A resource determination method and apparatus, electronic device, and storage medium

Abstract

Embodiments of the present application provide a resource determination method and device, electronic equipment and storage medium, and relate to the technical field of computer, which comprises: acquiring a plurality of preset network resources; for each network resource, determining the matching degree of the network resource and each user account based on the type label of the network resource; for each first network resource, determining a first user account from each user account according to the matching degree of the first network resource and each user account; for each first user account, determining a third network resource from each second network resource according to the matching degree of the first user account and the second network resource; determining the first network resource corresponding to the first user account and the third network resource corresponding to the first user account as the target network resource corresponding to the first user account, which can guarantee the demand of the service provider and the personalized demand of the user, and balance between the demand of the service provider and the personalized demand of the user.

Description

Technical Field

[0001] This invention relates to the field of computer technology, and in particular to a resource determination method, apparatus, electronic device, and storage medium. Background Technology

[0002] With the development of computer technology, users can browse online resources through clients, such as movies, TV series, and music.

[0003] In existing technologies, service providers can send network resources to clients. For example, a service provider can send newly popular resources to the client; these are network resources that have been accessed frequently within a certain period or are newly released network resources. Alternatively, a service provider can send network resources that match a user's interests and personalized needs; these resources are determined based on the activity logs of the user accounts logged into the client. Subsequently, the client can display the received network resources to users whose accounts it is logged into.

[0004] However, service providers need to send more new, trending resources to clients to attract users and increase user activity. Users, on the other hand, need more network resources that match their interests and personal preferences to improve user experience. Consequently, when clients display received network resources to users, it's difficult to balance the availability of new, trending resources with those that align with user interests and preferences, making it challenging to simultaneously meet both the service provider's needs and the user's personalized requirements. Summary of the Invention

[0005] The purpose of this invention is to provide a resource determination method, apparatus, electronic device, and storage medium to ensure a balance between the needs of service providers and the personalized needs of users. The specific technical solution is as follows:

[0006] In a first aspect of this invention, a resource determination method is provided, the method comprising:

[0007] Acquire multiple preset network resources; the multiple network resources include: a first network resource sent to a client logged in with a first user account;

[0008] For each network resource, the matching degree between the network resource and each user account is determined based on the type tag of the network resource;

[0009] For each first network resource, based on the matching degree between the first network resource and each user account, the user account corresponding to the first network resource is determined from each user account and used as the first user account.

[0010] For each first user account, based on the matching degree between the first user account and the second network resources other than the first network resources among the plurality of network resources, the third network resource corresponding to the first user account is determined from each second network resource;

[0011] The first network resource corresponding to the first user account and the third network resource corresponding to the first user account are determined as the target network resource corresponding to the first user account.

[0012] Optionally, for each network resource, determining the matching degree between the network resource and each user account based on the type tag of the network resource includes:

[0013] For each user account, obtain the user's operation information on the historical network resources browsed by the user account to obtain the first operation information;

[0014] Based on the first operation information and the type tags of the historical network resources that have been viewed, determine the matching degree between each type tag of the historical network resources that have been viewed and the user account.

[0015] For each network resource, determine the type tag that the network resource has from the various type tags of the historical network resources that have been browsed;

[0016] The matching degree between the network resource and the user account is calculated based on the matching degree between the type tag of the network resource and the user account.

[0017] Optionally, the first operation information includes: the number of operations performed by the user to which the user account belongs on historical network resources that have been viewed; the operation includes at least one of the following: click operation, forward operation, comment operation;

[0018] The step of determining the matching degree between the various type tags of the viewed historical network resources and the user account based on the first operation information and the type tags of the viewed historical network resources includes:

[0019] For each type of tag in the historical web resources that have been viewed, count the number of operations performed by the user account on the historical web resources with that type of tag, and use that count as the number of operations performed by the user account on that type of tag.

[0020] Generate a target feature vector containing the number of operations performed by the user account for each type of tag;

[0021] The target feature vector is input into a pre-trained probabilistic prediction model to obtain the matching degree between the various types of tags of the browsed historical network resources and the user account.

[0022] Optionally, for each first network resource, determining the user account corresponding to the first network resource from among the user accounts, based on the matching degree between the first network resource and each user account, and using it as the first user account, includes:

[0023] For each first network resource, the first number of user accounts corresponding to the first network resource are determined from the user accounts in descending order of the matching degree between the first network resource and each user account, and these are taken as the first user accounts.

[0024] or,

[0025] The user account whose matching degree with the first network resource is greater than the first threshold is determined from the user accounts and is designated as the first user account.

[0026] Optionally, for each first user account, determining the third network resource corresponding to the first user account from among the multiple network resources based on the matching degree between the first user account and the second network resources other than the first network resources includes:

[0027] For each second network resource other than the first network resource among the plurality of network resources, the first user account corresponding to the second network resource is determined from the first user accounts according to the matching degree between each first user account and the second network resource; wherein, one second network resource corresponding to one first user account means: sending the second network resource to the client logged in with the first user account;

[0028] For each first user account, the first second number of second network resources corresponding to the first user account are determined from the second network resources in descending order of the matching degree between the first user account and each second network resource, and these are used as candidate network resources.

[0029] If the candidate network resource is a second network resource corresponding to the first user account, then the candidate network resource is determined to be a third network resource corresponding to the first user account.

[0030] Optionally, for each first network resource, the first number corresponding to the first network resource decreases as the first time interval increases; the first threshold corresponding to the first network resource increases as the first time interval increases; the first time interval is: the time interval between the publication time of the first network resource and the current time.

[0031] For each second network resource, the number of first user accounts corresponding to the second network resource decreases as the second time interval increases; the second time interval is the time interval between the publication time of the second network resource and the current time.

[0032] Optionally, the decrease in the number of first network resources corresponding to the first network resource as the first time interval increases is less than the decrease in the number of first user accounts corresponding to the second network resource as the second time interval increases.

[0033] Optionally, before acquiring the preset plurality of network resources, the method further includes:

[0034] Receive a request to acquire a target network resource carrying the identifier of the first user account;

[0035] After determining the first network resource corresponding to the first user account and the third network resource corresponding to the first user account as the target network resource corresponding to the first user account, the method further includes:

[0036] Send the target network resources corresponding to the first user account to the client logged in with the first user account.

[0037] Optionally, after determining the user account corresponding to the first network resource from among the user accounts based on the matching degree between the first network resource and each user account for each first network resource, and using it as the first user account, the method further includes:

[0038] For each second user account other than the first user account, based on the matching degree between the second user account and the second network resources other than the first network resources, the fourth network resource corresponding to the second user account is determined from the second network resources.

[0039] Send the fourth network resource corresponding to the second user account to the client logged in with the second user account.

[0040] In a second aspect of the invention, a resource determination apparatus is also provided, the apparatus comprising:

[0041] The acquisition module is used to acquire multiple preset network resources; the multiple network resources include: a first network resource sent to a client logged in with a first user account;

[0042] The matching degree determination module is used to determine the matching degree between each network resource and each user account based on the type label of the network resource.

[0043] The first user account determination module is used to determine the user account corresponding to the first network resource from among the user accounts for each first network resource based on the matching degree between the first network resource and each user account, and use it as the first user account.

[0044] The first resource determination module is used to determine, for each first user account, the third network resource corresponding to the first user account from each of the second network resources based on the matching degree between the first user account and the second network resources other than the first network resources among the plurality of network resources;

[0045] The second resource determination module is used to determine the first network resource corresponding to the first user account and the third network resource corresponding to the first user account, as the target network resource corresponding to the first user account.

