Data Processing Method and Related Equipment

Abstract

An embodiment of the present disclosure provides a data processing method and related devices. The method includes: obtaining a first set of resources to be distributed that has a distribution association with a first set of account objects, and a second set of resources to be distributed that has a distribution association with a second set of account objects; distributing the first resources to be distributed in the first set of resources to be distributed to a first account object in the first set of account objects to obtain first attention index information of the first account object for the first resources to be distributed; determining first target resources to be distributed according to the first attention index information; distributing the first target resources to be distributed to some second account objects in the second set of account objects to obtain second attention index information of the some second account objects for the first target resources to be distributed; and adding the first target resources to be distributed to the second set of resources to be distributed when the second attention index information meets a preset condition. The method improves the diversity of the content ecosystem and realizes content circulation.

Description

Technical Field

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

[0002] Industrial-level recommendation systems in related technologies usually have the problem of overly strong positive feedback, which easily leads to the "information cocoon" problem for users, that is, the content recommended to users is too similar to the content they liked in the past. The information cocoon may limit the distribution ability of the recommendation system, resulting in user loss and author loss.

[0003] It should be noted that the information disclosed in the above background art section is only used to enhance the understanding of the background of the present disclosure, and thus may include information that does not constitute the prior art known to those of ordinary skill in the art. Summary of the Invention

[0004] Embodiments of the present disclosure provide a data processing method, a data processing device, an electronic device, a computer-readable storage medium, and a computer program product. The method improves the diversity of the content ecosystem, realizes content circulation, avoids the information cocoon problem in related technologies, improves user interest, and increases user stickiness.

[0005] Embodiments of the present disclosure provide a data processing method, the method comprising: obtaining a first set of resources to be distributed associated with a first set of account objects, and a second set of resources to be distributed associated with a second set of account objects; distributing a first resource to be distributed in the first set of resources to be distributed to a first account object in the first set of account objects to obtain first attention index information of the first account object for the first resource to be distributed; determining a first target resource to be distributed from the first set of resources to be distributed according to the first attention index information of the first account object for the first resource to be distributed; distributing the first target resource to be distributed to some second account objects in the second set of account objects to obtain second attention index information of the some second account objects for the first target resource to be distributed; and adding the first target resource to be distributed to the second set of resources to be distributed to update the second set of resources to be distributed when the second attention index information meets a preset condition.

[0006] In some exemplary embodiments of the present disclosure, the step of determining a first target resource to be distributed from the first resources to be distributed according to the first account object's first attention metric information for the first resources to be distributed includes: determining, as the first target resource to be distributed, the first resources to be distributed for which the first attention metric information is greater than a first attention metric information threshold; or, sorting the first attention metric information in descending order, and determining, as the first target resources to be distributed, the first resources to be distributed corresponding to the first preset number of the first attention metric information at the front; or, sorting the first attention metric information in descending order, and determining, as the first target resources to be distributed, the first resources to be distributed corresponding to the first preset proportion of the first attention metric information at the front.

[0007] In some exemplary embodiments of the present disclosure, it is characterized in that the step of adding the first target resource to be distributed to the second set of resources to be distributed to update the second set of resources to be distributed when the second attention metric information meets a preset condition includes: if the second attention metric information is greater than a second attention metric threshold, adding the first target resource to be distributed to the second set of resources to be distributed to update the second set of resources to be distributed.

[0008] In some exemplary embodiments of the present disclosure, the method further includes: obtaining third attention metric information of each second account object in the partial second account objects for the first target resource to be distributed; determining second target account objects from the partial second account objects according to the third attention metric information; and adding the second target account objects to the first account object set to update the first account object set.

[0009] In some exemplary embodiments of the present disclosure, the step of determining second target account objects from the partial second account objects according to the third attention metric information includes: determining, as the second target account objects, the second account objects for which the third attention metric information is greater than a third attention metric threshold; or, sorting the third attention metric information in descending order, and determining, as the second target account objects, the second account objects corresponding to the second preset number of the third attention metric information at the front; or, sorting the third attention metric in descending order, and determining, as the second target account objects, the second account objects corresponding to the second preset proportion of the third attention metric at the front.

[0010] In some exemplary embodiments of the present disclosure, when obtaining a first set of resources to be distributed associated with a first account object set and a second set of resources to be distributed associated with a second account object set, the first resources to be distributed are not distributed to the second account objects, and the second resources to be distributed are not distributed to the first account objects.

[0011] In some exemplary embodiments of the present disclosure, the step of obtaining a first set of resources to be distributed that has a distribution association with a first set of account objects includes: determining an initial set of account objects according to the similarity between account objects, where the initial set of account objects includes the first set of account objects; determining an initial set of resources to be distributed according to the similarity between resources to be distributed; and determining the first set of resources to be distributed that has a distribution association with the first set of account objects from the initial set of resources to be distributed according to the initial attention metrics of the initial set of account objects to the initial set of resources to be distributed.

[0012] In some exemplary embodiments of the present disclosure, the above method further includes: when distributing a first resource to be distributed in the first set of resources to be distributed to a first account object in the first set of account objects, randomly generating a first uncertainty parameter and a second uncertainty parameter, where the first uncertainty parameter is less than the second uncertainty parameter; obtaining the first attention metric information according to the first uncertainty parameter and the second uncertainty parameter; when distributing the first target resource to be distributed to some second account objects in the second set of account objects, adjusting the first uncertainty parameter and the second uncertainty parameter so that the first uncertainty parameter is greater than the second uncertainty parameter; and obtaining the second attention metric information according to the adjusted first uncertainty parameter and the second uncertainty parameter.

[0013] An embodiment of the present disclosure provides a data processing device, including: an obtaining module configured to obtain a first set of resources to be distributed that has a distribution association with a first set of account objects, and a second set of resources to be distributed that has a distribution association with a second set of account objects; a distributing module configured to distribute a first resource to be distributed in the first set of resources to be distributed to a first account object in the first set of account objects to obtain first attention metric information of the first account object to the first resource to be distributed; a determining module configured to determine a first target resource to be distributed from the first set of resources to be distributed according to the first attention metric information of the first account object to the first resource to be distributed; the distributing module is further configured to distribute the first target resource to be distributed to some second account objects in the second set of account objects to obtain second attention metric information of the some second account objects to the first target resource to be distributed; the determining module is further configured to add the first target resource to be distributed to the second set of resources to be distributed to update the second set of resources to be distributed when the second attention metric information meets a preset condition.

[0014] In some exemplary embodiments of the present disclosure, the determining module is further configured to determine a first resource to be distributed with the first attention metric information greater than a first attention metric information threshold as the first target resource to be distributed; or, the determining module is further configured to perform a descending order arrangement on the first attention metric information, and determine the first resources to be distributed corresponding to the first preset number of the first attention metric information as the first target resources to be distributed; or, the determining module is further configured to perform a descending order arrangement on the first attention metric information, and determine the first resources to be distributed corresponding to the first preset proportion of the first attention metric information as the first target resources to be distributed.

[0015] In some exemplary embodiments of the present disclosure, the determining module is further configured to perform, if the second attention metric information is greater than a second attention metric threshold, adding the first target resource to be distributed to the second set of resources to be distributed to update the second set of resources to be distributed.

