CN117668357A

CN117668357A - Resource recommendation method and device

Info

Publication number: CN117668357A
Application number: CN202311605529.1A
Authority: CN
Inventors: 郎鹏雪; 张晓敏
Original assignee: Beijing Baidu Netcom Science and Technology Co Ltd
Current assignee: Beijing Baidu Netcom Science and Technology Co Ltd
Priority date: 2023-11-28
Filing date: 2023-11-28
Publication date: 2024-03-08

Abstract

The disclosure provides a resource recommendation method and device, relates to the field of data processing, and particularly relates to the field of information flow. The specific implementation scheme is as follows: determining resources to be recommended to a target user according to a diversity threshold corresponding to a historical resource type of the target user, wherein the historical resource type is: the resource type of the historical resource recommended to the target user is determined according to the display statistical information and the access statistical information, wherein the display statistical information is as follows: displaying statistical information of the historical resources to the target user, wherein the access statistical information is: the statistical information of the historical resources accessed by the target user; recommending the determined resources to a client used by the target user. By applying the resource recommendation scheme provided by the embodiment of the disclosure, the diversity threshold can be set for the individuation of the user, so that the resource is recommended for the individuation of the user.

Description

Resource recommendation method and device

Technical Field

The disclosure relates to the technical field of data processing, in particular to the technical field of information flow, and particularly relates to a resource recommendation method and device.

Background

In the process that a user accesses resources such as images, videos and audios by using the client, the client sends a resource acquisition request to the server, the server responds to the resource acquisition request, determines resources to be recommended to the user from candidate resources, and recommends the determined resources to the client, so that the client can display the determined resources to the user.

In order to recommend rich and colorful resources to a user, the server may set the same diversity threshold corresponding to each resource type, where the diversity threshold corresponding to one resource type refers to the maximum duty ratio of the resource type in the resources to be recommended to the user. When determining the resources to be recommended to the user, the server can determine the resources of each resource type to be recommended to the user according to the diversity threshold corresponding to each resource type.

Disclosure of Invention

The disclosure provides a resource recommendation method and device.

In a first aspect, an embodiment of the present disclosure provides a resource recommendation method, including:

determining resources to be recommended to a target user according to a diversity threshold corresponding to a historical resource type of the target user, wherein the historical resource type is: the resource type of the historical resource recommended to the target user is determined according to the display statistical information and the access statistical information, wherein the display statistical information is as follows: displaying statistical information of the historical resources to the target user, wherein the access statistical information is: the statistical information of the historical resources accessed by the target user;

Recommending the determined resources to a client used by the target user.

In a second aspect, an embodiment of the present disclosure provides a resource recommendation apparatus, including:

the resource determining module is used for determining resources to be recommended to the target user according to a diversity threshold corresponding to the historical resource type of the target user, wherein the historical resource type is as follows: the resource type of the historical resource recommended to the target user is determined according to the display statistical information and the access statistical information, wherein the display statistical information is as follows: displaying statistical information of the historical resources to the target user, wherein the access statistical information is: the statistical information of the historical resources accessed by the target user;

and the resource recommending module is used for recommending the determined resources to the client used by the target user.

In a third aspect, an embodiment of the present disclosure provides an electronic device, including:

at least one processor; and

a memory communicatively coupled to the at least one processor; wherein,

the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the method of the first aspect described above.

In a fourth aspect, embodiments of the present disclosure provide a non-transitory computer-readable storage medium storing computer instructions for causing a computer to perform the method of the first aspect.

In a fifth aspect, embodiments of the present disclosure provide a computer program product comprising a computer program which, when executed by a processor, implements the method of the first aspect described above.

From the above, when the scheme provided by the embodiment of the present disclosure is applied to recommending resources, the diversity threshold corresponding to the type of the historical resources is determined according to the display statistics information and the access statistics information, where the display statistics information is the statistics information for displaying the information of the historical resources to the target user, and the access statistics information is the statistics information for the target user to access the historical resources, so that the display statistics information and the access statistics information can reflect the preference habit of the user for accessing the resources, and thus the diversity threshold of the type of the historical resources can be set for the target user according to the two statistics information, and the recommended resources can be personalized for the target user according to the diversity threshold.

It should be understood that the description in this section is not intended to identify key or critical features of the embodiments of the disclosure, nor is it intended to be used to limit the scope of the disclosure. Other features of the present disclosure will become apparent from the following specification.

Drawings

The drawings are for a better understanding of the present solution and are not to be construed as limiting the present disclosure. Wherein:

fig. 1 is a flowchart of a first resource recommendation method according to an embodiment of the present disclosure;

fig. 2 is a flow chart of a threshold determining method according to an embodiment of the disclosure;

fig. 3 is a flowchart of a second resource recommendation method according to an embodiment of the present disclosure;

fig. 4 is a schematic structural diagram of a resource recommendation device according to an embodiment of the present disclosure;

fig. 5 is a schematic structural diagram of a threshold determining device according to an embodiment of the present disclosure;

fig. 6 is a block diagram of an electronic device for implementing a resource recommendation method of an embodiment of the present disclosure.

Detailed Description

Exemplary embodiments of the present disclosure are described below in conjunction with the accompanying drawings, which include various details of the embodiments of the present disclosure to facilitate understanding, and should be considered as merely exemplary. Accordingly, one of ordinary skill in the art will recognize that various changes and modifications of the embodiments described herein can be made without departing from the scope and spirit of the present disclosure. Also, descriptions of well-known functions and constructions are omitted in the following description for clarity and conciseness.

First, an application scenario of the solution provided by the embodiment of the present disclosure is described.

1. Refreshing accessed resources by a user during access to the resources

When the user accesses the resource displayed in the client, the user can perform a resource refreshing operation, such as clicking a refresh button, sliding down a page, and the like. After detecting the refresh operation of the user, the client may send a resource acquisition request to the server. The server recommends new resources to the client in response to the resource acquisition request so that the client can present the new resources to the user, that is, the server recommends resources to the user.

2. User login or online

After a user logs in or goes on the client, the client can send a resource acquisition request to the server, so that the server recommends resources to the client, and the client displays the resources to the user.

3. User opens resource recommendation page provided by client

The client can provide various pages including a resource recommendation page, the resource recommendation page is provided with information of resources, the resource recommendation page provided by the client can be opened in the process of using the client by a user, when the user opens the resource recommendation page, the client can send a resource acquisition request to the server, so that the server recommends resources to the client, and the client displays the resources to the user in the resource recommendation page.

The three application scenarios are only part of the application scenarios of the resource recommendation scheme provided by the embodiments of the present disclosure, and the embodiments of the present disclosure are not limited to other application scenarios.

Next, an execution subject of the resource recommendation scheme provided by the embodiment of the present disclosure will be described.

The execution subject of the resource recommendation scheme provided by the embodiment of the disclosure is a server. And the server determines the resources to be recommended to the user according to the diversity threshold corresponding to the resource type, and recommends the determined resources to the client used by the user.

The execution subject for determining the diversity threshold corresponding to the resource type may be a server or may be another device other than the server having the diversity threshold determination function.

Under the condition that the other equipment determines the diversity threshold corresponding to the resource type, the other equipment can determine the diversity threshold corresponding to the historical resource type according to the statistical information aiming at the historical resource before the server responds to the resource acquisition request of the client, and sends the determined diversity threshold to the server.

