Media resource recommendation method, medium and product

Abstract

Embodiments of the present disclosure relate to a media resource recommendation method, medium and product, wherein the method comprises: in response to a resource recommendation request for a target object, first screening media resources of interest of the object based on object information to form an initial recommendation result; then selecting part of the target type media resources according to the urgency of the target type media resources that need to be recommended in an emergency, updating the initial recommendation result, and further recommending the updated recommendation result to the target object; in this way, by synergistically integrating the object interest-oriented media resources and the target type media resources with high urgency, the recommendation experience of the target object can be ensured, the cold start problem of the target type media resources that need to be recommended in an emergency can be alleviated, and the exposure rate of the target type media resources can be improved.

Description

Technical Field

[0001] This disclosure relates to the field of computer technology, and in particular to a method, medium, and product for recommending media resources. Background Technology

[0002] With the continuous development of computer technology, recommender systems have been widely used in many fields, capable of identifying and recommending relevant media resources from massive amounts of data for users. However, current resource recommendation methods have limitations, making it difficult for newly released media resources to gain exposure and achieve a cold start. Summary of the Invention

[0003] To solve the above-mentioned technical problems, or at least partially solve them, this disclosure provides a media resource recommendation method, medium, and product.

[0004] One aspect of this disclosure provides a media resource recommendation method, including: In response to a resource recommendation request for a target object, a resource set is determined from a media resource library based on the object information of the target object, the resource set including at least one media resource; Based on the recommendation parameters of the target type media resources in the media resource library, at least one first media resource is determined from the target type media resources, wherein the recommendation parameters are used to indicate the recommendation urgency of the target type media resource; The resource set is updated based on the at least one first media resource to obtain the updated resource set; The updated resource set of media resources is recommended to the target object.

[0005] Another aspect of this disclosure provides a media resource recommendation device, including: The determination module is configured to, in response to a resource recommendation request for a target object, determine a resource set from a media resource library based on the object information of the target object, the resource set including at least one media resource; The determining module is further configured to determine at least one first media resource from the target type media resources based on the recommendation parameters of the target type media resources in the media resource library, wherein the recommendation parameters are used to indicate the recommendation urgency of the target type media resource; The update module is used to update the resource set based on the at least one first media resource to obtain the updated resource set; The recommendation module is used to recommend media resources from the updated resource set to the target object.

[0006] In some embodiments, the determining module is configured to obtain the ranking parameters of the target type media resources based on the matching degree between the target type media resources and the target object and the recommendation parameters of the target type media resources; and to determine the at least one first media resource from the target type media resources based on the ranking parameters of the target type media resources, wherein the ranking parameters of the first media resource are higher than the ranking parameters of other target type media resources.

[0007] In some embodiments, the determining module is further configured to determine a second media resource from the target type media resources based on the recommendation parameters of the target type media resources, wherein the recommendation parameters of the second media resource are higher than the recommendation parameters of other target type media resources; The determining module is used to obtain the ranking parameters of the second media resource based on the matching degree between the second media resource and the target object and the recommendation parameters of the second media resource; and to determine the at least one first media resource from the second media resource based on the ranking parameters of the second media resource, wherein the ranking parameters of the first media resource are higher than the ranking parameters of the media resources other than the first media resource in the second media resource.

[0008] In some embodiments, the determining module is further configured to determine a third media resource that satisfies the recommendation strategy from the target type media resources based on the recommendation strategy; The determining module is used to determine the second media resource from the third media resource based on the recommendation parameters of the third media resource.

[0009] In some embodiments, the determining module is configured to perform at least one of the following: According to the first recommendation strategy, based on the resource information of the target type media resources and the object information of the target object, the third media resource is determined from the target type media resources; According to the second recommendation strategy, based on the recommendation parameters of the target type media resources, at least one target type media resource with the highest recommendation parameters is determined as the third media resource.

[0010] In some embodiments, the update module is further configured to obtain the traffic ratio of the target type media resource; and randomly generate a probability for updating the resource set for the at least one first media resource; The update module is used to update the resource set based on the at least one first media resource when the probability is less than the traffic ratio, so as to obtain the updated resource set.

[0011] In some embodiments, the update module is configured to perform any of the following: The updated resource set is obtained by randomly replacing the media resources in the resource set with at least one first media resource. The updated resource set is obtained by replacing all media resources in the resource set with at least one first media resource. All of the at least one first media resource are added to the resource set to obtain the updated resource set; The at least one first media resource is randomly added to the resource set to obtain the updated resource set.