[0016] In some exemplary embodiments of the present disclosure, the apparatus further includes: an obtaining module; the obtaining module is configured to obtain third attention metric information of each second account object in the partial second account objects with respect to the first target resource to be distributed; the determining module is further configured to determine a second target account object from the partial second account objects according to the third attention metric information; the determining module is further configured to add the second target account object to the first set of account objects to update the first set of account objects.

[0017] In some exemplary embodiments of the present disclosure, the determining module is further configured to determine a second account object with the third attention metric information greater than a third attention metric threshold as the second target account object; or, the determining module is further configured to perform a descending order arrangement on the third attention metric information, and determine the second account objects corresponding to the second preset number of the third attention metric information as the second target account objects; or, the determining module is further configured to perform a descending order arrangement on the third attention metric, and determine the second account objects corresponding to the second preset proportion of the third attention metric as the second target account objects.

[0018] In some exemplary embodiments of the present disclosure, when obtaining a first set of resources to be distributed having a distribution association with the first set of account objects and a second set of resources to be distributed having a distribution association with the second set of account objects, the first resources to be distributed are not distributed to the second account objects, and the second resources to be distributed are not distributed to the first account objects.

[0019] In some exemplary embodiments of the present disclosure, the determining module is further configured to determine an initial set of account objects according to the similarity between account objects, where the initial set of account objects includes the first set of account objects; the determining module is further configured to determine an initial set of resources to be distributed according to the similarity between the resources to be distributed; the determining module is further configured to determine, from the initial set of resources to be distributed, the first set of resources to be distributed that has a distribution association with the first set of account objects according to the initial attention metrics of the initial set of resources to be distributed with respect to the initial set of account objects.

[0020] In some exemplary embodiments of the present disclosure, the above device further includes: a generating module, an obtaining module, and an adjusting module; the generating module is configured to randomly generate a first uncertainty parameter and a second uncertainty parameter when distributing a first resource to be distributed in the first set of resources to be distributed to a first account object in the first set of account objects, where the first uncertainty parameter is less than the second uncertainty parameter; the obtaining module is configured to obtain the first attention metric information according to the first uncertainty parameter and the second uncertainty parameter; the adjusting module is configured to adjust the first uncertainty parameter and the second uncertainty parameter when distributing the first target resource to be distributed to some second account objects in the second set of account objects, so that the first uncertainty parameter is greater than the second uncertainty parameter; the obtaining module is further configured to obtain the second attention metric information according to the adjusted first uncertainty parameter and the second uncertainty parameter.

[0021] An embodiment of the present disclosure provides an electronic device, including: a processor; a memory for storing processor-executable instructions; wherein, the processor is configured to execute the executable instructions to implement the data processing method as described in any one of the above.

[0022] An embodiment of the present disclosure provides a computer-readable storage medium, when the instructions in the computer-readable storage medium are executed by the processor of the electronic device, enabling the electronic device to execute the data processing method as described in any one of the above.

[0023] An embodiment of the present disclosure provides a computer program product, a computer program / instructions, characterized in that when the computer program / instructions are executed by a processor, the data processing method as described in any one of the above is implemented.

[0024] Some embodiments of the present disclosure provide a data processing method. According to the first attention index information of the first resource to be distributed for the first account object, the first target resource to be distributed is determined from the first resource to be distributed. The first resource to be distributed that the first account object is interested in can be used as the first target resource to be distributed. Distributing the first target resource to be distributed to some second account objects in the second account object set only requires distributing the first target resource to be distributed to a small number of second account objects first, and then it can be tested whether the second account object set is interested in the first target resource to be distributed. Thus, the computational amount and storage space in the initial test stage can be reduced. Moreover, distributing the first target resource to be distributed to some second account objects in the second account object set can enable the first target resource to be distributed to more account objects to obtain more attention, and at the same time enable some second account objects to see more resources to be distributed, increasing the supply of target content on the supply side and enhancing the diversity of the content ecosystem. According to the second attention index of some second account objects to the first target resource to be distributed, it can be determined to add the first target resource to be distributed to the second resource to be distributed set, so that other account objects in the second account object set can see the first target resource to be distributed, realizing content circulation, avoiding the information cocoon problem in the related art, enhancing user interest, and increasing user stickiness.

[0025] It should be understood that the above general description and the following detailed description are only exemplary and explanatory, and cannot limit the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] The drawings herein are incorporated into the specification and constitute a part of this specification, showing embodiments consistent with the present disclosure and used together with the specification to explain the principles of the present disclosure. Obviously, the drawings in the following description are only some embodiments of the present disclosure, and those of ordinary skill in the art can obtain other drawings based on these drawings without creative efforts.

[0027] Figure 1 The figure shows a schematic diagram of the imbalance between the supply side and the utilization side in the related art.

[0028] Figure 2 The figure shows a schematic diagram of an exemplary system architecture to which the data processing method of the embodiments of the present disclosure can be applied.

[0029] Figure 3 It is a flowchart of a data processing method shown according to an exemplary embodiment.

[0030] Figure 4 It is a schematic diagram of a user group and a work group shown according to an example.

[0031] Figure 5It is a schematic diagram of another user group and work group shown according to an example.

[0032] Figure 6 It is a flowchart of another data processing method shown according to an exemplary embodiment.

[0033] Figure 7 It is a schematic diagram of a bilateral transfer mechanism shown according to an exemplary embodiment.

[0034] Figure 8 It is a block diagram of a data processing device shown according to an exemplary embodiment.

[0035] Figure 9 It is a schematic diagram of the structure of an electronic device suitable for implementing the exemplary embodiments of the present disclosure shown according to an exemplary embodiment. Detailed implementation manners

[0036] Example embodiments will now be described more fully with reference to the accompanying drawings. However, the example embodiments can be implemented in various forms and should not be construed as limited to the embodiments set forth herein; rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the concept of the example embodiments to those skilled in the art. Like reference numerals in the figures denote like or similar parts, and thus their repetitive description will be omitted.

[0037] The features, structures, or characteristics described in this disclosure can be combined in any suitable manner in one or more embodiments. In the following description, numerous specific details are provided to give a thorough understanding of the embodiments of this disclosure. However, those skilled in the art will realize that one or more of the specific details can be omitted in practicing the technical solutions of this disclosure, or other methods, components, devices, steps, etc. can be adopted. In other cases, well-known methods, devices, implementations, or operations are not shown or described in detail to avoid obscuring aspects of this disclosure.

[0038] The accompanying drawings are only schematic illustrations of this disclosure, and like reference numerals in the figures denote like or similar parts, and thus their repetitive description will be omitted. Some of the block diagrams shown in the drawings do not necessarily have to correspond to physically or logically independent entities. These functional entities can be implemented in software form, or in at least one hardware module or integrated circuit, or in different networks and / or processor devices and / or microcontroller devices.

[0039] The flowcharts shown in the accompanying drawings are only exemplary illustrations and do not necessarily include all the content and steps, nor do they necessarily have to be executed in the described order. For example, some steps can be decomposed, and some steps can be combined or partially combined, so the actual execution order may change according to the actual situation.

[0040] In this specification, the terms "a", "an", "the", "said", and "at least one" are used to indicate the existence of at least one element / component / etc.; the terms "comprising", "including", and "having" are used to mean an open inclusion and refer to the existence of additional elements / components / etc. in addition to the listed elements / components / etc.; the terms "first", "second", "third", etc. are only used as labels and are not a limitation on the quantity of their objects.

[0041] Figure 1 The schematic diagram showing the imbalance between the supply side and the utilization side in the related art is shown.