In the case that the server determines the diversity threshold corresponding to the resource type, the server may determine the diversity threshold corresponding to the historical resource type before responding to the resource acquisition request, and may determine the diversity threshold corresponding to the historical resource type first, then determine the resource to be recommended to the user according to the diversity threshold corresponding to the historical resource type, and recommend the determined resource to the server.

The resource recommendation scheme is described in detail below through specific embodiments.

In an embodiment of the present disclosure, referring to fig. 1, a flowchart of a first resource recommendation method is provided, where the method includes the following steps S101 to S102.

Step S101: and determining the resources to be recommended to the target user according to the diversity threshold corresponding to the historical resource types of the target user.

Wherein the resources may be images, video, audio, etc. Taking video resources as an example, there are a plurality of resource types of resources, such as sports, finance, movies, etc.

The above-mentioned historical resource types are: resource type of historical resources recommended to the target user.

The history resource is a resource recommended to the client by the server in the past, and the history resource may be a resource recommended to the client by the server in a past period of time. For example, the historical resources may be resources recommended by the server to the client during the past week.

The diversity threshold corresponding to the resource type can be regarded as: among the resources recommended to the target user, the larger the diversity threshold is, the larger the ratio of the resources of the resource type in the resources recommended to the target user is, and the smaller the diversity threshold is, the smaller the ratio of the resources of the resource type in the resources recommended to the target user is.

The diversity threshold is determined according to the display statistical information and the access statistical information, wherein the display statistical information is as follows: the statistical information of the historical resources is displayed to the target user, and the access statistical information is as follows: and the target user accesses the historical resources.

For a description of exposing statistics and accessing statistics, a specific implementation of determining a diversity threshold may be found in the subsequent embodiments, which are not described in detail herein.

Specifically, the server may determine the total number of resources to be recommended to the client, so that according to the diversity threshold corresponding to the historical resource type, the server may multiply the total number by the diversity threshold corresponding to the historical resource type, and the obtained product result is the maximum number of resources of the historical resource type to be recommended to the client, and then determine, from the candidate resources of the historical resource type, resources of which the number does not exceed the maximum number, as the resources to be recommended to the target user.

The total number can be a preset, fixed and invariable number, and the server recommends a fixed number of resources to the client each time; the total number may also be the number of resources carried by the client in the resource acquisition request, and when the server responds to the resource acquisition request, the number of resources carried in the resource acquisition request may be obtained and used as the total number.

Since the historical resources recommended to the target user may be one or more types of resources, there may be one or more of the above-mentioned types of historical resources. For each historical resource type, the server may determine the maximum number of resources for that historical resource type based on the diversity threshold corresponding to that historical resource type.

For example, if the server is to recommend 50 resources to the client, and the diversity threshold corresponding to the T1 type is 0.3, the server may calculate that the maximum number corresponding to the T1 type is 15, that is, the server may determine 15 resources from the candidate resources of the T1 type as the resources to be recommended to the client.

After determining the maximum number of resources of the history resource type, the optional number of resources not exceeding the maximum number can be selected from the candidate resources of the history resource type as the resources to be recommended to the target user, or the resources not exceeding the maximum number can be selected from the candidate resources of the history resource type according to a preset rule.

For example, the preset rule may be selected in order of the data amount of the resource from large to small, or may be selected in order of the resource names, or the like.

In addition, when selecting resources from the resources of the candidate history resource type, if the number of the resources of the candidate history resource type does not exceed the maximum number, all the resources of the candidate history resource type can be selected; if the number of the candidate resources of the historical resource type exceeds the maximum number, the maximum number of resources can be selected from the candidate resources, and the remaining unselected resources are used as the resource supplement set. And selecting resources from the candidate resources of each historical resource type according to the selection mode, and selecting resource complement from the resource complement set if the total number of the selected resources is insufficient to be recommended to the client.

Step S102: recommending the determined resources to a client used by the target user.

Specifically, the server may recommend resources to the client through either of the following two implementations.

In a first implementation manner, after determining the resource, the server may obtain the address information of the determined resource and send the obtained address information of the resource to the client, so that after receiving the address information sent by the server, the client may determine the resource at the location indicated by the address information and obtain the preview information of the determined resource, thereby displaying the obtained preview information of the resource to the target user in the page.

In a second implementation manner, after determining the resource, the server may obtain the preview information of the determined resource, and send the obtained preview information of the resource to the client, so that after receiving the preview information of the resource, the client directly displays the obtained preview information of the resource to the target user in the page.

In addition, after determining the resources, the server may sort the determined resources and recommend the sorted resources to the client.

In addition, the diversity threshold corresponding to one historical resource type refers to the maximum ratio of the resources of the historical resource type in the resources to be recommended to the user, after the diversity threshold corresponding to each historical resource type is determined, the ratio of the resources of each resource type in the resources to be recommended to the user can be allocated under the condition of conforming to the preference habit of the user according to the diversity threshold corresponding to each historical resource type, so that the resources with rich and colorful colors are recommended to the user, aesthetic fatigue and freshness of target users are not reduced, and the resource recommendation scheme provided by the embodiment of the disclosure can be used for individually recommending the resources to the user, and can recommend different resources to different users, so that thousands of faces of resource recommendation is realized, and the recommendation effect is optimized.

In one embodiment of the present disclosure, a plurality of resource aggregation modules may be included in a server, each for recalling resources using a different recall policy.

After receiving the resource acquisition request, the server can call each resource aggregation module to recall the resource and take the recalled resource as the candidate resource, so that the resource to be recommended to the target user is determined in the candidate resource according to the diversity threshold corresponding to the historical resource type.

In addition, after the resource aggregation module is called to recall the resource, the recalled resource can be subjected to processing such as de-duplication, fusion and filtering, and the processed resource is used as the candidate resource.

The presentation statistics and access statistics mentioned in step S101 above are first described below.

a. Displaying statistical information

The display statistical information is as follows: the statistical information of the information showing the history resource to the target user may be regarded as information obtained by counting the behavior of the information showing the history resource to the target user.

The information of the resource may be content, introduction, cover, etc. of the resource. And displaying the information of the history resources to the target user, namely displaying the content, the brief introduction, the cover and the like of the history resources to the target user.

For example, in the case where the resource is an image, the image may be directly presented to the target user; in the case that the resource is video, a video cover, a video brief introduction and the like can be displayed to the target user, and the video or the video fragment and the like can also be directly displayed to the target user.

The display statistics may include a number of resource displays, a rate of resource displays, and the like.

The number of the resource displays is as follows: the amount of information of the historical resources presented to the target user.

The historical resources may have multiple types, and the number of the displayed resources may be divided into a total number of displayed resources and a number of displayed resources corresponding to each resource type, where the total number of displayed resources is: the number of information of all historical resources displayed to the target user, and the corresponding display number of each resource type is as follows: the amount of information of the historical resources of the resource type presented to the target user.

In addition, each resource type corresponds to a resource display rate, and the resource display rate corresponding to one resource type is as follows: the ratio of the number of presentations corresponding to the resource type to the total number of presentations of resources.

b. Accessing statistics

The access statistics are: the statistical information of the target user accessing the history resource can be regarded as information obtained by counting the behavior of the target user accessing the history resource.

The access statistics may include the number of resources accessed and the rate at which the resources are accessed.

The number of accessed resources is: the number of historical resources accessed by the target user.

The number of accessed resources may also be divided into a total number of accessed resources and a number of accessed resources corresponding to each resource type, where the total number of accessed resources is: the number of all historical resources accessed by the target user, and the number of accessed resources corresponding to each resource type is as follows: the number of the resource types accessed by the target user corresponds to.