[0012] In some embodiments, the determining module is further configured to determine media resources in the media resource library that meet the recommendation criteria as the target type media resources, wherein the recommendation criteria indicate that the publication duration of the media resource is less than a threshold, or that the media resource is a specified media resource.

[0013] In some embodiments, the determining module is further configured to determine recommended parameters for the target type media resource based on the exposure parameters of the target type media resource, wherein the exposure parameters indicate the exposure amount or exposure progress of the target type media resource, and the exposure progress indicates the ratio of the exposure amount of the target type media resource to the target exposure amount.

[0014] In some embodiments, the determining module is configured to determine recommended parameters for the target type media resource based on the exposure parameters and time parameters of the target type media resource, wherein the time parameters indicate the release duration or time progress of the target type media resource, and the time progress indicates the ratio of the release duration to the target duration of the target type media resource.

[0015] In some embodiments, the exposure parameter indicates the exposure progress of the target type media resource, and the time parameter indicates the time progress of the target type media resource; the determining module is configured to determine a first parameter and a second parameter based on the exposure progress and the time progress of the target type media resource, wherein the first parameter is negatively correlated with the exposure progress and positively correlated with the time progress, the difference between the exposure progress and the target exposure progress is negatively correlated with the second parameter, and the second parameter is positively correlated with the time progress, and the target exposure progress is a fixed target value; the sum of the first parameter and the second parameter is determined as the recommended parameter for the target type media resource, and the target exposure progress is a fixed target value.

[0016] In some embodiments, the exposure parameter indicates the exposure amount of the target type media resource; the determining module is configured to, if the exposure amount of the target type media resource is not less than the target exposure amount, determine the exposure amount interval to which the exposure amount of the target type media resource belongs from a preset plurality of exposure amount intervals based on the exposure amount of the target type media resource, and determine the recommended parameter corresponding to the exposure amount interval to be the recommended parameter of the target type media resource.

[0017] Another aspect of this disclosure provides an electronic device, which includes a processor and a memory, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor performs the media resource recommendation method described above.

[0018] Another aspect of this disclosure provides a computer-readable storage medium storing a computer program that, when executed by a processor, can implement the media resource recommendation method described above.

[0019] Another aspect of this disclosure provides a computer program product, including a computer program that, when executed by a processor, implements the media resource recommendation method described above.

[0020] In the solution provided by this embodiment, in response to a resource recommendation request for a target object, media resources of interest to the object are first filtered based on the object information to form an initial recommendation result; then, in combination with the urgency of the target type media resources that need urgent recommendation, some resources are selected from the target type media resources to update the initial recommendation result, and then the updated recommendation result is recommended to the target object; in this way, by collaboratively integrating media resources that are interest-oriented to the object and target type media resources with high urgency, the recommendation experience of the target object can be guaranteed, the cold start problem of target type media resources that need urgent recommendation can be alleviated, and the exposure rate of similar target type media resources can be improved. Attached Figure Description

[0021] The accompanying drawings, which are incorporated in and form a part of this specification, illustrate embodiments consistent with this disclosure and, together with the description, serve to explain the principles of this disclosure.

[0022] To more clearly illustrate the technical solutions in the embodiments of this disclosure or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0023] Figure 1This is a flowchart illustrating a media resource recommendation method provided in an embodiment of this disclosure; Figure 2 This is a flowchart illustrating another media resource recommendation method provided in this embodiment of the disclosure; Figure 3 This is a schematic diagram of a booster provided in an embodiment of this disclosure; Figure 4 This is a flowchart illustrating another media resource recommendation method provided in this embodiment of the disclosure; Figure 5 This is a schematic diagram of a process for assisting in the recommendation of target type media resources provided in an embodiment of this disclosure; Figure 6 This is a schematic diagram of the structure of a media resource recommendation device provided in an embodiment of this disclosure; Figure 7 This is a schematic diagram of the structure of an electronic device according to an embodiment of this disclosure. Detailed Implementation

[0024] To better understand the above-mentioned objectives, features, and advantages of this disclosure, the solutions disclosed herein will be further described below. It should be noted that, unless otherwise specified, the embodiments and features described herein can be combined with each other.

[0025] Numerous specific details are set forth in the following description in order to provide a full understanding of this disclosure, but this disclosure may also be implemented in other ways different from those described herein; obviously, the embodiments in the specification are only some, and not all, of the embodiments of this disclosure.