[0042] Industrial-grade recommendation systems in the related art usually have the problem of overly strong positive feedback, which easily leads to the problem of "information cocoons" for users, that is, the content recommended to users is too similar to the content they liked in the past. To a certain extent, information cocoons improve the accuracy of recommendation distribution, but they also limit the distribution ability of the recommendation system, leading to user loss and author loss: users may be lost because they feel bored or think there is little content on the platform; authors may have a decline in operation willingness or be lost because the distribution volume of their works and the efficiency of gaining followers are limited.

[0043] Reference Figure 1 , the exploration and exploitation algorithms in the related art have the problem that the growth of the supply side (exploration side) is not efficient enough, which restricts the growth of the consumption side (utilization side) to a certain extent, may lead to the imbalance of the E&E (Explore&Exploit, exploration and exploitation) mechanism of the system, and affects the development of new businesses, commercialization, live broadcast, e-commerce, upgc (User&Professional GeneratedContent, user and expert generated content), mid-video and other businesses.

[0044] The traffic of content exploration in the related art (such as cold start, ramp-up, etc.) acts on the system in a forced-insertion manner: that is, x% of the traffic is forcibly allocated to content exploration, where x can be a number between 0 and 100. The problems of how to reasonably allocate cold-start traffic and whether there is an optimal balance point of cold-start traffic have not been solved. At the same time, there is a lack of effective solutions for user interest exploration, and the benefits of improving the diversity and fragmentation of conventional interest exploration are limited.

[0045] In view of the above technical problems existing in the related art, the embodiments of the present disclosure provide a data processing method for at least solving one or all of the above technical problems.

[0046] Figure 2 The schematic diagram showing an exemplary system architecture to which the data processing method of the embodiments of the present disclosure can be applied is shown.

[0047] AsFigure 2 As shown in the figure, the system architecture may include a server 201, a network 202, and a terminal device 203. The network 202 is used to provide a medium for a communication link between the terminal device 203 and the server 201. The network 202 may include various connection types, such as wired, wireless communication links, or fiber optic cables, etc.

[0048] The server 201 may be a server that provides various services, such as a background management server that supports the devices operated by users using the terminal device 203. The background management server may analyze and process data such as received requests, and feedback the processing results to the terminal device.

[0049] For example, the server 201 may obtain a first set of resources to be distributed that has a distribution association with the first set of account objects, and a second set of resources to be distributed that has a distribution association with the second set of account objects; for example, the server 201 may distribute the first resources to be distributed in the first set of resources to be distributed to the first account objects in the first set of account objects to obtain first attention metric information of the first account objects for the first resources to be distributed; for example, the server 201 may determine a first target resource to be distributed from the first set of resources to be distributed according to the first attention metric information of the first account objects for the first resources to be distributed; for example, the server 201 may distribute the first target resource to be distributed to some of the second account objects in the second set of account objects to obtain second attention metric information of some of the second account objects for the first target resource to be distributed; for example, when the second attention metric information meets a preset condition, the server 201 adds the first target resource to be distributed to the second set of resources to be distributed to update the second set of resources to be distributed.

[0050] It should be understood that Figure 2 the numbers of the terminal devices, networks, and servers in the figure are merely illustrative. The server 201 may be a physical server, may also be a server cluster composed of multiple servers, or may also be a cloud server. According to actual needs, there may be any number of terminal devices, networks, and servers.

[0051] Next, each step of the data processing method in the exemplary embodiments of the present disclosure will be described in more detail with reference to the accompanying drawings and embodiments.

[0052] Figure 3 is a flowchart of a data processing method shown according to an exemplary embodiment.

[0053] As Figure 3 shown, the method provided by the embodiments of the present disclosure may include the following steps.

[0054] In step S310, obtain a first set of resources to be distributed that has a distribution association with the first set of account objects, and a second set of resources to be distributed that has a distribution association with the second set of account objects.

[0055] In the embodiments of the present disclosure, the account object may be a user, such as a user using a video playback software or a shopping software. The resources to be distributed may be works, goods, etc. The works may include, but are not limited to, video works, text works, picture works, etc. In the following illustrative examples, the account object is taken as a user and the resource to be distributed is taken as a work for illustration, but the present disclosure is not limited thereto.

[0056] In the embodiments of the present disclosure, the set of account objects may be a user group. At this time, the first set of account objects may also be referred to as the first user group, and the second set of account objects may also be referred to as the second user group. The set of resources to be distributed may be a group of works. At this time, the first set of resources to be distributed may also be referred to as the first group of works, and the second set of resources to be distributed may also be referred to as the second group of works.

[0057] In the embodiments of the present disclosure, a first group of works having a distribution association with the first user group and a second group of works having a distribution association with the second user group may be obtained.

[0058] It should be noted that in the embodiments of the present disclosure, two groups of works (i.e., the first group of works and the second group of works) and two user groups (i.e., the first user group and the second user group) are taken as examples for illustration. In actual applications, there may be multiple groups of works and multiple user groups, and the number of groups of works and user groups may be the same or different.

[0059] Figure 4 It is a schematic diagram of a user group and a group of works shown according to an example.

[0060] Refer to Figure 4, the content side (Item) may include work group 1 and work group 2. Work group 1 may be a similar new work group 1, and work group 2 may be a similar new work group 2; the user side (User) may include user group 1 and user group 2. User group 1 may be a similar old user group 1, and user group 2 may be a similar old user group 2. Among them, a similar work group may be a work group obtained by clustering based on the similarity of work characteristics, and a similar user group may be a user group obtained by clustering based on the similarity of work characteristics of the works that each user was interested in historically. Among them, it can be determined whether the user is interested in the work based on whether the user has browsed, clicked, collected, placed an order, etc. on the work historically; different weights can also be set for different behaviors of the user on the work, and it can be determined whether the user is interested in the work based on the user's behavior and its corresponding weight; the duration of different behaviors of the user on the work can also be recorded, and it can be determined whether the user is interested in the work based on whether there are the above behaviors, the weight and duration of each behavior, etc.

[0061] In the embodiments of the present disclosure, for a user, a new work may refer to a work that the user has not seen, and an old work may refer to a work that the user has seen; for a work, a new user may refer to a user who has not seen the work, and an old user may refer to a user who has seen the work.

[0062] In an exemplary embodiment, when obtaining a first set of resources to be distributed that has a distribution association with a first set of account objects, and a second set of resources to be distributed that has a distribution association with a second set of account objects, the first set of resources to be distributed is not distributed to the second set of account objects, and the second set of resources to be distributed is not distributed to the first set of account objects.

[0063] In the embodiments of the present disclosure, when performing step S310, the first work group is not distributed to the second user group, and the second work group is not distributed to the first user group, that is, the users in the second user group have not watched the works in the first work group, and the users in the first user group have not watched the works in the second work group.

[0064] In an exemplary embodiment, the step of obtaining a first set of resources to be distributed that has a distribution association with a first set of account objects may include: determining an initial set of account objects according to the similarity between account objects, where the initial set of account objects includes the first set of account objects; determining an initial set of resources to be distributed according to the similarity between resources to be distributed; and determining a first set of resources to be distributed that has a distribution association with the first set of account objects from the initial set of resources to be distributed according to the initial attention index of the initial set of account objects to the initial set of resources to be distributed.

[0065] Among them, the concerned metric can be relevance, which can be obtained based on the user behavior of the user. The user behavior of the user can include, but is not limited to, viewing, collecting, forwarding, following, etc. of the work.