In addition, each resource type corresponds to a resource accessed rate, and the resource accessed rate corresponding to one resource type is: the ratio of resources of the resource type accessed by the target user in the historical resources.

Next, a specific implementation of determining the diversity threshold mentioned in the above step S101 will be described.

The diversity threshold may be determined by a server or by a device other than the server having a diversity threshold determination function. The following description will take an example of the server determining the diversity threshold.

In an embodiment of the present disclosure, referring to fig. 2, a flow chart of a threshold determining manner is provided, and in this embodiment, a diversity threshold corresponding to a historical resource type of a target user may be determined through the following steps S201 to S204.

Step S201: resource information of historical resources recommended to the target user is obtained.

The resource information of the history resource may include attributes of the history resource, such as a resource name, a resource type, and the like.

Specifically, each time the server recommends a resource to the client, the resource information or index information of the recommended resource may be recorded. Under the condition that the server records the resource information of the resource, the resource information recorded by the server can be directly read; in the case where the server has index information of resources recorded, it is possible to determine history resources from the index information and obtain resource information of the determined history resources.

Step S202: and determining the historical resource type of the historical resource according to the obtained resource information.

For example, the resource information obtained by the server may include a resource type of the resource, so that the resource type included in the resource information may be directly read, and the read resource type is a historical resource type.

For another example, the server may obtain a correspondence between the resource type and the keyword in advance, for example, for a sports such as a resource type, there may be a correspondence between the resource type and the keyword such as football, basketball, sprint, swimming, and the like. In this case, the resource information obtained by the server may include the name of the resource, so that the server may perform keyword detection on the resource name to obtain the target keyword included in the resource name, and then detect the corresponding relationship including the target keyword in the corresponding relationship obtained in advance, thereby determining the resource type in the corresponding relationship, as the historical resource type of the historical resource.

Step S203: and acquiring display statistical information and access statistical information.

The server may record the total number of recommended resources and the number of resources of each resource type in the recommended resources each time when recommending resources to the client, so that when acquiring the display statistics information, the server may use the recorded total number and the number of resources of each resource type as the display statistics information, and the server may calculate the resource display rate corresponding to each resource type according to the total number and the number of resources of each resource type, and also use the recorded total number and the number of resources of each resource type as the display statistics information.

In addition, each time the target user accesses the resource, the client can send a notification of the target user accessing the resource to the server, so that the server can record the total number of the resources accessed by the target user, the notification sent by the client can carry the resource type of the resources accessed by the target user, so that the server can record the number of the accessed resources corresponding to each resource type, and therefore, when the access statistical information is obtained, the server can take the recorded total number of the resources accessed and the number of the accessed resources corresponding to each resource type as the access statistical information. The server may also calculate the resource access rate corresponding to each resource type according to the total number of accessed resources and the number of accessed resources corresponding to each resource type, and also use the resource access rate as the access statistics information.

Step S204: and determining a diversity threshold corresponding to the historical resource type of the target user according to the display statistical information and the access statistical information.

The manner of determining the diversity threshold in step S204 is different if the presentation statistics and the access statistics obtained in step S203 are different, and in particular, reference may be made to the following embodiments, which will not be described in detail herein.

In addition, when classifying the classification type of the resource, the multi-layer classification type to which the resource belongs may be classified, so that for the history resource, the history resource has the history classification type to which the history resource belongs at each level. For this case, when determining the diversity threshold corresponding to the history resource type, the diversity threshold corresponding to the history classification type to which the history resource belongs at each hierarchy may be determined.

For example, if the history resource belongs to a ball type under the sports type, the history resource type is two types of the sports type and the ball type, so that when the diversity threshold corresponding to the history resource type is determined, the diversity threshold corresponding to the sports type can be determined according to the display statistical information and the access statistical information corresponding to the sports type; and determining a diversity threshold corresponding to the ball type according to the display statistical information and the access statistical information corresponding to the ball type.

From the above, when the resource is recommended by the scheme provided by the embodiment of the present disclosure, the server can accurately determine the historical resource type of the historical resource according to the resource information of the historical resource recommended to the target user, so that after acquiring the display statistical information and the access statistical information, the server can determine the diversity threshold conforming to the preference habit of the target user according to the display statistical information and the access statistical information reflecting the preference habit of the user accessing the resource, thereby recommending the resource by using the determined diversity threshold and personalizing the recommended resource for the target user.

In one embodiment of the present disclosure, referring to fig. 3, a determination flow of a server determining resources to be recommended to a target user is shown. The determination flow shown in fig. 3 includes two stages, one is a diversity threshold calculation stage and the other is a diversity screening stage.

In the diversity threshold calculation stage, a server or other equipment with a diversity threshold determination function obtains resource information of historical resources recommended to a target user, determines the type of the historical resources according to the obtained resource information, obtains presentation statistical information and access statistical information, and finally determines a diversity threshold corresponding to the type of the historical resources according to the presentation statistical information and the access statistical information.

In the diversity screening stage, the server calls each resource aggregation module to obtain candidate resources, calculates the corresponding quantity upper limit (namely the maximum quantity) of each historical resource type according to the diversity threshold value obtained in the diversity threshold value calculation stage, and selects the quantity upper limit resources at most in the resources of the historical resource type contained in the candidate resources according to each historical resource type, so that the candidate resources meet a diversity candidate set, and the unselected resources enter a supplementary set.

After the above processing is performed for each historical resource type, if the number of resources in the satisfied diversity candidate set is not enough to the total number of resources to be recommended to the client, selecting resources from the complement set, and adding the selected resources to the satisfied diversity candidate set to complement the missing part of the satisfied diversity candidate set, thereby obtaining the next-stage candidate set.

After obtaining the candidate set of the next stage, the server can directly recommend the resources in the candidate set of the next stage to the client; the resources in the candidate set of the next stage can also be ranked, and the ranked resources are recommended to the client.

The following describes the above-described step S203 and step S204.

In one embodiment of the present disclosure, the number of resource impressions and the rate of resource accesses may be obtained when the impressions statistics and the access statistics are obtained.

The implementation manner of obtaining the number of resource presentations and the resource access rate can be referred to the foregoing embodiments, and will not be described herein.

In the scheme, the number of the resource display reflects the number of the resource displayed to the target user, the resource accessed rate reflects the preference of the target user for accessing the resource, the higher the resource accessed rate corresponding to one resource type is, the more interested the target user is in the resource of the resource type corresponding to the resource accessed rate is, the lower the resource accessed rate corresponding to one resource type is, the less interested the target user is in the resource of the resource type corresponding to the resource accessed rate is, therefore, according to the number of the resource display and the resource accessed rate, the individualized diversity threshold meeting the preference habit of the target user can be determined, and the resources can be recommended for the target user in a personalized mode by utilizing the determined diversity threshold.

After the resource display quantity and the resource accessed rate are obtained, according to the two kinds of information, a diversity threshold corresponding to the historical resource type can be determined according to any one of the following two implementation modes.

In a first implementation manner, according to the number of the resource display, a first parameter corresponding to the historical resource type is determined, and according to the determined first parameter and the resource accessed rate, a diversity threshold corresponding to the historical resource type of the target user is determined.

For each historical resource type, the presentation of the resources of that historical resource type affects whether the target user is interested in the resources of that historical resource type.