[0046] Optionally, the matching degree determination module is specifically used for:

[0047] For each user account, obtain the user's operation information on the historical network resources browsed by the user account to obtain the first operation information;

[0048] Based on the first operation information and the type tags of the historical network resources that have been viewed, determine the matching degree between each type tag of the historical network resources that have been viewed and the user account.

[0049] For each network resource, determine the type tag that the network resource has from the various type tags of the historical network resources that have been browsed;

[0050] The matching degree between the network resource and the user account is calculated based on the matching degree between the type tag of the network resource and the user account.

[0051] Optionally, the first operation information includes: the number of operations performed by the user to which the user account belongs on historical network resources that have been viewed; the operation includes at least one of the following: click operation, forward operation, comment operation;

[0052] The matching degree determination module is specifically used for:

[0053] For each type of tag in the historical web resources that have been viewed, count the number of operations performed by the user account on the historical web resources with that type of tag, and use that count as the number of operations performed by the user account on that type of tag.

[0054] Generate a target feature vector containing the number of operations performed by the user account for each type of tag;

[0055] The target feature vector is input into a pre-trained probabilistic prediction model to obtain the matching degree between the various types of tags of the browsed historical network resources and the user account.

[0056] Optionally, the first user account determination module is specifically used for:

[0057] For each first network resource, the first number of user accounts corresponding to the first network resource are determined from the user accounts in descending order of the matching degree between the first network resource and each user account, and these are taken as the first user accounts.

[0058] or,

[0059] The user account whose matching degree with the first network resource is greater than the first threshold is determined from the user accounts and is designated as the first user account.

[0060] Optionally, the first resource determination module is specifically used for:

[0061] For each second network resource other than the first network resource among the plurality of network resources, the first user account corresponding to the second network resource is determined from the first user accounts according to the matching degree between each first user account and the second network resource; wherein, one second network resource corresponding to one first user account means: sending the second network resource to the client logged in with the first user account;

[0062] For each first user account, the first second number of second network resources corresponding to the first user account are determined from the second network resources in descending order of the matching degree between the first user account and each second network resource, and these are used as candidate network resources.

[0063] If the candidate network resource is a second network resource corresponding to the first user account, then the candidate network resource is determined to be a third network resource corresponding to the first user account.

[0064] Optionally, for each first network resource, the first number corresponding to the first network resource decreases as the first time interval increases; the first threshold corresponding to the first network resource increases as the first time interval increases; the first time interval is: the time interval between the publication time of the first network resource and the current time.

[0065] For each second network resource, the number of first user accounts corresponding to the second network resource decreases as the second time interval increases; the second time interval is the time interval between the publication time of the second network resource and the current time.

[0066] Optionally, the decrease in the number of first network resources corresponding to the first network resource as the first time interval increases is less than the decrease in the number of first user accounts corresponding to the second network resource as the second time interval increases.

[0067] Optionally, the device further includes:

[0068] The receiving module is configured to receive a target network resource acquisition request carrying the identifier of a first user account before the acquisition module performs the acquisition of multiple preset network resources.

[0069] The first sending module is configured to send the target network resource corresponding to the first user account to the client logged in with the first user account after the second resource determining module performs the process of determining the first network resource corresponding to the first user account and the third network resource corresponding to the first user account as the target network resource corresponding to the first user account.

[0070] Optionally, the device further includes:

[0071] The third resource determination module is used to, after the first user account determination module performs the following steps for each first network resource, determine the user account corresponding to the first network resource from each user account based on the matching degree between the first network resource and each user account, and then, for each second user account other than the first user account, determine the fourth network resource corresponding to the second user account from the second network resources based on the matching degree between the second user account and the second network resources other than the first network resources among the plurality of network resources;

[0072] The second sending module is used to send the fourth network resource corresponding to the second user account to the client that is logged in with the second user account.

[0073] In a third aspect of the present invention, an electronic device is provided, including a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other through the communication bus;

[0074] Memory, used to store computer programs;

[0075] When a processor executes a program stored in memory, it implements the resource determination method steps described in any of the first aspects above.

[0076] In another aspect of the present invention, a computer-readable storage medium is provided, wherein a computer program is stored therein, and when the computer program is executed by a processor, it implements the resource determination method described in any of the first aspects above.

[0077] In another aspect of the present invention, a computer program product containing instructions is also provided, which, when run on a computer, causes the computer to perform any of the resource determination methods described in the first aspect above.

[0078] Based on the above processing, the first network resource is the network resource sent to the client logged into the first user account from among a pre-defined pool of network resources. By identifying the first network resource as the target network resource, the target network resource includes network resources that can meet the service provider's needs. The third network resource corresponding to the first user account is determined based on the matching degree between the first user account and each second network resource; that is, the third network resource is a more suitable network resource for the first user account, and consequently, the user to whom the first user account belongs is more likely to be interested in the third network resource. By identifying the third network resource as the target network resource, the target network resource also includes network resources that are more suitable for the first user account, thus meeting the user's personalized needs. In other words, the identified target network resources can guarantee both the service provider's needs and the user's personalized needs, thereby achieving a balance between the service provider's needs and the user's personalized needs. Attached Figure Description

[0079] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below.

[0080] Figure 1 This is a first flowchart of a resource determination method provided in an embodiment of the present invention;

[0081] Figure 2 This is a second flowchart of the resource determination method provided in an embodiment of the present invention;

[0082] Figure 3 This is a third flowchart of the resource determination method provided in the embodiments of the present invention;

[0083] Figure 4 This is a fourth flowchart of the resource determination method provided in the embodiments of the present invention;

[0084] Figure 5 A fifth flowchart of the resource determination method provided in the embodiments of the present invention;

[0085] Figure 6 A sixth flowchart of the resource determination method provided in the embodiments of the present invention;

[0086] Figure 7 A seventh flowchart of the resource determination method provided in the embodiments of the present invention;

[0087] Figure 8An eighth flowchart of the resource determination method provided in the embodiments of the present invention;

[0088] Figure 9 A structural diagram of a resource determination device provided in an embodiment of the present invention;

[0089] Figure 10 This is a structural diagram of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0090] The technical solutions of the present invention will now be described with reference to the accompanying drawings in the embodiments of the present invention.

[0091] In existing technologies, service providers need to send a large number of new, trending resources to clients to attract users and increase their activity. Users, on the other hand, need to obtain more network resources that match their own interests and personalized requirements to improve user experience. Consequently, when clients display received network resources to users, it is difficult to simultaneously guarantee both the service provider's needs and the user's personalized requirements because it is challenging to strike a balance between new, trending resources and resources that match the user's interests and personalized needs.

[0092] To address the aforementioned problems, this invention provides a resource determination method applied to an electronic device, which can serve as a server for a service provider to send resources to clients. The electronic device acquires a plurality of preset network resources; these network resources include: a first network resource to be sent to a client logged into a first user account; for each network resource, determining the matching degree between the network resource and each user account based on the network resource's type tag; for each first network resource, determining the user account corresponding to the first network resource from among the user accounts based on the matching degree between the first network resource and each user account, and designating it as the first user account; for each first user account, determining the third network resource corresponding to the first user account from among the second network resources based on the matching degree between the first user account and a second network resource (excluding the first network resource); and determining the first network resource corresponding to the first user account and the third network resource corresponding to the first user account as the target network resource corresponding to the first user account. The determined target network resource can ensure both the service provider's needs and the user's personalized needs, thereby achieving a balance between these two requirements.