[0066] In the embodiments of the present disclosure, multiple initial user groups can be determined according to the similarity between users; multiple initial work groups can be determined according to the similarity between works; and according to the initial concerned metric of the user group for the work group, the work groups associated with each user group can be determined from multiple initial work groups.

[0067] Specifically, a robust relevance model can be constructed. This model can effectively characterize the relevance between side_info (content-side information), user_info (user-side information), and user_behavior (user-side behavior), that is, it can effectively associate users based on content. Constructing a relevance model can eliminate population bias, ensure the fairness of content competition, and then build a good content ecosystem. The present disclosure can train the XTR model (relevance model) by continuously optimizing user portraits and content understanding and weakening pid (photo ID, work identifier), such as using unbiased side info (mmu (MultiMediaUnderstanding, multimedia content understanding) embedding (mapping)) on the pid side to construct the model.

[0068] The above first relevance model can be a deep learning model. By inputting user behavior data into the above relevance model, the relevance between the user and the work can be output.

[0069] In the embodiments of the present disclosure, before obtaining the first set of resources to be distributed associated with the first set of account objects and the second set of resources to be distributed associated with the second set of account objects, authorization indication information can be displayed to the account objects; in response to the behavior of the first account object clicking on the authorization indication information, the first concerned metric of the first account object for the first resource to be distributed can be obtained.

[0070] In step S320, the first resource to be distributed in the first set of resources to be distributed is distributed to the first account object in the first set of account objects to obtain the first concerned metric information of the first account object for the first resource to be distributed.

[0071] Among them, there can be one or more first resources to be distributed, and there can be one or more first account objects; the attention metric can be the degree of interest of the account object in the resources to be distributed, where the attention metric can include a first attention metric, a second attention metric, and a third attention metric. For example, the attention metric can be the relevance between the user and the work, which can be represented by click-through rate, collection rate, forwarding rate, etc. In the following illustrative examples, the attention metric is taken as the click-through rate for illustration, but the present disclosure is not limited thereto.

[0072] In the embodiments of the present disclosure, each first resource to be distributed in the first set of resources to be distributed can be distributed to each first account object in the first set of first account objects; the first attention metric of each first account object for each first resource to be distributed can be obtained according to the behavior data of each first account object for each first resource to be distributed.

[0073] In the embodiments of the present disclosure, different groups of works can be distributed to different groups of users for viewing. For example, the first group of works can be distributed to the first group of users for viewing, and the second group of works can be distributed to the second group of users for viewing.

[0074] Among them, the process of distributing the first group of works to the first group of users for viewing and distributing the second group of works to the second group of users for viewing can be called the cold start process, that is, cold starting the first group of works on the first group of users and cold starting the second group of works on the second group of users.

[0075] In the embodiments of the present disclosure, cold start can include new content cold start and new user cold start. Among them, new content cold start can refer to when new content is added to the recommendation system, due to the lack of rich user preference evaluation information for the new content or even no user preference evaluation information at all, the recommendation system cannot match the corresponding users for the new content, which may result in low user interest when the new content is recommended to users; new user cold start can refer to when a new user joins the recommendation system, due to the lack of sufficient historical preference evaluation information for the new user, the recommendation system may not be able to make accurate content recommendations for the new user.

[0076] Reference Figure 4 ..., the similar group of works 1 can be distributed to the similar group of old users 1 for viewing, that is, cold starting the similar group of works 1 on the similar group of old users 1; the similar group of new works 2 can be distributed to the similar group of old users 2 for viewing, that is, cold starting the similar group of works 2 on the similar group of old users 2.

[0077] In the embodiments of the present disclosure, work distribution can be performed based on the cold start constraint relevance mechanism, that is, constraining similar works to compete in the sub-network, so that similar works are distributed to similar users, and thus the relevance obtained by similar works can be fairly compared, ensuring the fairness of content competition.

[0078] Figure 5 It is a schematic diagram of another user group and work group shown according to an example.

[0079] Reference Figure 5 In ① of [], each work in work group 1 can be distributed to each user in user group 1 for viewing, and each work in work group 2 can be distributed to each user in user group 2 for viewing. According to the behavior (such as clicking) of each user in the user group on each work in the work group, the attention index (such as click-through rate) of each user on each work is determined.

[0080] In step S330, according to the first attention index information of the first account object for the first resource to be distributed, the first target resource to be distributed is determined from the first set of resources to be distributed.

[0081] Among them, there can be one or more first target resources to be distributed.

[0082] In an exemplary embodiment, the first resource to be distributed with the first attention index information greater than the first attention index information threshold is determined as the first target resource to be distributed; or, the first attention index information is sorted in descending order, and the first resources to be distributed corresponding to the first preset number of the first attention index information are determined as the first target resources to be distributed; or, the first attention index information is sorted in descending order, and the first resources to be distributed corresponding to the first preset ratio of the first attention index information are determined as the first target resources to be distributed.

[0083] In the embodiments of the present disclosure, a first attention index information threshold (such as a click-through rate threshold) can be set, and the work with the first attention index information greater than the first attention index information threshold is determined as the target work.

[0084] In the embodiments of the present disclosure, the first attention index information can be sorted in descending order, and the first N or first M% of the works corresponding to the first attention index information are taken as the target works, where N can be an integer greater than or equal to 1, and M can be a number between 0 and 100.

[0085] Among them, the target work can be considered as a high-quality work that wins or stands out.

[0086] Reference Figure 5 , for example, if the click-through rate of work 501 in work group 1 exceeds the click-through rate threshold, then work 501 is determined as the target work.

[0087] In the embodiments of the present disclosure, when a work wins through fair competition in a sub-network (a user group and its associated work group can form a sub-network), the system can know that this is a vertically high-quality work, and gradually expand the distribution scope of these "winning" works, that is, try to explore content among some people with less strong relevance to it, which can ensure that it can continue to climb and break through the circle. At the same time, from the user's perspective, this process of breaking through the circle may be to recommend works that the user is not very familiar with, so the user can consider it as an interest exploration.

[0088] In step S340, the first target resource to be distributed is distributed to some second account objects in the second account object set to obtain the second attention index information of some second account objects for the first target resource to be distributed.

[0089] In the embodiments of the present disclosure, the first target resource to be distributed can be distributed to some second account objects in the second account object set; the second attention index information of some second account objects for the first target resource to be distributed can be obtained according to the behavior data of some second account objects for the first target resource to be distributed.

[0090] Reference Figure 5 In ② of, the target work 501 in work group 1 can be distributed to some users in user group 2 for viewing, and according to the behavior (such as clicking) of the above-mentioned some users in user group 2 for the target work 501, the attention index (such as click-through rate) of some users in user group 2 for the target work 501 is determined.

[0091] In the embodiments of the present disclosure, the target work can be considered as a high-quality work that wins or breaks through the slope, and the target work can be distributed to user group 2 for climbing and interest exploration.

[0092] In the embodiments of the present disclosure, the second attention index information of some second account objects for the first target resource to be distributed can be obtained by responding to the behavior of the second account object clicking on the authorization indication information.

[0093] In the embodiments of the present disclosure, the first target resource to be distributed is a resource to be distributed that the first account object is interested in selected from the first resources to be distributed. Distributing the first target resource to some second account objects in the stage where it cannot be determined at the beginning (that is, the stage where it cannot be determined whether the second account object is interested in the first target resource to be distributed) can ensure the accuracy of the distribution and prevent the loss of these second account objects; at the same time, in the stage where it cannot be determined at the beginning, only a small number of second account objects need to be distributed the first target resource to test whether the second account object set is interested in the first target resource to be distributed, thereby reducing the computational amount and storage space in the initial test stage.