[0026] The term "comprising" and its variations as used herein are open-ended, meaning "including but not limited to". The term "based on" means "at least partially based on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments". Definitions of other terms will be given in the description below. The concepts of "first", "second", etc., used in this disclosure are used only to distinguish different devices, modules, or units and are not intended to limit the order of functions performed by these devices, modules, or units or their interdependencies. The modifications of "a" and "a plurality" used in this disclosure are illustrative and not restrictive, and those skilled in the art should understand that unless explicitly indicated in the context, they should be understood as "one or more".

[0027] In recommendation systems, a common "Matthew effect" exists, meaning that media resources with high exposure and high click-through rates continue to receive more traffic, while newly released media resources struggle to gain exposure, resulting in a large number of media resources being buried. However, current resource recommendation methods have limitations, making it difficult for newly released media resources to gain exposure and achieve a cold start. To address the above problems, this disclosure provides a media resource recommendation method, which will be described below with reference to specific embodiments.

[0028] Figure 1 This is a flowchart illustrating a media resource recommendation method provided in an embodiment of this disclosure. This method can be executed by an electronic device. The electronic device can be exemplarily understood as a device such as a mobile phone, tablet computer, laptop computer, desktop computer, or smart TV. Figure 1 As shown, the method provided in this embodiment includes the following steps: S101. In response to a resource recommendation request for a target object, a resource set is determined from a media resource library based on the object information of the target object. The resource set includes at least one media resource.

[0029] In this embodiment of the disclosure, there are special types of media resources, namely target type media resources, among the media resources included in the media resource library. The recommendation of target type media resources may not meet expectations. Therefore, a recommendation parameter is set for target type media resources to indicate the urgency of recommendation. When recommending media resources to an object, the recommendation parameter of the target type media resources will be combined to recommend some target type media resources to the object as much as possible, so as to improve the recommendation of target type media resources and increase the exposure of target type media resources.

[0030] The target object is the recipient of the resource recommendation request, specifically the user who needs to receive media resource recommendations. The resource recommendation request is used to request media resources to be recommended to the target object. The media resource library is a database used to store media resources. It includes a large number of media resources, which exist in the form of text, images, audio, video, animation, etc., and are used to convey information, express content, or meet specific usage needs. For example, media resources might be housing listings, used to display the apartment layout, location, decoration details, price, etc. The target object's information describes the user it represents. For example, object information can describe the user's interests and preferences, such as historical interaction records of media resources the target object has interacted with, interest tags, etc. Object information can be represented in any form; for example, it can be represented as text. The media resources in the resource set are determined based on the target object's object information. Therefore, the media resources in the resource set are essentially media resources that match the object information, and also media resources that the target object is interested in.

[0031] S102. Based on the recommendation parameters of the target type media resources in the media resource library, determine at least one first media resource from the target type media resources. The recommendation parameters are used to indicate the urgency of recommending the target type media resources.

[0032] In this embodiment, the media resource library stores target type media resources, as well as non-target type media resources. Target type media resources are those in the media resource library that require urgent recommendation. To recommend these target type media resources to the target object, upon receiving a resource recommendation request for the target object, at least one first media resource is determined from the target type media resources in the media resource library according to the urgency of the recommendation. This first first media resource is then used to update the set of resources subsequently recommended to the target object.

[0033] In this embodiment, media resources in the media resource library that require urgent recommendation are referred to as target type media resources. That is, target type is equivalent to urgent recommendation type, and media resources belonging to the urgent recommendation type are target type media resources. From the perspective of whether urgent recommendation is needed, media resources in the media resource library can be divided into two categories: target type media resources and non-target type media resources. The recommendation parameter of target type media resources is positively correlated with the recommendation urgency of target type media resources; that is, the larger the recommendation parameter value of a target type media resource, the higher its recommendation urgency; the smaller the recommendation parameter value of a target type media resource, the lower its recommendation urgency.

[0034] S103. Update the resource set based on at least one first media resource to obtain the updated resource set.

[0035] In this embodiment of the disclosure, the media resources in the resource set are aligned with the interests and preferences of the target audience, while the first media resource is a media resource selected based on the urgency of recommendation. By updating the resource set with at least one first media resource, the updated resource set can include media resources of this type that require urgent recommendation. Consequently, when recommending media resources to the target audience in the future, it is possible to recommend media resources that match their interests and media resources of the type that meet the urgent recommendation requirements.

[0036] S104. Recommend media resources from the updated resource set to the target object.

[0037] In this embodiment of the disclosure, by recommending media resources from the updated resource set to the target object, the response to the resource recommendation request is realized, enabling the target object to conveniently query and obtain recommended media resources that meet its needs, thereby improving response efficiency and user experience.