[0093] See Figure 1 , Figure 1 A first flowchart of a resource determination method provided in an embodiment of the present invention may include the following steps:

[0094] S101: Obtain multiple preset network resources.

[0095] Multiple network resources include: the first network resource sent to a client logged in with the first user account.

[0096] S102: For each network resource, determine the matching degree between the network resource and each user account based on the type label of the network resource.

[0097] S103: For each first network resource, based on the matching degree between the first network resource and each user account, determine the user account corresponding to the first network resource from each user account and use it as the first user account.

[0098] S104: For each first user account, based on the matching degree between the first user account and a second network resource other than the first network resource among multiple network resources, determine the third network resource corresponding to the first user account from each second network resource.

[0099] S105: Determine the first network resource corresponding to the first user account and the third network resource corresponding to the first user account as the target network resource corresponding to the first user account.

[0100] Based on the resource determination method provided in this embodiment of the invention, the first network resource is a network resource sent to a client logged into a first user account from among a preset plurality of network resources. Determining the first network resource as the target network resource means that the target network resource includes network resources that can meet the needs of the service provider. The third network resource corresponding to the first user account is determined based on the matching degree between the first user account and each second network resource; that is, the third network resource is a network resource that is more suitable for the first user account, and correspondingly, the user to whom the first user account belongs is more likely to be interested in the third network resource. Determining the third network resource as the target network resource means that the target network resource also includes network resources that are more suitable for the first user account, which can meet the user's personalized needs. In other words, the determined target network resource can guarantee both the needs of the service provider and the personalized needs of the user, thereby achieving a balance between the needs of the service provider and the personalized needs of the user.

[0101] Regarding step S101, the multiple network resources can be network resources that the electronic device can currently provide to the user. For example, the network resources that the electronic device can currently provide to the user can be videos, music, etc. The multiple network resources include a first network resource sent to a client logged into the first user account.

[0102] To ensure the needs of service providers are met, electronic devices can identify the first network resource as the target network resource. For example, the first network resource could be a network resource that has been clicked many times, or a network resource that the service provider has participated in creating.

[0103] To ensure personalized user needs, electronic devices can select a second network resource from multiple network resources (excluding the first network resource) that is more suitable for the first user account. For example, the second network resource could be a network resource with fewer clicks, a newly released network resource, etc.

[0104] For step S102, for each network resource, the type label of the network resource can be pre-set, or it can be extracted by feature extraction.

[0105] The type tag of a web resource can indicate the content it includes. For example, the type tag can indicate the genre of the web resource, its release date, etc. A type tag for a web resource could include: Actor A, sitcom, new July release, etc., indicating that the lead actor is A, the web resource is a sitcom, and its release date is July.

[0106] In some embodiments, Figure 1 Based on this, see Figure 2 Step S102 may include the following steps:

[0107] S1021: For each user account, obtain the operation information of the user to which the user account belongs regarding the historical network resources that have been browsed, and obtain the first operation information.

[0108] S1022: Based on the first operation information and the type tags of the historical network resources that have been viewed, determine the matching degree between the type tags of the historical network resources that have been viewed and the user account.

[0109] S1023: For each network resource, determine the type tag that the network resource has from the type tags of the historical network resources that have been browsed.

[0110] S1024: Calculate the matching degree between the network resource and the user account based on the matching degree between the type label of the network resource and the user account.

[0111] For each user account, the user to whom the user account belongs is the user who uses that user account to browse network resources.

[0112] Historical online resources include those viewed by users belonging to the user account. For each user belonging to that account, while browsing online resources, that user may take actions related to resources that interest them. For example, they might forward resources of interest to other users or post comments about resources they are interested in. Accordingly, the electronic device can obtain the user's first action information regarding the viewed historical online resources. This first action information can represent the probability that the user is interested in historical online resources, and consequently, it can represent the matching degree between the user account and historical online resources.

[0113] In some embodiments, the first operation information includes: the number of operations performed by the user to which the user account belongs on historical network resources that have been viewed; the operation includes at least one of the following: click operation, forward operation, comment operation.

[0114] Correspondingly, in Figure 2 Based on this, see Figure 3 Step S1022 may include the following steps:

[0115] S10221: For each type of tag of the historical network resources that have been viewed, count the number of operations performed by the user account on the historical network resources with that type of tag, and use that count as the number of operations performed by the user account on that type of tag.

[0116] S10222: Generate a target feature vector containing the number of operations performed by the user account on each type of tag.

[0117] S10223: Input the target feature vector into the pre-trained probabilistic prediction model to obtain the matching degree between the various types of tags of the browsed historical network resources and the user account.

[0118] For each historical web resource that has been viewed, since that historical web resource may have multiple type tags, meaning different historical web resources may have the same type tag, after the electronic device obtains the first operation information of the user account to which the user belongs regarding the historical web resources that have been viewed, for each type tag, the electronic device can count the number of operations of the user account to which the user belongs regarding historical web resources with that type tag, and use the count result as the number of operations of the user account for that type tag.

[0119] For example, an electronic device can calculate the sum of the number of operations performed by the user account on historical network resources with that type of label, or calculate the average number of operations performed by the user account on historical network resources with that type of label.

[0120] Furthermore, to prevent users of a particular account from accidentally accessing network resources with a certain type of tag, thus affecting the subsequent determination of the target network resource corresponding to that user account, the electronic device can perform data cleaning on the number of operations performed by the user of that user account for each type of tag. Specifically, for each type of tag, if the electronic device obtains a low number of operations performed by the user of that user account for that type of tag, it indicates that the user of that user account may have accidentally accessed a network resource with that type of tag. Therefore, the electronic device can record the number of operations performed by that user account for that type of tag as 0.

[0121] Subsequently, for each user account, the electronic device can generate a target feature vector containing the number of operations performed by that user account on each type of label, thus obtaining the target feature vector corresponding to that user account. Each element in the target feature vector corresponds one-to-one with each type of label, with each element representing the number of operations performed by the user account on the type of label corresponding to that element. The target feature vector can represent the degree of matching between the user account and each type of label.

[0122] For example, the category tags of the historical online resources browsed include: Actor A, Actor B, Sitcom, Suspense, Historical Drama, and New July Drama. If the user account has performed 10 operations on the category tag "Actor A", 20 operations on the category tag "Actor B", 30 operations on the category tag "Sitcom", 40 operations on the category tag "Suspense", 50 operations on the category tag "Historical Drama", and 60 operations on the category tag "New July Drama", then the target feature vector generated by the electronic device, which contains the number of operations performed by the user account on each category tag, can be represented as: (10, 20, 30, 40, 50, 60).

[0123] In some embodiments, in order to improve the efficiency of resource determination, for each user account, the electronic device may also normalize the target feature vector containing the number of operations of the user account for each type of tag, and then use the result of the normalization process as the target feature vector corresponding to the user account.

[0124] Then, the electronic device inputs the target feature vector corresponding to the user account into a pre-trained probabilistic prediction model. It then obtains the matching degree between the various types of tags of the browsed historical web resources and the user account, as output by the probabilistic prediction model. The probabilistic prediction model can be a DIN (Deep Interest Network) model, or a CAN (Co-Action Network) model, but is not limited to these.

[0125] In some embodiments, the probabilistic prediction model can be trained based on the sample feature vector corresponding to the sample user account and the sample matching degree between the sample type label of the sample network resource and the sample user account. For each sample user account, the sample feature vector corresponding to the sample user account contains the number of operations performed by the sample user account for each sample type label.