[0094] In step S350, when the second attention metric information meets the preset conditions, add the first target resource to be distributed to the second set of resources to be distributed, so as to update the second set of resources to be distributed.

[0095] In the embodiments of the present disclosure, the first target resource to be distributed can be added to the second set of resources to be distributed, and the second set of resources to be distributed containing the first target resource to be distributed is used as the updated second set of resources to be distributed.

[0096] Among them, adding the first target resource to be distributed to the second set of resources to be distributed can be moving the first target resource to be distributed to the second set of resources to be distributed, and the first target resource to be distributed is used as the second resource to be distributed to be distributed to the second set of account objects; it can also be copying the first target resource to be distributed into the second set of resources to be distributed, and the first target resource to be distributed is used as the first resource to be distributed and the second resource to be distributed at the same time, and is distributed to the first set of account objects and the second set of account objects respectively.

[0097] In an exemplary embodiment, if the second attention metric information is greater than the second attention metric information threshold, add the first target resource to be distributed to the second set of resources to be distributed.

[0098] In the embodiments of the present disclosure, the second attention metric threshold (such as the click-through rate threshold) can be set. If the second attention metric is greater than the second attention metric threshold, add the first target resource to be distributed to the second set of resources to be distributed.

[0099] Refer to Figure 5 In ③ of, for example, the attention metric of the users in user group 2 for the target work 501 is greater than the second attention metric threshold, then the target work 501 can be added to work group 2 and distributed to the users in user group 2 for viewing as a work in work group 2.

[0100] In the embodiments of the present disclosure, if the target work meets the above conditions, it can be considered that the target work has also obtained a good attention metric on the sub-network composed of the second user group and the second work group, then the target work can be pulled into the second work group, and the target work has achieved breaking the circle.

[0101] The data processing method provided by some embodiments of the present disclosure determines a first target to-be-distributed resource from a first to-be-distributed resource according to the first attention index information of the first to-be-distributed resource for a first account object. The first to-be-distributed resource that the first account object is interested in can be used as the first target to-be-distributed resource. Distributing the first target to-be-distributed resource to some second account objects in the second account object set only requires distributing the first target to-be-distributed resource to a small number of second account objects first, and then the interest of the second account object set in the first target to-be-distributed resource can be tested. Thus, the computational workload and storage space in the initial test stage can be reduced. Moreover, distributing the first target to-be-distributed resource to some second account objects in the second account object set can enable the first target to-be-distributed resource to be distributed to more account objects to obtain more attention, and at the same time enable some second account objects to see more to-be-distributed resources, increasing the supply of target content on the supply side and enhancing the diversity of the content ecosystem. According to the second attention index of some second account objects for the first target to-be-distributed resource, it can be determined to add the first target to-be-distributed resource to the second to-be-distributed resource set, enabling other account objects in the second account object set to see the first target to-be-distributed resource, realizing content circulation, avoiding the information cocoon problem in related technologies, enhancing user interest, and increasing user stickiness.

[0102] Figure 6 It is a flowchart of another data processing method shown according to an exemplary embodiment.

[0103] Based on Figure 3 the shown placement data processing method, Figure 6 the shown placement data processing method may further include the following steps.

[0104] In step S610, third attention index information of each second account object in some second account objects for the first target to-be-distributed resource is obtained.

[0105] In the embodiments of the present disclosure, the first target to-be-distributed resource can be distributed to some second account objects in the second account object set; based on the behavior data of each second account object in some second account objects for the first target to-be-distributed resource, the third attention index of each second account object in some second account objects for the first target to-be-distributed resource can be obtained.

[0106] Referring to Figure 5 , the target work 501 can be distributed to some users in user group 2, and based on the behavior data (such as click behavior) of each user in some users in user group 2 for the target work 501, the attention index (such as click-through rate) of each user in some users in user group 2 for the target work 501 can be determined.

[0107] In step S620, according to the third attention index information, determine the second target account object from some of the second account objects.

[0108] In the embodiments of the present disclosure, the target users with successful interest exploration can be determined according to the third attention index information.

[0109] In an exemplary embodiment, the second account object with the third attention index information greater than the third attention index threshold is determined as the second target account object; or, the third attention index information is sorted in descending order, and the second account objects corresponding to the first second preset quantity of the third attention index information are determined as the second target account objects; or, the third attention index is sorted in descending order, and the second account objects corresponding to the first second preset ratio of the third attention index are determined as the second target account objects.

[0110] In the embodiments of the present disclosure, a third attention index threshold (such as a click-through rate threshold) can be set, and the users with the third attention index greater than the third attention index threshold are determined as target users.

[0111] In the embodiments of the present disclosure, the third attention index can be sorted in descending order, and the first N or the first M% of the users corresponding to the third attention index are taken as target users.

[0112] Reference Figure 5 , for example, if the click-through rate of user 502 in user group 2 on target work 501 exceeds the click-through rate threshold, then user 502 is determined as the target user.

[0113] In step S630, add the second target account object to the first account object set to update the first account object set.

[0114] Among them, adding the second target account object to the first account object set can be moving the second target account object to the first account object set, and the second target account object serves as the first account object to be distributed the first set of accounts to be distributed; or copying the second target account object to the first account object set, and the second target account object serves as both the first account object and the second account object to be distributed the first set of resources to be distributed and the second set of resources to be distributed.

[0115] Reference Figure 5 in ④, for example, adding target user 502 to user group 1, as a user in user group 1, to be distributed the works in work group 1 for viewing.

[0116] In the embodiments of the present disclosure, the winning works can compete with works in other sub - networks for users. Therefore, when the DAU (Daily Active User) is limited, if the system discovers more confident interests of users, that is, a user belongs to many sub - networks at the same time, then the needs of each sub - network can be more abundant, increasing the opportunities for content growth and breaking through the circle.

[0117] In an exemplary embodiment, when distributing the first resource to be distributed in the first set of resources to be distributed to the first account object in the first set of account objects, a first uncertainty parameter and a second uncertainty parameter are randomly generated, and the first uncertainty parameter is less than the second uncertainty parameter; the first attention index information is obtained according to the first uncertainty parameter and the second uncertainty parameter.

[0118] In an exemplary embodiment, when distributing the first target resource to be distributed to some second account objects in the second set of account objects, the first uncertainty parameter and the second uncertainty parameter are adjusted so that the first uncertainty parameter is greater than the second uncertainty parameter; the second attention index information is obtained according to the adjusted first uncertainty parameter and the second uncertainty parameter.

[0119] In the embodiments of the present disclosure, when distributing the first resource to be distributed in the first set of resources to be distributed to the first account object in the first set of account objects, it can be considered as the cold start stage (such as Figure 5 shown as ①); when distributing the first target resource to be distributed to some second account objects in the second set of account objects, it can be considered as the ramp - up stage (such as Figure 5 shown as ②).