For example, if the number of resource presentations for a resource of a historical resource type is small, the target user can browse less resources of the historical resource type, which may be difficult to arouse the user's interest in the resources of the historical resource type. In other words, the target user is not interested in resources of one historical resource type, possibly not due to the target user's own cause, but rather due to the small number of resources of that historical resource type recommended to the target user.

The first parameter characterizes the influence degree of the display condition of the resources of the history resource type on the recommended resources of the history resource type.

If the display of the resources of a history resource type is smaller, the user is more likely to have less interest in the resources of the history resource type, and at this time, more recommends of the resources of the history resource type are needed, that is, the higher the influence degree on recommends of the resources of the history resource type is, the higher the first parameter is.

In one embodiment of the present disclosure, a first parameter corresponding to a historical resource type is calculated according to the following expression:

wherein the index_show_ratio represents a first parameter corresponding to the historical resource type, the total_show represents the total resource display quantity of the resources of all the historical resource types, the show represents the display quantity of the resources of the historical resource type, and f ₁ () Representing a square root taking operation.

In the above expression, the smaller the number of the exhibition of the resources of the history resource typeThe larger the size of the container,the larger the square root of the first parameter, i.e., the larger the first parameter, it can be seen that according to the above expression, the first parameter can be accurately calculated, so that the resource is recommended according to the first parameter, and the accuracy of resource recommendation can be improved.

In another embodiment of the present disclosure, when calculating the first parameter, the total resource display number may be divided by the display number of resources of the historical resource type, and the obtained quotient is multiplied by a preset coefficient to obtain a multiplication calculation result as the first parameter.

The above-mentioned resource access rate characterizes the interest degree of the target user in the resources of the historical resource types due to the interest of the target user, and the higher the resource access rate corresponding to one historical resource type, the more interested the user is in the resources of the historical resource types, so that the server needs to recommend the resources of the historical resource types to the target user.

In one embodiment of the present disclosure, the diversity threshold corresponding to the historical resource type of the target user is calculated according to the following expression:

ratio＝α*click_ratio+β*inver_show_ratio

the ratio represents a diversity threshold corresponding to the historical resource type of the target user, the click_ratio represents the resource access rate of the resource of the historical resource type, alpha represents a first preset weight, and beta represents a second preset weight.

The first preset weight and the second preset weight are weight coefficients determined in advance. The two preset weights can be two weights which are mutually independent, namely the two preset weights are weight coefficients which are respectively and independently set; the two preset weights may also be two weights associated with each other.

For example, in the case where the above two preset weights are associated with each other, α and β are determined according to the following expression:

α+β＝1

It can be seen from the expression that if α is larger, β is smaller, so that interest of the target user is emphasized when calculating the diversity threshold; if alpha is smaller, beta is larger, and thus, when the diversity threshold is calculated, the display condition of the resources of the historical resource types is emphasized, therefore, by applying the expression, the two preset weights can be set according to the aspect of the emphasis of actual needs, and the diversity threshold meeting the actual needs can be determined, so that the resources are recommended to the target user according to the diversity threshold, and the accuracy of resource recommendation can be improved.

The higher the first parameter is, the more resources of the recommended historical resource types are needed, namely the higher the diversity threshold corresponding to the historical resource types is needed; and the higher the resource access rate is, the more interested the user is in the resources of the historical resource type, so that the server also needs to recommend the resources of the historical resource type to the target user, namely the higher the diversity threshold corresponding to the historical resource type is also needed. In the above expression for calculating the diversity threshold, the higher the click_ratio and/or the reverse_show_ratio, the higher the ratio, so that the diversity threshold corresponding to the historical resource type can be accurately calculated by using the above expression for calculating the diversity threshold, and resources can be recommended to the target user according to the calculated diversity threshold, and the accuracy of resource recommendation can be improved.

From the above, in the scheme, the diversity threshold is determined by combining the first parameter and the accessed rate of the resources, which is equivalent to determining the diversity threshold by considering both the resource display condition of the historical resource type and the interest and hobbies of the target user, so that the accuracy of determining the diversity threshold can be improved, resources can be recommended to the target user according to the diversity threshold, and the accuracy of recommending the resources can be improved.

In the second implementation manner, weights can be directly allocated to the display quantity of the resources of the historical resource types and the accessed rate of the resources, and the weighted summation is performed to obtain a calculation result, and the calculation result is used as the diversity threshold.

In a first implementation manner of determining the diversity threshold according to the number of resource display and the accessed rate of the resource, after calculating the first parameter, the diversity threshold may also be determined through the following embodiments.

In one embodiment of the disclosure, an initial diversity threshold corresponding to a historical resource type is determined according to the determined first parameter and the resource accessed rate; if the initial diversity threshold is smaller than or equal to a preset first threshold, determining the first threshold as a diversity threshold corresponding to the historical resource type of the target user; if the initial diversity threshold is larger than the first threshold and smaller than a preset second threshold, determining the initial diversity threshold as a diversity threshold corresponding to the historical resource type of the target user; and if the initial diversity threshold is greater than or equal to the second threshold, determining the second threshold as the diversity threshold corresponding to the historical resource type of the target user.

The first threshold and the second threshold are preset thresholds, and the first threshold is smaller than the second threshold.

The specific implementation of determining the initial diversity threshold is similar to the specific implementation of determining the diversity threshold in the foregoing embodiment, and the difference is only that the threshold determined in the foregoing embodiment is referred to as the diversity threshold, and the determined threshold is referred to as the initial diversity threshold and is not repeated herein.

After the initial diversity threshold is calculated, the initial diversity threshold, the first threshold and the second threshold can be compared, if the initial diversity threshold is smaller than or equal to the first threshold, the initial diversity threshold is too small, and the number of resources of the historical resource types recommended to the user is small; if the initial diversity threshold is greater than or equal to the second threshold, it is indicated that the initial diversity threshold is too large, which results in a large number of resources of the historical resource type recommended to the user, so that the user is tired aesthetically and the freshness is reduced.

When the display statistics information and the access statistics information are obtained in the step S203, the resource display rate may be obtained in addition to the resource display number and the resource accessed rate, where the resource display rate is: the ratio of resources of the historical resource type that are presented to the target user in the historical resources.

The method for obtaining the resource presentation rate may be referred to the description in step S203, and will not be described here again.

After the resource display number, the resource display rate and the resource accessed rate are obtained, a diversity threshold corresponding to the historical resource type can be determined according to the following steps one to three.

Step one: and predicting the access probability of the target user to the resources of the historical resource type according to the resource presentation rate and the resource accessed rate.

The higher the resource display rate, the higher the probability that the target user repeatedly browses the resources of the same historical resource type, so that the more easily the user is subjected to aesthetic fatigue, and the target user is unwilling to access the resources of the historical resource type; the higher the resource is accessed, the higher the target user's interest in resources of the historical resource type, and thus the more willing the target user to access resources of the historical resource type.

The access probability can be predicted according to any one of the following two implementations according to the resource presentation rate and the resource accessed rate.

In a first implementation, first, determining a second parameter representing the interest degree of a target user on the resources of the historical resource type according to the resource display rate and the resource accessed rate; and then predicting the access probability of the target user to access the resources of the historical resource type according to the second parameter and the resource display rate.

Specifically, the higher the resource display rate, the easier the target user is to produce aesthetic fatigue on the resources of the historical resource type, so that the target user is not interested in the resources of the historical resource type; the higher the resource accessed rate is, the more interested the target user is in the resources of the historical resource types, so that the second parameter can be accurately determined from two aspects according to the resource display rate and the resource accessed rate, the access probability can be accurately predicted according to the second parameter and the resource display rate, and the accuracy of resource recommendation can be improved by utilizing the access probability.