[0126] For each sample user account, the electronic device can input the sample feature vector corresponding to that sample user account into the probabilistic prediction model. The probabilistic prediction model outputs the predicted matching degree between the sample network resource's sample type label and the sample user account. Then, based on the difference between the predicted matching degree and the sample matching degree, the parameters of the probabilistic prediction model are adjusted, and the probabilistic prediction model is trained.

[0127] In some embodiments, in order to improve the accuracy of the probabilistic prediction model, for each user account, the electronic device can also periodically acquire the operation information of the user account on the network resources displayed in the client logged in with the user account, and generate the target feature vector corresponding to the user account based on the acquired operation information of the user account on the network resources displayed in the client logged in with the user account, as a sample feature vector for training the probabilistic prediction model, and continue to train the probabilistic prediction model based on the sample feature vector.

[0128] Based on the above processing, the probabilistic prediction model is trained using the target feature vectors corresponding to each user account obtained from real-world application scenarios. This improves the accuracy of the matching between the network resource labels output by the probabilistic prediction model and the user accounts. Furthermore, subsequent electronic devices can continue to adjust the model parameters of the probabilistic prediction model based on the target feature vectors corresponding to each user account. This allows the probabilistic prediction model to better adapt to the interests and personalized needs of the users associated with each account, thereby enhancing the user experience.

[0129] For each user account, after the electronic device determines the matching degree between the various type tags of the historical network resources browsed and the user account, for each network resource, it can determine the type tag of the network resource from the type tags of the historical network resources browsed by the user account.

[0130] Furthermore, the electronic device can calculate the matching degree between the user account and the network resource based on the matching degree between the various types of tags of the network resource and the user account. For example, the electronic device can use the highest matching degree among the matching degrees of the various types of tags of the network resource and the user account as the matching degree between the user account and the network resource. Alternatively, the electronic device can calculate the average matching degree of the types of tags of the network resource and the user account, and use the calculation result as the matching degree between the user account and the network resource, but it is not limited to these methods.

[0131] In step S103, the electronic device determines the matching degree between each network resource and each user account, thus determining the matching degree between each user account and the first network resource, as well as the matching degree between each user account and each second network resource. Accordingly, for each first network resource, the electronic device can determine the first user account with a higher matching degree to that first network resource based on the matching degree between each user account and the first network resource. Consequently, the user belonging to the first user account is more likely to be interested in that first network resource.

[0132] In some embodiments, Figure 1 Based on this, see Figure 4 Step S103 may include the following steps:

[0133] S1031: For each first network resource, in descending order of the matching degree between the first network resource and each user account, determine the first number of user accounts corresponding to the first network resource as the first user accounts; or, determine the user accounts whose matching degree with the first network resource is greater than a first threshold as the first user accounts.

[0134] For each first network resource, after obtaining the matching degree between each user account and the first network resource, the electronic device can sort the user accounts in descending order of matching degree, with the user account ranking higher in the sorting having a greater matching degree with the first network resource.

[0135] Therefore, the electronic device can determine the first number of user accounts corresponding to the first network resource from the sorted user accounts, and use them as the first user accounts. The first number can be determined according to the actual needs of the service provider. For example, the first number can be 20% of the total number of user accounts, or it can be 30% of the total number of user accounts, but it is not limited to this.

[0136] Alternatively, a first threshold can be preset. For each user account, if the matching degree between the user account and the first network resource is greater than the first threshold, it indicates that the user to whom the user account belongs is more likely to be interested in the first network resource. Accordingly, the electronic device identifies user accounts from among the user accounts whose matching degree with the first network resource is greater than the first threshold as the first user account. The first threshold can be determined according to the actual needs of the service provider. For example, the first threshold can be 0.4, or it can be 0.5, but it is not limited to these.

[0137] If the electronic device identifies a large number of user accounts whose matching degree with the first network resource is greater than the first threshold, the electronic device then selects a first number of user accounts from the identified user accounts as the first user accounts.

[0138] Based on the above processing, for each first network resource, the electronic device can determine the first user account with a higher matching degree with the first network resource from among the user accounts according to the matching degree between each user account and the first network resource, thereby improving the accuracy of the determined first user account.

[0139] Regarding steps S104 and S105, for each first user account, the target network resources corresponding to that first user account include the first network resources corresponding to that first user account, which can meet the service provider's needs. Furthermore, in order to achieve a balance between the service provider's needs and the user's personalized needs, the electronic device also needs to determine the network resources that satisfy the interests and personalized needs of the user to which the first user account belongs as target network resources, thereby satisfying the personalized needs of the user to which the first user account belongs.

[0140] In some embodiments, Figure 4 Based on this, see Figure 5 Step S104 may include the following steps:

[0141] S1041: For each second network resource other than the first network resource among multiple network resources, determine the first user account corresponding to the second network resource from among the first user accounts based on the matching degree between each first user account and the second network resource.

[0142] In this context, the correspondence between a second network resource and a first user account indicates that the second network resource is sent to a client logged into the first user account.

[0143] S1042: For each first user account, in descending order of the matching degree between the first user account and each second network resource, determine the second-largest number of second network resources corresponding to the first user account as candidate network resources.

[0144] S1043: If the candidate network resource is the second network resource corresponding to the first user account, determine the candidate network resource as the third network resource corresponding to the first user account.

[0145] For each first user account, in order to meet the personalized needs of the user to which the first user account belongs, the electronic device needs to first determine from the various second network resources the alternative network resources that match the user's interests and personalized needs.

[0146] The electronic device determines the matching degree between each user account and each network resource, thus determining the probability that the user to which each user account belongs is interested in each second network resource. Furthermore, for each second network resource, the electronic device sorts the first user accounts in descending order of their matching degree with that second network resource; the higher the ranking of the first user account, the greater the probability that the user to which it belongs is interested in that second network resource.

[0147] Therefore, the electronic device can determine the second number of first user accounts corresponding to the second network resource from the sorted first user accounts, and use these as the first user accounts corresponding to the second network resource. The second number can be determined according to the actual needs of the service provider. For example, the second number can be 80% of the total number of user accounts, or it can be 90% of the total number of user accounts, but it is not limited to these.

[0148] Alternatively, a second threshold can be preset. For each first user account, if the matching degree between the first user account and the second network resource is greater than the second threshold, it indicates that the user belonging to the first user account is more likely to be interested in the second network resource. Accordingly, the electronic device determines the first user accounts from among the first user accounts whose matching degree with the second network resource is greater than the second threshold, and uses these as the first user accounts corresponding to the second network resource. The second threshold can be determined according to the actual needs of the service provider. For example, the second threshold can be 0.1, or it can be 0.2, but it is not limited to these.

[0149] If the electronic device identifies a large number of first user accounts corresponding to the second network resource, the electronic device will then select a second number of first user accounts from the identified first user accounts as the first user accounts corresponding to the second network resource.

[0150] Furthermore, for each first user account, the electronic device determines the second network resources that the user to which the first user account belongs may be interested in based on the matching degree between the first user account and each second network resource. That is, from each second network resource, the first second number of second network resources corresponding to the first user account are determined as candidate network resources.

[0151] After obtaining the matching degree between the first user account and the second network resource, the electronic device can sort the second network resources in descending order of matching degree between the first user account and each second network resource. The user to whom the first user account belongs is more interested in the second network resource that ranks higher in the sort.