[0120] In the embodiments of the present disclosure, based on the Contextual Bandit idea in the multi - armed bandit, uncertainty can be introduced into the sorting target. In the cold start stage, a first uncertainty parameter and a second uncertainty parameter are randomly generated, where the randomly generated first uncertainty parameter is less than the second uncertainty parameter, and the first attention index information is obtained according to the first uncertainty parameter and the second uncertainty parameter; in the ramp - up stage, the first uncertainty parameter and the second uncertainty parameter are adjusted so that the first uncertainty parameter is greater than the second uncertainty parameter, and the second attention index information is obtained according to the adjusted first uncertainty parameter and the second uncertainty parameter.

[0121] For example, when introducing uncertainty into the sorting target, the form of sorting can be:

[0122]

[0123] where u r is the first uncertainty parameter, ue is the second uncertainty parameter, rxtr can be the first correlation, extr can be the second correlation, pxtr can be the third correlation, u p is the uncertainty parameter of pxtr, f can be a monotonically decreasing function, and ~ represents the equivalence symbol.

[0124] Among them, the first correlation rxtr can be p(u|i), that is, the attention index of the first user group for the works in each first work group; the second correlation extr can be p(i), that is, the attention index of part of the second user group for the target work; the third correlation pxtr can be p(i|u), that is, the attention index of each user in part of the second user group for the target work. Where u can represent user, and i can represent item.

[0125] Among them, the first correlation rxtr, the second correlation extr, and the third correlation pxtr can all be obtained through a correlation model deployed online (such as a deep learning model).

[0126] In the embodiments of the present disclosure, the sorting target can be determined according to the first correlation rxtr, the first uncertainty parameter u r , the second correlation extr, and the second uncertainty parameter u e (that is, pxtr + u on the left side of the formula p ). When distributing the first resources to be distributed in the first set of resources to be distributed to the first account objects in the first set of account objects, the sorting target is used as the first attention index information; when distributing the first target resources to be distributed to some second account objects in the second set of account objects, the sorting target is used as the second attention index information.

[0127] In the embodiments of the present disclosure, in the cold start stage, the first uncertainty parameter u r and the second uncertainty parameter u e can be randomly generated. The randomly generated first uncertainty parameter u r is less than the second uncertainty parameter u e , which can make the weight of the first correlation rxtr in the sorting target larger. This is because in the cold start stage (such as Figure 5 shown in ①), the exposure of the works is low, the third correlation pxtr estimated by the correlation model is not confident, and the second correlation extr is also not confident. The sorting can be performed according to the first correlation rxtr, that is, the optimization target can be the first correlation rxtr.

[0128] In the embodiments of the present disclosure, in the ramp-up stage, the first uncertainty parameter u r and the second uncertainty parameter u esuch that the randomly generated first uncertainty parameter u r is greater than the second uncertainty parameter u e , which can make the second correlation extr have a greater weight in the sorting target. This is because when the content enters the climbing stage after going down the slope (for example Figure 5 shown in ②), the exposure continues to increase. At this time, the exploration of breaking through the circle of the work will gradually reduce the weight of the first correlation pxtr and increase the weight of the second correlation extr. If the work can still maintain a high second correlation extr, the content will gradually break through the circle, that is, penetrate the user's interest, and may finally pass the high-heat review, topk review, and even become a hit.

[0129] In the embodiments of the present disclosure, some hyperparameters can also be added for regulation in this sorting formula. The system takes the user duration, the number of works, etc. corresponding to the long period in the sub-network as the target, learns reasonable hyperparameters to adjust the supply-demand balance of the sub-network, and thus improves the practice of forced insertion according to a fixed ratio in the related art.

[0130] Figure 7 is a schematic diagram of a bilateral transfer mechanism shown according to an exemplary embodiment.

[0131] Refer to Figure 7 , the embodiments of the present disclosure implement a bilateral transfer mechanism based on uncertainty estimation through the above method, connect interest exploration and content exploration, achieve the construction of a diversified content ecosystem, and realize effective global E&E.

[0132] Specifically, the present disclosure draws on the idea of Contextual Bandit. Based on the bilateral transfer mechanism of uncertainty, new content can be distributed to old users according to the correlation to achieve content exploration, and verified good content can be distributed to new users to achieve interest exploration. It should be noted that in the embodiments of the present disclosure, for users, new works can refer to works that the user has not seen, and old works can refer to works that the user has seen; for works, new users can refer to users who have not seen the work, and old users can refer to users who have seen the work.

[0133] The data processing method provided by the embodiments of the present disclosure determines to add a second target account object to the first account object set according to the third attention index of each second account object in some second account objects for the first target resource to be distributed. The second target account object interested in the first target resource to be distributed can be added to the first account object set, so that other resources to be distributed in the first resource set to be distributed can be seen by the second target account object, realizing user transfer, thus realizing bilateral transfer, avoiding the problem of information cocoons in the related art, enhancing user interest, and increasing user viscosity.

[0134] Figure 8 is a block diagram of a data processing apparatus shown according to an exemplary embodiment. Referring to Figure 8 , the apparatus 800 may include an obtaining module 810, a distributing module 820, and a determining module 830.

[0135] Among them, the obtaining module 810 may be configured to obtain a first set of resources to be distributed that has a distribution association with a first set of account objects, and a second set of resources to be distributed that has a distribution association with a second set of account objects; the distributing module 820 may be configured to distribute the first resources to be distributed in the first set of resources to be distributed to the first account objects in the first set of account objects to obtain first attention metric information of the first account objects for the first resources to be distributed; the determining module 830 may be configured to determine a first target resource to be distributed from the first set of resources to be distributed according to the first attention metric information of the first account objects for the first resources to be distributed; the distributing module 820 may also be configured to distribute the first target resource to be distributed to some of the second account objects in the second set of account objects to obtain second attention metric information of some of the second account objects for the first target resource to be distributed; the determining module 830 may also be configured to add the first target resource to be distributed to the second set of resources to be distributed to update the second set of resources to be distributed when the second attention metric information meets a preset condition.

[0136] In some exemplary embodiments of the present disclosure, the determining module 830 may also be configured to determine the first resources to be distributed with the first attention metric information greater than the first attention metric information threshold as the first target resources to be distributed; or, the determining module 830 may also be configured to perform a descending order arrangement on the first attention metric information, and determine the first resources to be distributed corresponding to the first preset number of the first attention metric information as the first target resources to be distributed; or, the determining module 830 may also be configured to perform a descending order arrangement on the first attention metric information, and determine the first resources to be distributed corresponding to the first preset proportion of the first attention metric information as the first target resources to be distributed.

[0137] In some exemplary embodiments of the present disclosure, the determining module 830 may also be configured to add the first target resource to be distributed to the second set of resources to be distributed to update the second set of resources to be distributed if the second attention metric information is greater than the second attention metric threshold.

[0138] In some exemplary embodiments of the present disclosure, the apparatus 800 may further include: an obtaining module; the obtaining module may be configured to obtain the third attention metric information of each second account object in a part of the second account objects for the first target resource to be distributed; the determining module 830 may further be configured to determine a second target account object from the part of the second account objects according to the third attention metric information; the determining module 830 may further be configured to add the second target account object to the first account object set to update the first account object set.

[0139] In some exemplary embodiments of the present disclosure, the determining module 830 may further be configured to determine a second account object with the third attention metric information greater than the third attention metric threshold as the second target account object; alternatively, the determining module 830 may further be configured to perform a descending order arrangement on the third attention metric information, and determine the second account objects corresponding to the first second preset quantity of the third attention metric information as the second target account objects; alternatively, the determining module 830 may further be configured to perform a descending order arrangement on the third attention metric, and determine the second account objects corresponding to the first second preset ratio of the third attention metric as the second target account objects.