In view of this, the second parameter may be determined according to the following expression:

reward＝click_ratio-f ₂ (show_ratio)

wherein click_ratio represents the resource accessed rate of the resources of the historical resource type, show_ratio represents the resource display rate of the resources of the historical resource type, and rewind represents the second parameter corresponding to the historical resource type ，f ₂ () Representing the operation of taking an index of a preset base.

The preset base number is a preset value, for example, 1.3, 1.4, etc.

In the above expression for calculating the second parameter, the greater the accessed rate of the resource, the higher the interested degree of the target user on the resource of the historical resource type, and the greater the second parameter; the larger the resource display rate is, the easier the target user is to produce aesthetic fatigue on the resources of the historical resource types, the smaller the second parameter is, and therefore, the second parameter can be accurately calculated by applying the expression for calculating the second parameter, the diversity threshold can be accurately determined according to the second parameter, and the accuracy of resource recommendation can be improved according to the diversity threshold.

According to the second parameter and the resource exposure rate, the access probability can be predicted according to the following expression:

wherein x represents the access probability corresponding to the historical resource type, reward represents the second parameter corresponding to the historical resource type, and show_ratio represents the resource display rate of the resources of the historical resource type.

In the expression for calculating the access probability, the greater the second parameter, the higher the interest degree of the target user in the resources of the historical resource type is, and the greater the access probability of the target user in accessing the resources of the historical resource type is; the larger the resource display rate is, the easier the target user is to produce aesthetic fatigue on the resources of the historical resource types, so that the smaller the access probability of the target user to the resources of the historical resource types is, and the more the access probability can be calculated accurately by applying the expression for calculating the access probability, so that the diversity threshold can be determined accurately, and the accuracy of resource recommendation can be improved according to the diversity threshold.

In a second implementation manner, the division calculation result may be obtained by dividing the resource accessed rate by the resource presentation rate, and the division calculation result may be used as the access probability.

Step two: and determining the target resource quantity of the resources of the historical resource type to be recommended to the target user according to the access probability and the resource display quantity.

After the access probability is calculated, the number of the historical resource types of the resources accessed by the target user after recommending the resources to the target user can be predicted according to the access probability and the resource display number.

For example, the target resource quantity of resources of the history resource type to be recommended to the target user is determined according to the following expression:

w＝show*exp(γ*x)

wherein w represents the number of target resources corresponding to the historical resource type, show represents the number of resource display of the resources of the historical resource type, x represents the access probability corresponding to the historical resource type, and gamma represents a third preset weight.

In the expression for calculating the number of target resources, the higher the access probability corresponding to the historical resource type is, the more likely the target user accesses the resources of the historical resource type is, so that the larger the number of target resources is, the more the number of target resources can be accurately calculated according to the expression for calculating the number of target resources, the diversity threshold can be accurately determined, and the accuracy of resource recommendation can be improved according to the diversity threshold.

Step three: and calculating the quantity ratio of the target resource quantity corresponding to each historical resource type as a diversity threshold corresponding to each historical resource type.

After the number of target resources corresponding to each historical resource type is calculated, the sum of the number of target resources corresponding to all the historical resource types can be calculated, so that the duty ratio of the number of target resources corresponding to each historical resource type in the sum is calculated and used as the diversity threshold value corresponding to each historical resource type.

From the above, in the scheme, the diversity threshold is determined by using the plurality of information such as the resource display quantity, the resource display rate and the resource accessed rate, so that the accuracy of determining the diversity threshold can be improved, and the resource display rate and the resource accessed rate can respectively reflect the interest degree of the user on the resources of the historical resource types from two aspects, so that the prediction accuracy can be improved according to the resource display rate and the resource accessed rate to predict the access probability, and the accuracy of determining the diversity threshold can be improved by using the predicted access probability, so that the accuracy of recommending the resources can be improved according to the diversity threshold.

After the number of target resources corresponding to each of the historical resource types is calculated in the step two, the diversity threshold may also be determined by the following embodiment.

In one embodiment of the disclosure, for each historical resource type, a target duty ratio of a target number of resources corresponding to the historical resource type is calculated; if the target duty ratio is smaller than or equal to a preset first duty ratio, determining that the first duty ratio is a diversity threshold corresponding to the historical resource type; if the target duty ratio is larger than the first duty ratio and smaller than a preset second duty ratio, determining that the target duty ratio is a diversity threshold corresponding to the historical resource type; and if the target duty ratio is greater than or equal to the second duty ratio, determining the second duty ratio as a diversity threshold corresponding to the historical resource type.

The first duty ratio and the second duty ratio are preset numbers, and the first duty ratio is smaller than the second duty ratio.

The specific implementation of calculating the above target duty ratio is similar to the specific implementation of calculating the number duty ratio in the foregoing embodiment, and only the difference is that the duty ratio calculated in the foregoing embodiment is referred to as the number duty ratio, and the calculated duty ratio is referred to as the target duty ratio, which is not described herein.

After the target duty ratio is calculated, the target duty ratio, the first duty ratio and the second duty ratio can be compared, if the target duty ratio is smaller than or equal to the first duty ratio, the target duty ratio is determined to be a diversity threshold, and the diversity threshold is too small, so that the number of resources of the historical resource types recommended to the user is small; if the target duty ratio is greater than or equal to the second duty ratio, the target duty ratio is determined to be the diversity threshold value, the diversity threshold value is too large, so that the number of resources of the history resource type recommended to the user is large, and accordingly aesthetic fatigue and freshness of the user are reduced.

Different presentation statistics and access statistics may be utilized to determine diversity thresholds for different users.

In one embodiment of the present disclosure, before acquiring presentation statistics and access statistics, determining whether a target user is an active user; if the target user is an active user, acquiring the resource display quantity and the resource accessed rate; and if the target user is an inactive user, acquiring the resource display quantity, the resource display rate and the resource accessed rate.

Specifically, when judging whether the target user is an active user, information such as continuous login days, client use time, client use times, resource display quantity, resource access quantity and the like of the target user can be obtained, and whether the target user is an active user is judged according to the information.

If the value of the acquired information is larger, the target user can be determined to be an active user, otherwise, the target user can be determined to be an inactive user.

Taking continuous login days as an example, if the continuous login days of the target user exceed a preset day threshold, determining that the target user is an active user; and if the number of continuous login days of the target user does not exceed the number of days threshold, determining that the target user is an inactive user.

And under the condition that the target user is an active user, acquiring the resource display quantity and the resource accessed rate, and determining a diversity threshold according to the resource display quantity and the resource accessed rate.

The specific implementation manner of determining the diversity threshold according to the above-mentioned resource display number and the resource accessed rate may refer to the foregoing embodiments, and will not be described herein again.

And under the condition that the target user is an inactive user, acquiring the resource display quantity, the resource display rate and the resource accessed rate, and determining a diversity threshold according to the resource display quantity, the resource display rate and the resource accessed rate.

The specific implementation manner of determining the diversity threshold according to the above-mentioned resource display number, resource display rate and resource accessed rate can be referred to the foregoing embodiments, and will not be repeated here.

From the above, when the resource is recommended by the scheme provided by the embodiment of the disclosure, different information is acquired for different users, and the diversity threshold is determined in different modes, so that the accuracy of determining the diversity threshold can be improved, and the accuracy of recommending the resource can be improved.