[0152] Therefore, the electronic device can determine the top third number of second network resources corresponding to the first user account from the sorted second network resources, and use these as the second network resources corresponding to the first user account. The third number is preset by the service provider, and the third number is not less than the total number of network resources that the service provider presets to be sent to the client logged into the first user account.

[0153] If the alternative network resource is the second network resource corresponding to the first user account, meaning the electronic device can determine that the alternative network resource matches the interests and personalized needs of the user to whom the first user account belongs, then the electronic device can determine the alternative network resource as the third network resource corresponding to the first user account. Subsequently, the electronic device can determine the third network resource corresponding to the first user account as the target network resource corresponding to the first user account, thus ensuring the user's personalized needs are met.

[0154] The number of third network resources corresponding to the first user account can be the difference between the total number of network resources that need to be sent to clients logged into the first user account and the number of first network resources corresponding to the first user account. The total number of network resources that need to be sent to clients logged into the first user account can be preset by the service provider. For example, if the service provider presets that 8 network resources need to be sent to each client logged into the first user account, and the number of first network resources corresponding to the first user account is 3, then the number of third network resources corresponding to the first user is 5.

[0155] Based on the above processing, the third network resource is the network resource that the user of the first user account is interested in. By determining the third network resource corresponding to the first user account as the target network resource, the target network resource can also guarantee the user's personalized needs and improve the user experience.

[0156] In some embodiments, the electronic device can sort the second network resources according to their publication time to obtain a sorted resource queue, with the second network resource published later appearing earlier in the resource queue. For each second network resource, the corresponding second number can correspond to its position in the resource queue. The second number corresponding to each second network resource in the resource queue can be adjusted according to actual needs.

[0157] For example, if the second network resource is the most recently released second network resource, then it is the first second network resource in the resource queue, and the second number corresponding to this second network resource can be 100% of the total number of user accounts, meaning this second network resource is the target network resource for all user accounts. When a new second network resource is released, it becomes the second second network resource in the resource queue, and the second number corresponding to the second second network resource in the resource queue can be 90% of the total number of user accounts. The second number corresponding to the third second network resource in the resource queue can be 80% of the total number of user accounts. The second number corresponding to the fourth second network resource in the resource queue can be 70% of the total number of user accounts. The second number corresponding to the fifth second network resource in the resource queue can be 50% of the total number of user accounts. The second number corresponding to the sixth second network resource in the resource queue can be 20% of the total number of user accounts.

[0158] In some embodiments, for each first network resource, the first number corresponding to the first network resource decreases as the first time interval increases; the first threshold corresponding to the first network resource increases as the first time interval increases; the first time interval is the time interval between the publication time of the first network resource and the current time.

[0159] The first number corresponding to the first network resource can decrease stepwise as the first time interval increases, and the first threshold corresponding to the first network resource can increase stepwise as the first time interval increases. For example, when the first time interval is 1 week, the first number can be 30% of the total number of user accounts, and the first threshold can be 0.4; when the first time interval is 2 weeks, the first number can be 20% of the total number of user accounts, and the first threshold can be 0.5.

[0160] For each second network resource, the number of first user accounts corresponding to the second network resource decreases as the second time interval increases; the second time interval is the time interval between the publication time of the second network resource and the current time.

[0161] The second number corresponding to the second network resource can decrease in a stepwise manner as the second time interval increases. For example, when the second time interval is 1 day, the second number can be 90% of the total number of user accounts; when the second time interval is 2 days, the second number can be 80% of the total number of user accounts.

[0162] Alternatively, the second number corresponding to the second network resource can decrease functionally as the second time interval increases, such as linear decay, logarithmic decay, etc.

[0163] Alternatively, the second number corresponding to the second network resource can decrease stepwise as the second network resource is positioned in the resource queue, and decrease functionally as the second time interval increases while the position of the second network resource in the resource queue remains unchanged.

[0164] In some embodiments, the magnitude by which the number of first network resources corresponding to the first network resources decreases as the first time interval increases is less than the magnitude by which the number of first user accounts corresponding to the second network resources decreases as the second time interval increases.

[0165] For example, the decrease in the number of first network resources corresponding to the first number as the first time interval increases can be: after the first time interval increases by 1 week, the first number decreases by 10% of the total number of user accounts; the decrease in the number of first user accounts corresponding to the second network resources as the second time interval increases can be: after the second time interval increases by 1 day, the second number decreases by 10% of the total number of user accounts.

[0166] In some embodiments, Figure 1 Based on this, see Figure 6 Before step S101, the method may further include the following steps:

[0167] S106: Receive a request to acquire target network resources carrying the identifier of the first user account.

[0168] Accordingly, after step S105, the method may further include the following steps:

[0169] S107: Send the target network resources corresponding to the first user account to the client logged in with the first user account.

[0170] When a user needs to browse online resources, the user can use their first user account to input a target online resource acquisition command into the client. Correspondingly, the electronic device can also receive the target online resource acquisition request carrying the first user account.

[0171] Furthermore, in order to improve user experience, for each first user account, since the first user account has a high degree of matching with the first network resource corresponding to the first user account, the probability that the user to which the first user account belongs is also greater is higher. The electronic device can identify the first network resource corresponding to the first user account as the target network resource, and from each second network resource other than the first network resource, identify the third network resource that meets the interests and personalized needs of the user to which the first user account belongs, and identify the third network resource corresponding to the first user account as the target network resource.

[0172] Compared to related technologies, where electronic devices directly identify all first network resources as target network resources corresponding to the first user account, since the first network resources included in the target network resources are network resources with a high degree of matching with the first user account, it can avoid identifying network resources with a low degree of matching with the first user account as target network resources, thus improving the user experience.

[0173] Compared to related technologies, where electronic devices directly identify all second network resources as target network resources corresponding to the first user account, the third network resources included in the target network resources are network resources that have a high degree of matching with the first user account and meet the interests, preferences, and personalized needs of the user to whom the first user account belongs. Therefore, network resources that meet the interests, preferences, and personalized needs of the user to whom the first user account belongs can be identified as target network resources, which can improve the user experience.

[0174] Subsequently, the electronic device can send the identified target network resources to the client logged into the first user account. The client can then display the received target network resources, effectively recommending them to the user.

[0175] Based on the above processing, the identified target network resources can guarantee both the service provider's needs and the user's personalized needs, thus achieving a balance between the two. Subsequently, sending the identified target network resources to the client logged into the first user account allows for the recommendation of the target network resources to the user belonging to that account, thereby meeting the user's needs.

[0176] In some embodiments, when a user browses network resources through a client, they typically first view the network resources displayed in the upper part of the display interface. For each user belonging to a first user account, the electronic device can display the first network resource corresponding to that first user account in the upper part of the display interface, or it can display the third network resource corresponding to that first user account in the upper part of the display interface.

[0177] Alternatively, the electronic device can sort the first network resources and third network resources corresponding to the first user account in descending order of matching degree between the first user account and each network resource, and then display the sorted first network resources and third network resources in the client's display interface from top to bottom.

[0178] Alternatively, a display area for displaying target network resources can be pre-set in the client's display interface. The electronic device sorts the first network resources and third network resources in the target network resources corresponding to the first user account, and then displays the sorted first network resources and third network resources in the display area in a scrolling manner.

[0179] In some embodiments, Figure 1 Based on this, see Figure 7 After step S103, the method may further include the following steps:

[0180] S108: For each second user account other than the first user account, determine the fourth network resource corresponding to the second user account from the second network resources based on the matching degree between the second user account and the second network resources other than the first network resources.