[0140] In some exemplary embodiments of the present disclosure, when obtaining the first set of resources to be distributed associated with the first account object set and the second set of resources to be distributed associated with the second account object set, the first resource to be distributed is not distributed to the second account object, and the second resource to be distributed is not distributed to the first account object.

[0141] In some exemplary embodiments of the present disclosure, the determining module 830 may further be configured to determine an initial account object set according to the similarity between account objects, where the initial account object set includes the first account object set; the determining module 830 may further be configured to determine an initial set of resources to be distributed according to the similarity between the resources to be distributed; the determining module 830 may further be configured to determine a first set of resources to be distributed associated with the first account object set from the initial set of resources to be distributed according to the initial attention metric of the initial account object set for the initial set of resources to be distributed.

[0142] In some exemplary embodiments of the present disclosure, the above-mentioned apparatus may further include: a generation module, an acquisition module, and an adjustment module; the generation module may be configured to randomly generate a first uncertainty parameter and a second uncertainty parameter when distributing a first resource to be distributed in a first set of resources to be distributed to a first set of account objects, the first uncertainty parameter being less than the second uncertainty parameter; the acquisition module may be configured to obtain first focus index information according to the first uncertainty parameter and the second uncertainty parameter; the adjustment module may be configured to adjust the first uncertainty parameter and the second uncertainty parameter when distributing a first target resource to be distributed to some second account objects in a second set of account objects, so that the first uncertainty parameter is greater than the second uncertainty parameter; the acquisition module may also be configured to obtain second focus index information according to the adjusted first uncertainty parameter and the second uncertainty parameter.

[0143] Regarding the apparatus in the above embodiments, the specific manners in which each module performs operations have been described in detail in the embodiments related to the method, and will not be elaborated herein.

[0144] The following refers to Figure 9 to describe the electronic device 900 according to such an embodiment of the present disclosure. Figure 9 The electronic device 900 shown is merely an example and should not impose any limitations on the functions and usage scope of the embodiments of the present disclosure.

[0145] As Figure 9 shown, the electronic device 900 is presented in the form of a general-purpose computing device. The components of the electronic device 900 may include, but are not limited to: the above-mentioned at least one processing unit 910, the above-mentioned at least one storage unit 920, a bus 930 connecting different system components (including the storage unit 920 and the processing unit 910), and a display unit 940.

[0146] Among them, the storage unit stores program codes, and the program codes can be executed by the processing unit 910, so that the processing unit 910 executes the steps according to various exemplary embodiments of the present disclosure described in the above "exemplary method" section of this specification. For example, the processing unit 910 can execute as Figure 3In step S310 shown in the figure, obtain a first set of resources to be distributed that has a distribution association with the first set of account objects, and a second set of resources to be distributed that has a distribution association with the second set of account objects; in step S320, distribute the first resources to be distributed in the first set of resources to be distributed to the first account object in the first set of account objects to obtain first attention metric information of the first account object for the first resources to be distributed; in step S330, determine a first target resource to be distributed from the first set of resources to be distributed according to the first attention metric information of the first account object for the first resources to be distributed; in step S340, distribute the first target resource to be distributed to some of the second account objects in the second set of account objects to obtain second attention metric information of some of the second account objects for the first target resource to be distributed; in step S350, when the second attention metric information meets a preset condition, add the first target resource to be distributed to the second set of resources to be distributed to update the second set of resources to be distributed.

[0147] For another example, the electronic device can implement each step as Figure 3 shown.

[0148] The storage unit 920 may include a readable medium in the form of a volatile storage unit, such as a random access storage unit (RAM) 921 and / or a cache storage unit 922, and may further include a read-only storage unit (ROM) 923.

[0149] The storage unit 920 may further include a program / utilities 924 having a set (at least one) of program modules 925. Such program modules 925 include, but are not limited to: an operating system, one or more application programs, other program modules, and program data. Each or some combination of these examples may include an implementation of a network environment.

[0150] The bus 930 may represent one or more of several types of bus structures, including a memory bus or memory controller, a peripheral bus, an accelerated graphics port, a processing unit, or a local bus using any of the various bus structures.

[0151] The electronic device 900 can also communicate with one or more external devices 970 (such as a keyboard, a pointing device, a Bluetooth device, etc.), and can also communicate with one or more devices that enable a user to interact with the electronic device 900, and / or communicate with any device (such as a router, a modem, etc.) that enables the electronic device 900 to communicate with one or more other computing devices. Such communication can be carried out through the input / output (I / O) interface 950. Moreover, the electronic device 900 can also communicate with one or more networks (such as a local area network (LAN), a wide area network (WAN), and / or a public network, such as the Internet) through the network adapter 960. As shown in the figure, the network adapter 960 communicates with other modules of the electronic device 900 through the bus 930. It should be understood that, although not shown in the figure, other hardware and / or software modules can be used in combination with the electronic device 900, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems, etc.

[0152] Through the description of the above embodiments, those skilled in the art can easily understand that the example embodiments described herein can be implemented by software or by a combination of software and necessary hardware. Therefore, the technical solutions according to the embodiments of the present disclosure can be embodied in the form of a software product, and the software product can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.) or on a network, including several instructions to enable a computing device (which can be a personal computer, a server, a terminal device, or a network device, etc.) to execute the method according to the embodiments of the present disclosure.

[0153] In an exemplary embodiment, there is also provided a computer-readable storage medium including instructions, such as a memory including instructions, and the above instructions can be executed by a processor of the device to complete the above method. Optionally, the computer-readable storage medium can be a ROM, a random access memory (RAM), a CD-ROM, a magnetic tape, a floppy disk, and an optical data storage device, etc.

[0154] In an exemplary embodiment, there is also provided a computer program product including a computer program / instructions, and when the computer program / instructions are executed by a processor, the data processing method in the above embodiments is implemented.

[0155] After considering the specification and practicing the invention disclosed herein, those skilled in the art will readily conceive of other embodiments of the present disclosure. This application is intended to cover any variations, uses, or adaptations of the present disclosure, which follow the general principles of the present disclosure and include known common general knowledge or conventional technical means in the technical field not disclosed by the present disclosure. The specification and embodiments are only regarded as exemplary, and the true scope and spirit of the present disclosure are pointed out by the following claims.

[0156] It should be understood that the present disclosure is not limited to the exact structures described above and shown in the drawings, and various modifications and changes can be made without departing from its scope. The scope of the present disclosure is only limited by the appended claims.

Claims

1. A data processing method, characterized in that, Including: Obtaining a first set of resources to be distributed that has a distribution association with a first set of account objects, and a second set of resources to be distributed that has a distribution association with a second set of account objects; Distributing a first resource to be distributed in the first set of resources to be distributed to a first account object in the first set of account objects to obtain first attention metric information of the first account object for the first resource to be distributed; Determining a first target resource to be distributed from the first set of resources to be distributed according to the first attention metric information of the first account object for the first resource to be distributed; Distributing the first target resource to be distributed to some second account objects in the second set of account objects to obtain second attention metric information of the some second account objects for the first target resource to be distributed; When the second attention metric information meets a preset condition, adding the first target resource to be distributed to the second set of resources to be distributed to update the second set of resources to be distributed; Among them, the step of obtaining the first set of resources to be distributed that has a distribution association with the first set of account objects includes: Determining an initial set of account objects according to the similarity between account objects, where the initial set of account objects includes the first set of account objects; Determining an initial set of resources to be distributed according to the similarity between resources to be distributed; Determining the first set of resources to be distributed that has a distribution association with the first set of account objects from the initial set of resources to be distributed according to the initial attention metrics of the initial set of account objects for the initial set of resources to be distributed.