Corresponding to the resource recommendation method, the embodiment of the disclosure also provides a resource recommendation device.

In an embodiment of the present disclosure, referring to fig. 4, a schematic structural diagram of a resource recommendation device is provided, where in this embodiment, the device includes:

the resource determining module 401 is configured to determine a resource to be recommended to a target user according to a diversity threshold corresponding to a historical resource type of the target user, where the historical resource type is: the resource type of the historical resource recommended to the target user is determined according to the display statistical information and the access statistical information, wherein the display statistical information is as follows: displaying statistical information of the historical resources to the target user, wherein the access statistical information is: the statistical information of the historical resources accessed by the target user;

and a resource recommending module 402, configured to recommend the determined resource to the client used by the target user.

In one embodiment of the present disclosure, referring to fig. 5, a schematic structural diagram of a threshold determining device is provided, in this embodiment, the following modules are used to determine a diversity threshold corresponding to the historical resource type of the target user:

a resource information obtaining module 501, configured to obtain resource information of the historical resources recommended to the target user;

A type determining module 502, configured to determine a historical resource type of the historical resource according to the obtained resource information;

a statistics information obtaining module 503, configured to obtain the presentation statistics information and the access statistics information;

the threshold determining module 504 is configured to determine a diversity threshold corresponding to the historical resource type of the target user according to the presentation statistics and the access statistics.

In one embodiment of the disclosure, the resource information obtaining module 501 includes:

the information acquisition sub-module is used for acquiring the resource display quantity and the resource accessed rate, wherein the resource display quantity is as follows: the amount of information of the historical resources presented to the target user, the resource accessed rate being: and the ratio of the resources accessed by the target user in the historical resources.

In one embodiment of the present disclosure, the threshold determining module 504 includes:

the parameter determination submodule is used for determining a first parameter corresponding to the historical resource type according to the resource display quantity, wherein the first parameter represents the influence degree of the display condition of the resources of the historical resource type on the recommended resources of the historical resource type;

and the threshold value determining submodule is used for determining a diversity threshold value corresponding to the historical resource type of the target user according to the determined first parameter and the resource accessed rate.

In one embodiment of the disclosure, the parameter determination submodule is specifically configured to:

calculating a first parameter corresponding to the historical resource type according to the following expression:

From the above, in the present solution, according to the above expression, the first parameter can be accurately calculated, so that the resource is recommended according to the first parameter, and the accuracy of resource recommendation can be improved.

In one embodiment of the disclosure, the threshold determination submodule is specifically configured to:

Calculating a diversity threshold corresponding to the historical resource type of the target user according to the following expression:

ratio＝α*click_ratio+β*inver_show_ratio

According to the scheme, the diversity threshold corresponding to the historical resource type can be accurately calculated by using the expression for calculating the diversity threshold, so that resources are recommended to a target user according to the calculated diversity threshold, and the accuracy of resource recommendation can be improved.

In one embodiment of the present disclosure, the α and the β are determined according to the following expression:

α+β＝1

in the scheme, the expression is applied, the two preset weights can be set according to the aspect of important consideration of actual needs, so that the diversity threshold meeting the actual needs can be determined, resources are recommended to a target user according to the diversity threshold, and the accuracy of resource recommendation can be improved.

determining an initial diversity threshold corresponding to the historical resource type according to the determined first parameter and the resource accessed rate;

If the initial diversity threshold is smaller than or equal to a preset first threshold, determining the first threshold as the diversity threshold corresponding to the historical resource type of the target user;

if the initial diversity threshold is larger than the first threshold and smaller than a preset second threshold, determining that the initial diversity threshold is a diversity threshold corresponding to the historical resource type of the target user, wherein the first threshold is smaller than the second threshold;

and if the initial diversity threshold is greater than or equal to the second threshold, determining that the second threshold is the diversity threshold corresponding to the historical resource type of the target user.

In the scheme, after the initial diversity threshold value is calculated, the initial diversity threshold value, the first threshold value and the second threshold value can be compared, if the initial diversity threshold value is smaller than or equal to the first threshold value, the initial diversity threshold value is too small, and the number of resources of the historical resource types recommended to the user is small; if the initial diversity threshold is greater than or equal to the second threshold, it is indicated that the initial diversity threshold is too large, which results in a large number of resources of the historical resource type recommended to the user, so that the user is tired aesthetically and the freshness is reduced.

In one embodiment of the disclosure, the information obtaining sub-module is specifically configured to:

obtaining the resource display quantity, the resource display rate and the resource accessed rate, wherein the resource display rate is as follows: a ratio of resources of the historical resource type presented to the target user in the historical resources;

the threshold determination module 504 includes:

the probability prediction sub-module is used for predicting the access probability of the target user to the resources of the historical resource type according to the resource display rate and the resource accessed rate;

a quantity determination submodule, configured to determine, according to the access probability and the resource display quantity, a target resource quantity of resources of the historical resource type to be recommended to the target user;

the threshold value calculating sub-module is used for calculating the quantity ratio of the target resource quantity corresponding to each historical resource type and taking the quantity ratio as the diversity threshold value corresponding to each historical resource type.

In one embodiment of the disclosure, the probability prediction submodule includes:

a parameter determining unit, configured to determine a second parameter that characterizes a degree of interest of the target user in the resource of the historical resource type according to the resource presentation rate and the resource accessed rate;

and the probability prediction unit is used for predicting the access probability of the target user to access the resources of the historical resource type according to the second parameter and the resource display rate.

In the scheme, the higher the resource display rate is, the easier the target user is to produce aesthetic fatigue on the resources of the historical resource types, so that the target user is not interested in the resources of the historical resource types; the higher the resource accessed rate is, the more interested the target user is in the resources of the historical resource types, so that the second parameter can be accurately determined from two aspects according to the resource display rate and the resource accessed rate, the access probability can be accurately predicted according to the second parameter and the resource display rate, and the accuracy of resource recommendation can be improved by utilizing the access probability.

In one embodiment of the disclosure, the parameter determining unit is specifically configured to:

The second parameter is determined according to the following expression:

reward＝click_ratio-f ₂ (show_ratio)

wherein click_ratio represents the resource accessed rate of the resources of the historical resource type, show_ratio represents the resource display rate of the resources of the historical resource type, and rewind represents the second parameter corresponding to the historical resource type, f ₂ () Representing the operation of taking an index of a preset base.

In the scheme, the second parameter can be accurately calculated by applying the expression for calculating the second parameter, so that the diversity threshold can be accurately determined according to the second parameter, and the accuracy of resource recommendation can be improved according to the diversity threshold.

In one embodiment of the disclosure, the probability prediction unit is specifically configured to include:

the access probability is predicted according to the following expression:

In the scheme, the access probability can be accurately calculated by applying the expression for calculating the access probability, so that the diversity threshold can be accurately determined, and the accuracy of resource recommendation can be improved according to the diversity threshold.

In one embodiment of the present disclosure, the number determination submodule is specifically configured to:

determining a target resource quantity of the resources of the history resource type to be recommended to the target user according to the following expression:

w＝show*exp(γ*x)

According to the scheme, the target resource quantity can be accurately calculated according to the expression for calculating the target resource quantity, so that the diversity threshold can be accurately determined, and the accuracy of resource recommendation can be improved according to the diversity threshold.