[0181] S109: Send the fourth network resource corresponding to the second user account to the client logged in with the second user account.

[0182] For each second user account other than the first user account, since the matching degree between the second user account and each first network resource is low, it means that the user to which the second user account belongs is not interested in each first network resource. Therefore, it is not necessary to determine the first network resource as the target network resource corresponding to the second user account. In order to determine the target network resource corresponding to the second user account, the electronic device needs to determine the network resource with a high matching degree with the second user account from the second network resources, that is, the fourth network resource corresponding to the second user account.

[0183] The way in which the electronic device determines the fourth network resource corresponding to the second user account from each of the second network resources based on the matching degree between the second user account and each of the second network resources is similar to the way in the foregoing embodiment that the third network resource corresponding to the first user account is determined from each of the second network resources based on the matching degree between the first user account and each of the second network resources. Please refer to the relevant description in the foregoing embodiment.

[0184] Based on the above processing, the fourth network resource is a network resource with a high degree of matching with the second user account, that is, a network resource that the user of the second user account is interested in. Sending the fourth network resource corresponding to the second user account to the client logged in with the second user account can recommend network resources that match the user's interests and personalized needs, thereby satisfying the personalized needs of the user of the second user account and improving the user experience.

[0185] In some embodiments, see Figure 8 , Figure 8 An eighth flowchart of a resource determination method provided in an embodiment of the present invention may include the following steps:

[0186] S801: Preparation phase.

[0187] In the steps, the preparation phase is performed by the data preparation module, the deep personalized recommendation model training module, and the personalized score distribution construction module.

[0188] The data preparation module first collects user behavior data, content information, and other relevant data of the users to which the sample user accounts belong; and performs data preprocessing and feature extraction. For each sample user account, the user behavior data of the users to which the sample user account belongs is: the number of operations performed by the sample user account on each sample type tag in the aforementioned embodiments. Content information and other relevant data include the sample type tags of the sample network resources in the aforementioned embodiments. For each sample network resource, the electronic device obtains the sample type tag of the sample network resource through feature extraction. A sample type tag of a sample network resource can indicate the content included in the sample network resource.

[0189] For each sample user account, the electronic device obtains the user's operation information on the browsed sample network resources. For each sample type tag of the browsed sample network resources, it counts the number of operations performed by the user on sample network resources with that sample type tag, and uses this count as the user's operation count for that sample type tag. A sample feature vector containing the user's operation count for each sample type tag is then generated, resulting in the sample feature vector corresponding to the user account. Furthermore, the electronic device can perform data preprocessing, such as data cleaning of the obtained operation counts for each sample type tag, or normalizing the sample feature vector generated based on the user's operation counts for each sample type tag, and using the normalized result as the sample feature vector corresponding to the user account.

[0190] Subsequently, the electronic device inputs the feature vectors of each sample into the deep personalized recommendation model training module to train the personalized recommendation model.

[0191] The deep personalized recommendation model training module trains the personalized recommendation model. This personalized recommendation model is the probabilistic prediction model described in the previous embodiment, which can be a DIN model. Based on the feature vectors of each sample input from the data preparation module, the DIN model is trained to obtain a trained DIN model.

[0192] Next, the data preparation module collects user behavior data, content information, and other relevant data of the user account to which the target network resource needs to be acquired; and performs data preprocessing and feature extraction to obtain the target feature vector corresponding to the user account to which the target network resource needs to be acquired. Then, the target feature vector corresponding to the user account to which the network resource needs to be acquired is input into the trained DIN model.

[0193] The personalized score distribution module constructs a personalized score distribution based on the characteristics of new and trending content and the user's personalized score. The new and trending content characteristics are the type tags of each network resource in the aforementioned embodiments; the user's personalized score is the matching degree between the user account and the type tags of the historically viewed network resources in the aforementioned embodiments. The personalized score distribution includes the personalized scores of each user. The electronic device traverses all users and content, where the content is the network resource in the aforementioned embodiments; it calculates the score for each user account for each piece of content and statistically obtains the personalized score distribution of all user accounts. The personalized score distribution of all users can represent the overall matching degree of each user account with different content, providing a basis for the quantity preservation stage and the personalization stage in the subsequent two-stage distribution model.

[0194] For each user account, the electronic device uses a pre-trained probabilistic prediction model to obtain the matching degree between the user account and the type tags of the historically viewed online resources. This yields the matching degree of each user account with the type tags of the historically viewed online resources. Subsequently, based on the matching degree of each user account with the type tags of the historically viewed online resources, the electronic device determines the target online resource corresponding to each user account.

[0195] S802: Quantity Preservation Algorithm Design Phase: Select users with a high interest in new and trending content based on the characteristics of the new and trending content and the personalized score distribution of users.

[0196] In this step, "quantity preservation" refers to the aforementioned embodiment, where for each network resource, the network resource is sent to a number of clients logged into user accounts corresponding to that network resource. The characteristics of the new and trending content are the type tags possessed by the first network resource in the aforementioned embodiment. Users with a high interest in the new and trending content are the first users belonging to the first user accounts with a high degree of matching with the first network resource in the aforementioned embodiment.

[0197] For each first network resource, the electronic device calculates the matching degree between each user account and the first network resource based on the matching degree of the type tags of each user account and the first network resource, and determines the user account corresponding to the first network resource from among the user accounts, which is then used as the first user account. In other words, the first network resource corresponding to each first user account is determined.

[0198] S803: Differentiated User Acquisition Phase: Identify high-quality users for different new and trending content.

[0199] In this process, differentiated content delivery refers to the process described in the previous embodiment where, for each first user account, in determining the first network resource corresponding to that first user account, it is also necessary to determine a third network resource that matches the user's interests and personalized needs. A high-quality user refers to the first user account corresponding to the second network resource described in the previous embodiment. For different trending content, high-quality users are determined based on the characteristics of the trending content and the user's interest score, and they are provided with differentiated content delivery. Highly trending content receives a higher level of content delivery guarantee, while medium-to-high-trending content receives a moderate level. Highly trending content refers to the most recently published network resource with a high number of clicks; a higher level of content delivery guarantee means a larger number of corresponding third network resources. Medium-to-high-trending content refers to network resources that were published later or have fewer clicks; a medium-to-high level of content delivery guarantee means a smaller number of corresponding third network resources.

[0200] For each second network resource, based on the matching degree between each first user account and the second network resource, the first user account corresponding to the second network resource is determined from among the first user accounts. For each first user account, in descending order of the matching degree between the first user account and each second network resource, the second-to-last number of second network resources corresponding to the first user account are determined as candidate network resources.

[0201] If the candidate network resource is the second network resource corresponding to the first user account, then the candidate network resource is determined to be the third network resource corresponding to the first user account.

[0202] S804: Two-stage distribution mode: In the quantity guarantee stage, new and popular content is distributed to users with high interest; in the personalization stage, based on the personalized score distribution and differentiated quantity guarantee function, higher quality personalized content is recommended to users.

[0203] In this process, the two-stage distribution mode, as described in the previous embodiment, involves the electronic device sending the target network resources corresponding to each first user account to the client logged into that first user account. The electronic device can send both the first network resources and the third network resources corresponding to the first user account to the client logged into that first user account. During the quantity guarantee phase, the electronic device sends the first network resources corresponding to the first user account to the client logged into that first user account; during the personalization phase, the electronic device sends the third network resources corresponding to the first user account to the client logged into that first user account.