2. The data processing method according to claim 1, wherein The step of determining a first target resource to be distributed from the first resources to be distributed according to the first attention metric information of the first account object for the first resource to be distributed includes: Determining the first resource to be distributed with the first attention metric information greater than the first attention metric information threshold as the first target resource to be distributed; or, Performing a descending order arrangement on the first attention metric information, and determining the first resources to be distributed corresponding to the first preset number of the first attention metric information as the first target resources to be distributed; or, Performing a descending order arrangement on the first attention metric information, and determining the first resources to be distributed corresponding to the first preset proportion of the first attention metric information as the first target resources to be distributed.

3. The data processing method according to claim 1 or 2, characterized in that, The step of, when the second attention metric information meets a preset condition, adding the first target resource to be distributed to the second set of resources to be distributed to update the second set of resources to be distributed includes: If the second attention metric information is greater than the second attention metric threshold, adding the first target resource to be distributed to the second set of resources to be distributed to update the second set of resources to be distributed.

4. The data processing method according to claim 1, wherein The method further includes: Obtaining third attention metric information of each second account object in the some second account objects for the first target resource to be distributed; Determining a second target account object from the some second account objects according to the third attention metric information; Add the second target account object to the first account object set to update the first account object set.

5. The data processing method according to claim 4, wherein The step of determining the second target account object from the partial second account objects according to the third attention index information includes: Determine the second account object whose third attention index information is greater than the third attention index threshold as the second target account object; or, Arrange the third attention index information in descending order, and determine the second account objects corresponding to the first second preset quantity of the third attention index information as the second target account objects; or, Arrange the third attention index in descending order, and determine the second account objects corresponding to the first second preset ratio of the third attention index as the second target account objects.

6. The data processing method according to claim 1, wherein When obtaining the first set of resources to be distributed associated with the first account object set and the second set of resources to be distributed associated with the second account object set, the first resource to be distributed is not distributed to the second account object, and the second resource to be distributed is not distributed to the first account object.

7. The data processing method according to claim 1, wherein Further includes: When distributing the first resource to be distributed in the first set of resources to be distributed to the first account object in the first account object set, randomly generate a first uncertainty parameter and a second uncertainty parameter, where the first uncertainty parameter is less than the second uncertainty parameter; Obtain the first attention index information according to the first uncertainty parameter and the second uncertainty parameter; When distributing the first target resource to be distributed to some of the second account objects in the second account object set, adjust the first uncertainty parameter and the second uncertainty parameter so that the first uncertainty parameter is greater than the second uncertainty parameter; Obtain the second attention index information according to the adjusted first uncertainty parameter and the second uncertainty parameter.

8. A data processing device, characterized in that, Includes: An acquisition module configured to execute acquiring a first set of resources to be distributed associated with the first account object set and a second set of resources to be distributed associated with the second account object set; A distribution module configured to execute distributing the first resource to be distributed in the first set of resources to be distributed to the first account object in the first account object set to obtain the first attention index information of the first account object for the first resource to be distributed; A determination module configured to execute determining a first target resource to be distributed from the first set of resources to be distributed according to the first attention index information of the first account object for the first resource to be distributed; The distribution module is further configured to execute distributing the first target resource to be distributed to some of the second account objects in the second account object set to obtain the second attention index information of the some second account objects for the first target resource to be distributed; The determination module is further configured to execute adding the first target resource to be distributed to the second set of resources to be distributed to update the second set of resources to be distributed when the second attention index information meets a preset condition; The determining module is further configured to execute to determine an initial set of account objects according to the similarity between account objects, where the initial set of account objects includes the first set of account objects; The determining module is further configured to execute to determine an initial set of resources to be distributed according to the similarity between the resources to be distributed; The determining module is further configured to execute to determine, from the initial set of resources to be distributed, the first set of resources to be distributed that has a distribution association with the first set of account objects according to the initial attention metrics of the initial set of resources to be distributed with respect to the initial set of account objects.

9. The data processing device according to claim 8, wherein The determining module is further configured to execute to determine the first resources to be distributed with the first attention metric information greater than the first attention metric information threshold as the first target resources to be distributed; or, The determining module is further configured to execute to sort the first attention metric information in descending order, and determine the first resources to be distributed corresponding to the first preset number of the first attention metric information at the front as the first target resources to be distributed; or, The determining module is further configured to execute to sort the first attention metric information in descending order, and determine the first resources to be distributed corresponding to the first preset proportion of the first attention metric information at the front as the first target resources to be distributed.

10. The data processing device according to claim 8 or 9, wherein The determining module is further configured to execute to add the first target resources to be distributed to the second set of resources to be distributed to update the second set of resources to be distributed if the second attention metric information is greater than the second attention metric threshold.

11. The data processing device according to claim 8, wherein The device further includes: an obtaining module; The obtaining module is configured to execute to obtain the third attention metric information of each of the second account objects in the partial second account objects with respect to the first target resources to be distributed; The determining module is further configured to execute to determine second target account objects from the partial second account objects according to the third attention metric information; The determining module is further configured to execute to add the second target account objects to the first set of account objects to update the first set of account objects.

12. The data processing device according to claim 11, wherein The determining module is further configured to execute to determine the second account objects with the third attention metric information greater than the third attention metric threshold as the second target account objects; or, The determining module is further configured to execute to sort the third attention metric information in descending order, and determine the second account objects corresponding to the second preset number of the third attention metric information at the front as the second target account objects; or, The determining module is further configured to execute to sort the third attention metric in descending order, and determine the second account objects corresponding to the second preset proportion of the third attention metric at the front as the second target account objects.

13. The data processing device according to claim 8, wherein When obtaining a first set of resources to be distributed that has a distribution association with a first set of account objects, and a second set of resources to be distributed that has a distribution association with a second set of account objects, the first resources to be distributed are not distributed to the second account objects, and the second resources to be distributed are not distributed to the first account objects.

14. The data processing device according to claim 8, wherein It further includes: A generation module, an acquisition module, and an adjustment module; The generation module is configured to execute generating a first uncertainty parameter and a second uncertainty parameter randomly when distributing a first resource to be distributed in the first set of resources to be distributed to a first account object in the first set of account objects, where the first uncertainty parameter is less than the second uncertainty parameter; The acquisition module is configured to execute obtaining the first attention index information according to the first uncertainty parameter and the second uncertainty parameter; The adjustment module is configured to execute adjusting the first uncertainty parameter and the second uncertainty parameter when distributing the first target resource to be distributed to some second account objects in the second set of account objects, so that the first uncertainty parameter is greater than the second uncertainty parameter; The acquisition module is further configured to execute obtaining the second attention index information according to the adjusted first uncertainty parameter and the second uncertainty parameter.

15. An electronic device, characterized in that, It includes: A processor; A memory for storing instructions executable by the processor; Wherein, the processor is configured to execute the executable instructions to implement the data processing method according to any one of claims 1 to 7.

16. A computer-readable storage medium, when the instructions in the computer-readable storage medium are executed by a processor of an electronic device, enabling the electronic device to execute the data processing method according to any one of claims 1 to 7.

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

Images