based on the service status, and service recommendation data corresponding to the service recommendation element is generated.

5. The recommendation processing method according to claim 1, wherein generating a recommendation strategy comprises: Determine the recommendation channel for recommendation processing based on the user's location; A recommended topic is determined based on the attribute data, and service recommendation data under the recommended topic is generated according to the service status.

6. The recommendation processing method according to claim 5, wherein determining a recommendation channel for recommendation processing according to the user location comprises: If the location type of the user location is the first type, using the first recommendation channel as the recommendation channel; The first type includes an indoor type; If the location type is the second type, the second recommended channel is used as the recommended channel; the second type includes an outdoor type.

7. The recommendation processing method according to claim 1, wherein the step of identifying the service status of the user based on the attribute data, the first behavior data and the second behavior data, and obtaining the service status of the user in the resource service is performed based on a recognition engine; The method further comprises: Comparing the response result of the user group to the test recommendation strategy with the response result of the user group to the benchmark recommendation strategy; The test recommendation strategy is generated based on a large language model; The parameters of the recognition engine are adjusted according to the comparison results.

8. The recommendation processing method according to claim 1, wherein the identifying the service status of the user based on the attribute data, the first behavior data and the second behavior data to obtain the service status of the user in the resource service comprises: Determine the user group to which the user belongs according to the attribute data, and read the service status sequence of the resource service corresponding to the user group; The service state is screened out from the service state sequence based on the first behavior information and the second behavior information.

9. The recommendation processing method according to claim 8, wherein the step of reading the service status sequence of the resource service corresponding to the user group comprises: If the user group is the first user group, eliminating the service states in the preset service state sequence of the resource service according to the number of intervals to obtain the service state sequence; If the user group is the second user group, the preset service state sequence is used as the service state sequence.

10. The recommendation processing method according to claim 1, wherein the identifying the service status of the user based on the attribute data, the first behavior data and the second behavior data to obtain the service status of the user in the resource service comprises: If the first behavior data is empty, calculating the resource application data of the user in the third-party resource service based on the second behavior data; The service state is determined in a service state sequence of the resource service according to the attribute data and the resource application data.

11. The recommendation processing method according to claim 1, wherein the credit data is obtained by: Obtain the credit data of the user on the third-party platform; and / or obtain the application materials submitted by the user during the resource application process of the resource service, and extract the credit data of the user from the application materials.

12. A recommendation processing device, comprising: A data acquisition module, configured to acquire attribute data of a user, first behavior data of the user in a resource service, and credit data of the user; a behavior prediction module, configured to predict the user's behavior of a third-party resource service according to the credit data to obtain second behavior data; A state identification module is configured to identify the service state of the user based on the attribute data, the first behavior data and the second behavior data, and obtain the service state of the user in the resource service; The strategy generation module is configured to input the service status and the attribute data into a large language model to generate a recommendation strategy to obtain a recommendation strategy, and perform recommendation processing.

13. A recommendation processing device, comprising: processor; and a memory configured to store computer executable instructions that, when executed, cause the processor to: Acquire attribute data of a user, first behavior data of the user in a resource service, and credit data of the user; Predicting the user's behavior of providing a third-party resource service based on the credit data to obtain second behavior data; Based on the attribute data, the first behavior data and the second behavior data, the service status of the user is identified to obtain the service status of the user in the resource service; The service status and the attribute data are input into a large language model to generate a recommendation strategy to obtain a recommendation strategy, and then a recommendation process is performed.

14. A computer-readable storage medium for storing computer-executable instructions, wherein the computer-executable instructions implement the steps of the method of claim 1 when executed.

Images