[0204] Based on the above processing, the first network resource is the network resource sent to the client logged into the first user account from among a pre-defined pool of network resources. By identifying the first network resource as the target network resource, the target network resource includes network resources that can meet the service provider's needs. The third network resource corresponding to the first user account is determined based on the matching degree between the first user account and each second network resource; that is, the third network resource is a more suitable network resource for the first user account, and consequently, the user to whom the first user account belongs is more likely to be interested in the third network resource. By identifying the third network resource as the target network resource, the target network resource also includes network resources that are more suitable for the first user account, thus meeting the user's personalized needs. In other words, the identified target network resources can guarantee both the service provider's needs and the user's personalized needs, thereby achieving a balance between the service provider's needs and the user's personalized needs.

[0205] Furthermore, based on the characteristics of newly trending content and users' personalized scores, a similarity metric between users and content can be obtained. This similarity metric represents the probability that a user is interested in the type tags of historically viewed online resources. Based on the user-content similarity metric, the characteristics of each content item, and users' personalized scores, a personalized score distribution for each user and a content-preserving algorithm are constructed. In the content-preserving phase, the algorithm selects users with a high interest in newly trending content for distribution, ensuring the efficiency and coverage of content distribution and meeting the service provider's needs for recommending new and trending content. Based on the characteristics of newly trending content and users' interest scores, differentiated content preservation functions are implemented for different content items. In the personalization phase, higher-quality personalized content is recommended to users, providing a more personalized recommendation experience, meeting users' personalized needs, and enhancing their personalized experience. These two phases are interconnected, jointly achieving a balanced distribution of newly trending content and personalized recommendations—that is, achieving a balance between the needs of service providers and the personalized needs of users.

[0206] Based on the same inventive concept as the resource determination method described above, embodiments of the present invention also provide a resource determination apparatus. See also... Figure 9 , Figure 9 A structural diagram of a resource determination device provided in an embodiment of the present invention, the device comprising:

[0207] The acquisition module 901 is used to acquire a plurality of preset network resources; the plurality of network resources include: a first network resource sent to a client logged in with a first user account;

[0208] The matching degree determination module 902 is used to determine the matching degree between each network resource and each user account based on the type label of the network resource.

[0209] The first user account determination module 903 is used to determine the user account corresponding to the first network resource from each user account based on the matching degree between the first network resource and each user account for each first network resource, and use it as the first user account.

[0210] The first resource determination module 904 is used to determine, for each first user account, the third network resource corresponding to the first user account from each second network resource based on the matching degree between the first user account and the second network resource other than the first network resource among the plurality of network resources;

[0211] The second resource determination module 905 is used to determine the first network resource corresponding to the first user account and the third network resource corresponding to the first user account as the target network resource corresponding to the first user account.

[0212] Optionally, the matching degree determination module 902 is specifically used for:

[0213] For each user account, obtain the user's operation information on the historical network resources browsed by the user account to obtain the first operation information;

[0214] Based on the first operation information and the type tags of the historical network resources that have been viewed, determine the matching degree between each type tag of the historical network resources that have been viewed and the user account.

[0215] For each network resource, determine the type tag that the network resource has from the various type tags of the historical network resources that have been browsed;

[0216] The matching degree between the network resource and the user account is calculated based on the matching degree between the type tag of the network resource and the user account.

[0217] Optionally, the first operation information includes: the number of operations performed by the user to which the user account belongs on historical network resources that have been viewed; the operation includes at least one of the following: click operation, forward operation, comment operation;

[0218] The matching degree determination module 902 is specifically used for:

[0219] For each type of tag in the historical web resources that have been viewed, count the number of operations performed by the user account on the historical web resources with that type of tag, and use that count as the number of operations performed by the user account on that type of tag.

[0220] Generate a target feature vector containing the number of operations performed by the user account for each type of tag;

[0221] The target feature vector is input into a pre-trained probabilistic prediction model to obtain the matching degree between the various types of tags of the browsed historical network resources and the user account.

[0222] Optionally, the first user account determination module 903 is specifically used for:

[0223] For each first network resource, the first number of user accounts corresponding to the first network resource are determined from the user accounts in descending order of the matching degree between the first network resource and each user account, and these are taken as the first user accounts.

[0224] or,

[0225] The user account whose matching degree with the first network resource is greater than the first threshold is determined from the user accounts and is designated as the first user account.

[0226] Optionally, the first resource determination module 904 is specifically used for:

[0227] For each second network resource other than the first network resource among the plurality of network resources, the first user account corresponding to the second network resource is determined from the first user accounts according to the matching degree between each first user account and the second network resource; wherein, one second network resource corresponding to one first user account means: sending the second network resource to the client logged in with the first user account;

[0228] For each first user account, the first second number of second network resources corresponding to the first user account are determined from the second network resources in descending order of the matching degree between the first user account and each second network resource, and these are used as candidate network resources.

[0229] If the candidate network resource is a second network resource corresponding to the first user account, then the candidate network resource is determined to be a third network resource corresponding to the first user account.

[0230] Optionally, for each first network resource, the first number corresponding to the first network resource decreases as the first time interval increases; the first threshold corresponding to the first network resource increases as the first time interval increases; the first time interval is: the time interval between the publication time of the first network resource and the current time.

[0231] For each second network resource, the number of first user accounts corresponding to the second network resource decreases as the second time interval increases; the second time interval is the time interval between the publication time of the second network resource and the current time.

[0232] Optionally, the decrease in the number of first network resources corresponding to the first network resource as the first time interval increases is less than the decrease in the number of first user accounts corresponding to the second network resource as the second time interval increases.

[0233] Optionally, the device further includes:

[0234] The receiving module is configured to receive a target network resource acquisition request carrying the identifier of a first user account before the acquisition module 901 performs the acquisition of a plurality of preset network resources.

[0235] The first sending module is configured to send the target network resource corresponding to the first user account to the client logged in with the first user account after the second resource determining module 905 performs the process of determining the first network resource corresponding to the first user account and the third network resource corresponding to the first user account as the target network resource corresponding to the first user account.

[0236] Optionally, the device further includes:

[0237] The third resource determination module is used to, after the first user account determination module 903 performs the following steps: for each first network resource, determine the user account corresponding to the first network resource from each user account based on the matching degree between the first network resource and each user account, and then, for each second user account other than the first user account, determine the fourth network resource corresponding to the second user account from the second network resources based on the matching degree between the second user account and the second network resources other than the first network resources among the plurality of network resources;

[0238] The second sending module is used to send the fourth network resource corresponding to the second user account to the client that is logged in with the second user account.

[0239] Based on the resource determination apparatus provided in this embodiment of the invention, the first network resource is a network resource sent to a client logged into a first user account from among a preset plurality of network resources. Determining the first network resource as the target network resource means that the target network resource includes network resources that can meet the needs of the service provider. The third network resource corresponding to the first user account is determined based on the matching degree between the first user account and each second network resource; that is, the third network resource is a more suitable network resource for the first user account, and correspondingly, the user to whom the first user account belongs is more likely to be interested in the third network resource. Determining the third network resource as the target network resource means that the target network resource also includes network resources that are more suitable for the first user account, which can meet the user's personalized needs. In other words, the determined target network resource can ensure that the needs of the service provider and the personalized needs of the user are met, thereby achieving a balance between the needs of the service provider and the personalized needs of the user.

[0240] This invention also provides an electronic device, such as... Figure 10 As shown, it includes a processor 1001, a communication interface 1002, a memory 1003, and a communication bus 1004, wherein the processor 1001, the communication interface 1002, and the memory 1003 communicate with each other through the communication bus 1004.