In one embodiment of the disclosure, the threshold calculation submodule is specifically configured to:

the diversity threshold corresponding to each historical resource type is determined as follows:

calculating the target duty ratio of the target resource quantity corresponding to the historical resource type;

if the target duty ratio is smaller than or equal to a preset first duty ratio, determining that the first duty ratio is a diversity threshold corresponding to the historical resource type;

if the target duty ratio is larger than the first duty ratio and smaller than a preset second duty ratio, determining that the target duty ratio is a diversity threshold corresponding to the historical resource type, wherein the first duty ratio is smaller than the second duty ratio;

And if the target duty ratio is greater than or equal to the second duty ratio, determining that the second duty ratio is a diversity threshold corresponding to the historical resource type.

In the scheme, after the target duty ratio is calculated, the target duty ratio, the first duty ratio and the second duty ratio can be compared, if the target duty ratio is smaller than or equal to the first duty ratio, the target duty ratio is determined to be a diversity threshold value, and the diversity threshold value is too small, so that the quantity of resources of the historical resource types recommended to the user is small; if the target duty ratio is greater than or equal to the second duty ratio, the target duty ratio is determined to be the diversity threshold value, the diversity threshold value is too large, so that the number of resources of the history resource type recommended to the user is large, and accordingly aesthetic fatigue and freshness of the user are reduced.

In one embodiment of the present disclosure, the apparatus further comprises:

the user judging module is used for judging whether the target user is an active user or not before the display statistical information and the access statistical information are acquired;

The statistical information obtaining module 503 is specifically configured to:

if the target user is an active user, acquiring the resource display quantity and the resource accessed rate, wherein the resource display quantity is as follows: the amount of information of the historical resources presented to the target user, the resource accessed rate being: a ratio of resources accessed by the target user among the historical resources;

if the target user is an inactive user, the resource display quantity, the resource display rate and the resource accessed rate are obtained, wherein the resource display rate is as follows: and the ratio of resources of the historical resource type, which are displayed to the target user, in the historical resources.

In the technical scheme of the disclosure, the related processes of collecting, storing, using, processing, transmitting, providing, disclosing and the like of the personal information of the user accord with the regulations of related laws and regulations, and the public order colloquial is not violated.

According to embodiments of the present disclosure, the present disclosure also provides an electronic device, a readable storage medium and a computer program product.

In one embodiment of the present disclosure, there is provided an electronic device including:

at least one processor; and

a memory communicatively coupled to the at least one processor; wherein,

the memory stores instructions executable by the at least one processor to enable the at least one processor to perform any one of the resource recommendation methods of the method embodiments described above.

In one embodiment of the present disclosure, a non-transitory computer-readable storage medium storing computer instructions for causing a computer to perform any one of the resource recommendation methods of the foregoing method embodiments is provided.

In one embodiment of the present disclosure, a computer program product is provided, comprising a computer program which, when executed by a processor, implements a resource recommendation method of any of the preceding method embodiments.

Fig. 6 illustrates a schematic block diagram of an example electronic device 600 that may be used to implement embodiments of the present disclosure. Electronic devices are intended to represent various forms of digital computers, such as laptops, desktops, workstations, personal digital assistants, servers, blade servers, mainframes, and other appropriate computers. The electronic device may also represent various forms of mobile devices, such as personal digital processing, cellular telephones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions, are meant to be exemplary only, and are not meant to limit implementations of the disclosure described and/or claimed herein.

As shown in fig. 6, the apparatus 600 includes a computing unit 601 that can perform various appropriate actions and processes according to a computer program stored in a Read Only Memory (ROM) 602 or a computer program loaded from a storage unit 608 into a Random Access Memory (RAM) 603. In the RAM 603, various programs and data required for the operation of the device 600 may also be stored. The computing unit 601, ROM 602, and RAM 603 are connected to each other by a bus 604. An input/output (I/O) interface 605 is also connected to bus 604.

Various components in the device 600 are connected to the I/O interface 605, including: an input unit 606 such as a keyboard, mouse, etc.; an output unit 607 such as various types of displays, speakers, and the like; a storage unit 608, such as a magnetic disk, optical disk, or the like; and a communication unit 609 such as a network card, modem, wireless communication transceiver, etc. The communication unit 609 allows the device 600 to exchange information/data with other devices via a computer network, such as the internet, and/or various telecommunication networks.

The computing unit 601 may be a variety of general and/or special purpose processing components having processing and computing capabilities. Some examples of computing unit 601 include, but are not limited to, a Central Processing Unit (CPU), a Graphics Processing Unit (GPU), various specialized Artificial Intelligence (AI) computing chips, various computing units running machine learning model algorithms, a Digital Signal Processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 601 performs the respective methods and processes described above, such as a resource recommendation method. For example, in some embodiments, the resource recommendation method may be implemented as a computer software program tangibly embodied on a machine-readable medium, such as storage unit 608. In some embodiments, part or all of the computer program may be loaded and/or installed onto the device 600 via the ROM 602 and/or the communication unit 609. When the computer program is loaded into the RAM 603 and executed by the computing unit 601, one or more steps of the resource recommendation method described above may be performed. Alternatively, in other embodiments, the computing unit 601 may be configured to perform the resource recommendation method by any other suitable means (e.g. by means of firmware).

Various implementations of the systems and techniques described here above may be implemented in digital electronic circuitry, integrated circuit systems, field Programmable Gate Arrays (FPGAs), application Specific Integrated Circuits (ASICs), application Specific Standard Products (ASSPs), systems On Chip (SOCs), complex Programmable Logic Devices (CPLDs), computer hardware, firmware, software, and/or combinations thereof. These various embodiments may include: implemented in one or more computer programs, the one or more computer programs may be executed and/or interpreted on a programmable system including at least one programmable processor, which may be a special purpose or general-purpose programmable processor, that may receive data and instructions from, and transmit data and instructions to, a storage system, at least one input device, and at least one output device.

Program code for carrying out methods of the present disclosure may be written in any combination of one or more programming languages. These program code may be provided to a processor or controller of a general purpose computer, special purpose computer, or other programmable data processing apparatus such that the program code, when executed by the processor or controller, causes the functions/operations specified in the flowchart and/or block diagram to be implemented. The program code may execute entirely on the machine, partly on the machine, as a stand-alone software package, partly on the machine and partly on a remote machine or entirely on the remote machine or server.

In the context of this disclosure, a machine-readable medium may be a tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. The machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. The machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of a machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

To provide for interaction with a user, the systems and techniques described here can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to a user; and a keyboard and pointing device (e.g., a mouse or trackball) by which a user can provide input to the computer. Other kinds of devices may also be used to provide for interaction with a user; for example, feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form, including acoustic input, speech input, or tactile input.

The systems and techniques described here can be implemented in a computing system that includes a background component (e.g., as a data server), or that includes a middleware component (e.g., an application server), or that includes a front-end component (e.g., a user computer having a graphical user interface or a web browser through which a user can interact with an implementation of the systems and techniques described here), or any combination of such background, middleware, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: local Area Networks (LANs), wide Area Networks (WANs), and the internet.

The computer system may include a client and a server. The client and server are typically remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other. The server may be a cloud server, a server of a distributed system, or a server incorporating a blockchain.

It should be appreciated that various forms of the flows shown above may be used to reorder, add, or delete steps. For example, the steps recited in the present disclosure may be performed in parallel, sequentially, or in a different order, provided that the desired results of the disclosed aspects are achieved, and are not limited herein.