[0241] Memory 1003 is used to store computer programs;

[0242] When the processor 1001 executes the program stored in the memory 1003, it implements the steps of any of the resource determination methods in the above embodiments.

[0243] The communication bus mentioned in the above electronic devices can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. This communication bus can be divided into address bus, data bus, control bus, etc. For ease of illustration, only one thick line is used to represent it in the diagram, but this does not indicate that there is only one bus or one type of bus.

[0244] The communication interface is used for communication between the aforementioned electronic devices and other devices.

[0245] The memory may include random access memory (RAM) or non-volatile memory, such as at least one disk storage device. Optionally, the memory may also be at least one storage device located remotely from the aforementioned processor.

[0246] The processors mentioned above can be general-purpose processors, including central processing units (CPUs), network processors (NPs), etc.; they can also be digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.

[0247] In another embodiment of the present invention, a computer-readable storage medium is also provided, wherein a computer program is stored therein, and when the computer program is executed by a processor, it implements any of the resource determination methods described in the above embodiments.

[0248] In another embodiment of the present invention, a computer program product containing instructions is also provided, which, when run on a computer, causes the computer to perform any of the resource determination methods described in the above embodiments.

[0249] In the above embodiments, implementation can be achieved entirely or partially through software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented entirely or partially in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of the present invention are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that integrates one or more available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., solid state disk (SSD)).

[0250] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0251] The various embodiments in this specification are described in a related manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the embodiments of apparatus, electronic devices, computer-readable storage media, and computer program products are basically similar to the method embodiments, and therefore the descriptions are relatively simple; relevant parts can be referred to the descriptions of the method embodiments.

[0252] The above description is merely a preferred embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention are included within the scope of protection of the present invention.

Claims

1. A method for determining resources, characterized in that, The method includes: Acquire multiple preset network resources; the multiple network resources include: a first network resource sent to a client logged in with a first user account; For each network resource, the matching degree between the network resource and each user account is determined based on the type tag of the network resource; the type tag of the network resource is used to indicate the content included in the network resource. For each first network resource, based on the matching degree between the first network resource and each user account, the user account corresponding to the first network resource is determined from each user account and used as the first user account. For each first user account, based on the matching degree between the first user account and the second network resources other than the first network resources, the third network resource corresponding to the first user account is determined from each second network resource; The first network resource corresponding to the first user account and the third network resource corresponding to the first user account are determined as the target network resource corresponding to the first user account.

2. The method according to claim 1, characterized in that, For each network resource, determining the matching degree between the network resource and each user account based on the type tag of the network resource includes: For each user account, obtain the user's operation information on the historical network resources browsed by the user account to obtain the first operation information; Based on the first operation information and the type tags of the historical network resources that have been viewed, determine the matching degree between each type tag of the historical network resources that have been viewed and the user account. For each network resource, determine the type tag that the network resource has from the various type tags of the historical network resources that have been browsed; The matching degree between the network resource and the user account is calculated based on the matching degree between the type tag of the network resource and the user account.

3. The method according to claim 2, characterized in that, The first operation information includes: the number of operations performed by the user to which the user account belongs on historical network resources that have been viewed; the operation includes at least one of the following: click operation, forward operation, comment operation; The step of determining the matching degree between the various type tags of the viewed historical network resources and the user account based on the first operation information and the type tags of the viewed historical network resources includes: For each type of tag in the historical web resources that have been viewed, count the number of operations performed by the user account on the historical web resources with that type of tag, and use that count as the number of operations performed by the user account on that type of tag. Generate a target feature vector containing the number of operations performed by the user account for each type of tag; The target feature vector is input into a pre-trained probabilistic prediction model to obtain the matching degree between the various types of tags of the browsed historical network resources and the user account.

4. The method according to claim 1, characterized in that, For each first network resource, determining the user account corresponding to the first network resource from among the user accounts based on the matching degree between the first network resource and each user account, and using it as the first user account, includes: For each first network resource, the first number of user accounts corresponding to the first network resource are determined from the user accounts in descending order of the matching degree between the first network resource and each user account, and these are taken as the first user accounts. or, The user account whose matching degree with the first network resource is greater than the first threshold is determined from the user accounts and is designated as the first user account.

5. The method according to claim 4, characterized in that, For each first user account, based on the matching degree between the first user account and a second network resource (excluding the first network resource) among the plurality of network resources, the third network resource corresponding to the first user account is determined from each second network resource, including: For each second network resource other than the first network resource among the plurality of network resources, the first user account corresponding to the second network resource is determined from the first user accounts according to the matching degree between each first user account and the second network resource; wherein, one second network resource corresponding to one first user account means: sending the second network resource to the client logged in with the first user account; For each first user account, the first second number of second network resources corresponding to the first user account are determined from the second network resources in descending order of the matching degree between the first user account and each second network resource, and these are used as candidate network resources. If the candidate network resource is a second network resource corresponding to the first user account, then the candidate network resource is determined to be a third network resource corresponding to the first user account.

6. The method according to claim 5, characterized in that, For each first network resource, the first number corresponding to the first network resource decreases as the first time interval increases; the first threshold corresponding to the first network resource increases as the first time interval increases; the first time interval is: the time interval between the publication time of the first network resource and the current time. For each second network resource, the number of first user accounts corresponding to the second network resource decreases as the second time interval increases; the second time interval is the time interval between the publication time of the second network resource and the current time.

7. The method according to claim 6, characterized in that, The magnitude by which the number corresponding to the first network resource decreases as the first time interval increases is less than the magnitude by which the number of the first user accounts corresponding to the second network resource decreases as the second time interval increases.

8. The method according to claim 1, characterized in that, Before acquiring the preset multiple network resources, the method further includes: Receive a request to acquire a target network resource carrying the identifier of the first user account; After determining the first network resource corresponding to the first user account and the third network resource corresponding to the first user account as the target network resource corresponding to the first user account, the method further includes: Send the target network resources corresponding to the first user account to the client logged in with the first user account.

9. The method according to claim 1, characterized in that, After determining the user account corresponding to the first network resource from among the user accounts based on the matching degree between the first network resource and each user account for each first network resource, and using it as the first user account, the method further includes: For each second user account other than the first user account, based on the matching degree between the second user account and the second network resources other than the first network resources, the fourth network resource corresponding to the second user account is determined from the second network resources. Send the fourth network resource corresponding to the second user account to the client logged in with the second user account.

10. A resource determination device, characterized in that, The device includes: The acquisition module is used to acquire multiple preset network resources; the multiple network resources include: a first network resource sent to a client logged in with a first user account; The matching degree determination module is used to determine the matching degree between each network resource and each user account based on the type label of the network resource; the type label of the network resource is used to indicate the content included in the network resource. The first user account determination module is used to determine the user account corresponding to the first network resource from among the user accounts for each first network resource based on the matching degree between the first network resource and each user account, and use it as the first user account. The first resource determination module is used to determine, for each first user account, the third network resource corresponding to the first user account from each of the second network resources based on the matching degree between the first user account and the second network resources other than the first network resources among the plurality of network resources; The second resource determination module is used to determine the first network resource corresponding to the first user account and the third network resource corresponding to the first user account, as the target network resource corresponding to the first user account.

11. An electronic device, characterized in that, It includes a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other through the communication bus; Memory, used to store computer programs; A processor, when executing a program stored in memory, implements the steps of the method described in any one of claims 1-9.

12. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the steps of the method described in any one of claims 1-9.

Images