The above detailed description should not be taken as limiting the scope of the present disclosure. It will be apparent to those skilled in the art that various modifications, combinations, sub-combinations and alternatives are possible, depending on design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present disclosure are intended to be included within the scope of the present disclosure.

Claims

1. A resource recommendation method, comprising:

Recommending the determined resources to a client used by the target user.

2. The method of claim 1, wherein the diversity threshold corresponding to the historical resource type of the target user is determined as follows:

acquiring resource information of the historical resources recommended to the target user;

according to the obtained resource information, determining a historical resource type of the historical resource;

acquiring the display statistical information and the access statistical information;

and determining a diversity threshold corresponding to the historical resource type of the target user according to the display statistical information and the access statistical information.

3. The method of claim 2, wherein the obtaining the presentation statistics and the access statistics comprises:

obtaining the number of resource display and the resource access rate, wherein the number of resource display is as follows: the amount of information of the historical resources presented to the target user, the resource accessed rate being: and the ratio of the resources accessed by the target user in the historical resources.

4. A method according to claim 3, wherein said determining a diversity threshold corresponding to said historical resource type of said target user from said presentation statistics and said access statistics comprises:

Determining a first parameter corresponding to the historical resource type according to the resource display quantity, wherein the first parameter characterizes the influence degree of the display condition of the resources of the historical resource type on recommending the resources of the historical resource type;

and determining a diversity threshold corresponding to the historical resource type of the target user according to the determined first parameter and the resource accessed rate.

5. The method of claim 4, wherein the determining, according to the resource exposure number, the first parameter corresponding to the historical resource type includes:

wherein the index_show_ratio represents a first parameter corresponding to the historical resource type, the total_show represents a total resource display number of resources of all the historical resource types, the show represents a display number of the resources of the historical resource type, and the f ₁ () Representing a square root taking operation.

6. The method according to claim 4 or 5, wherein said determining a diversity threshold corresponding to the historical resource type of the target user according to the determined first parameter and the resource accessed rate comprises:

ratio＝α*click_ratio+β*inver_show_ratio

the ratio represents a diversity threshold corresponding to the historical resource type of the target user, the click_ratio represents a resource access rate of the resource of the historical resource type, the alpha represents a first preset weight, and the beta represents a second preset weight.

7. The method of claim 6, wherein the α and the β are determined according to the following expression:

α+β＝1。

8. the method according to claim 4 or 5, wherein said determining a diversity threshold corresponding to the historical resource type of the target user according to the determined first parameter and the resource accessed rate comprises:

9. The method of claim 3, wherein the obtaining the resource exposure quantity and the resource accessed rate comprises:

the determining the diversity threshold corresponding to the historical resource type of the target user according to the display statistical information and the access statistical information comprises the following steps:

predicting the access probability of the target user to the resources of the historical resource type according to the resource display rate and the resource accessed rate;

determining a target resource quantity of the resources of the historical resource type to be recommended to the target user according to the access probability and the resource display quantity;

and calculating the quantity ratio of the target resource quantity corresponding to each historical resource type as a diversity threshold corresponding to each historical resource type.

10. The method of claim 9, wherein the predicting the probability of access of the target user to the resource of the historical resource type based on the resource exposure rate and the resource accessed rate comprises:

determining a second parameter representing the interest degree of the target user on the resources of the historical resource type according to the resource display rate and the resource accessed rate;

and predicting the access probability of the target user to the resources of the historical resource type according to the second parameter and the resource display rate.

11. The method of claim 10, wherein the determining a second parameter characterizing the target user's level of interest in the resource of the historical resource type based on the resource exposure rate and the resource access rate comprises:

the second parameter is determined according to the following expression:

reward＝click_ratio-f ₂ (show_ratio)

wherein the click_ratio represents a resource access rate of the resources of the historical resource type, the show_ratio represents a resource display rate of the resources of the historical resource type, the rewind represents a second parameter corresponding to the historical resource type, and the f ₂ () Representing the operation of taking an index of a preset base.

12. The method according to claim 10 or 11, wherein said predicting the access probability of the target user to access the resource of the historical resource type based on the second parameter and the resource exposure rate comprises:

the access probability is predicted according to the following expression:

wherein x represents the access probability corresponding to the historical resource type, the request represents the second parameter corresponding to the historical resource type, and the show_ratio represents the resource display rate of the resource of the historical resource type.

13. The method according to any of claims 9-11, wherein the determining a target number of resources of the historical resource type to be recommended to the target user according to the access probability and the number of resource impressions comprises:

w＝show*exp(γ*x)

the w represents the number of target resources corresponding to the historical resource type, the show represents the number of resource display of the resources of the historical resource type, the x represents the access probability corresponding to the historical resource type, and the gamma represents a third preset weight.

14. The method according to any one of claims 9-11, wherein the calculating a number-to-number ratio of the target resource number corresponding to each historical resource type as the diversity threshold corresponding to each historical resource type includes:

15. The method of claim 2, further comprising, prior to the obtaining the presentation statistics and the access statistics:

judging whether the target user is an active user or not;

the obtaining the presentation statistics and the access statistics includes:

16. A resource recommendation device, comprising:

17. The apparatus of claim 16, wherein the diversity threshold corresponding to the historical resource type of the target user is determined using:

the resource information obtaining module is used for obtaining resource information of the historical resources recommended to the target user;

the type determining module is used for determining the historical resource type of the historical resource according to the obtained resource information;

the statistical information acquisition module is used for acquiring the display statistical information and the access statistical information;

and the threshold determining module is used for determining a diversity threshold corresponding to the historical resource type of the target user according to the display statistical information and the access statistical information.

18. The apparatus of claim 17, wherein the resource information obtaining module comprises:

19. The apparatus of claim 18, wherein the threshold determination module comprises:

20. The apparatus of claim 19, wherein the parameter determination submodule is configured to:

21. The apparatus of claim 19 or 20, wherein the threshold determination submodule is specifically configured to:

ratio＝α*click_ratio+β*inver_show_ratio

22. The apparatus of claim 21, wherein the α and the β are determined according to the following expression:

α+β＝1。

23. the apparatus of claim 19 or 20, wherein the threshold determination submodule is specifically configured to:

24. The apparatus of claim 18, wherein the information acquisition sub-module is specifically configured to:

the threshold determination module includes:

25. The apparatus of claim 24, wherein the probability prediction submodule comprises:

26. The apparatus of claim 25, wherein the parameter determination unit is specifically configured to:

the second parameter is determined according to the following expression:

reward＝click_ratio-f ₂ (show_ratio)

27. The apparatus according to claim 25 or 26, wherein the probability prediction unit is specifically configured to include:

the access probability is predicted according to the following expression:

28. The apparatus according to any of claims 24-26, wherein the number determination submodule is specifically configured to:

w＝show*exp(γ*x)

29. The apparatus of any of claims 24-26, wherein the threshold calculation submodule is specifically configured to:

30. The apparatus of claim 17, further comprising:

the statistical information acquisition module is specifically configured to:

31. An electronic device, comprising:

at least one processor; and

a memory communicatively coupled to the at least one processor; wherein,

The memory stores instructions executable by the at least one processor to enable the at least one processor to perform the method of any one of claims 1-15.

32. A non-transitory computer readable storage medium storing computer instructions for causing the computer to perform the method of any one of claims 1-15.

33. A computer program product comprising a computer program which, when executed by a processor, implements the method according to any of claims 1-15.