[0038] In the solution provided by this embodiment, in response to a resource recommendation request for a target object, media resources of interest to the object are first filtered based on the object information to form an initial recommendation result; then, in combination with the urgency of the target type media resources that need urgent recommendation, some resources are selected from the target type media resources to update the initial recommendation result, and then the updated recommendation result is recommended to the target object; in this way, by collaboratively integrating media resources that are interest-oriented to the object and target type media resources with high urgency, the recommendation experience of the target object can be guaranteed, the cold start problem of target type media resources that need urgent recommendation can be alleviated, and the exposure rate of similar target type media resources can be improved.

[0039] In the above Figure 1 Based on the embodiments shown, this disclosure will combine the recommended parameters of the target type media resources and the matching degree with the target object to determine the first media resource, and combine the traffic ratio to update the resource set. For details of the process, please refer to the following embodiments. Figure 2 A flowchart illustrating another media resource recommendation method provided in an embodiment of this disclosure is shown. This method can be executed by an electronic device, such as... Figure 2 As shown, this media resource recommendation method may include the following steps: S201. In response to a resource recommendation request for a target object, a resource set is determined from a media resource library based on the object information of the target object. The resource set includes at least one media resource.

[0040] In some embodiments, the process of determining the resource set includes: based on the matching degree between the object information of the target object and the resource information of each media resource in the media resource library, forming a resource set by selecting at least one media resource in the media resource library with the highest matching degree. The resource information describes the corresponding media resource; for example, it describes the type and content of the corresponding media resource. The resource information can be represented in any form, such as text. The matching degree between the object information and the resource information can be calculated in any way, such as using similarity calculation, with the calculated similarity used as the matching degree. Similarity calculation methods include cosine similarity, Euclidean distance, etc.

[0041] In this embodiment of the disclosure, resource information is used to characterize media resources, and the matching degree between object information and resource information is equivalent to the target object's interest in media resources. By selecting at least one media resource with the largest matching degree from the media resource library to form a resource set, it is ensured that the media resources in the resource set are media resources that the target object is interested in, thereby ensuring the accuracy of subsequent resource recommendations.

[0042] For example, the media resources in the media resource library are sorted in descending order of the matching degree between the object information of the target object and the resource information of each media resource in the media resource library. The first media resource in the sort is the first media resource with the highest matching degree, and the first media resource in the sort constitutes a resource set.

[0043] In some embodiments, the matching degree between the resource information of the candidate media resource and the object information of the target object is, in other words, the matching degree between the candidate media resource and the target object. The process of obtaining the matching degree includes: processing the resource information of the candidate media resource and the object information of the target object using a content ranking model to obtain the matching degree. The content ranking model is a model that predicts the click-through rate of a media resource based on object information and resource information. After inputting the resource information of the candidate media resource and the object information of the target object into the content ranking model, the model outputs a predicted click-through rate, which is the probability that the target object will click on the candidate media resource when it is recommended to the target object. This predicted click-through rate is determined as the matching degree between the candidate media resource and the target object.

[0044] In some embodiments, the process of determining the resource set includes: determining candidate media resources from a media resource library based on at least one recall strategy; and forming a resource set by selecting at least one candidate media resource with the highest matching degree based on the matching degree between the resource information of the candidate media resources and the object information of the target object. Different recall strategies indicate different recall methods. For example, if one recall strategy is a personalized recall strategy, it will recall media resources of interest to the target object based on the object information; another recall strategy is a fallback recall strategy, which will recall highly popular media resources from the media resource library.

[0045] In this embodiment of the disclosure, different recall strategies indicate different recall methods. Candidate media resources are recalled from the media resource library based on at least one recall strategy, and the candidate media resources recalled by different recall strategies are merged. On this basis, they are further filtered according to the matching degree with the object information to ensure that the media resources contained in the resource set are media resources that users are interested in, thus ensuring the accuracy of subsequent recommendations.

[0046] In some embodiments, the media resource library includes media resources of multiple different resource types. The process of determining the resource set includes: for each resource type, based on the matching degree between the object information of the target object and the resource information of each media resource under the current resource type, selecting at least one media resource with the highest matching degree from among the media resources under the current resource type to form a candidate resource set corresponding to the current resource type. The candidate resource set includes at least one media resource under the current resource type. The candidate resource sets corresponding to multiple resource types are then merged to obtain a merged candidate resource set. Based on the matching degree between the object information of the target object and the resource information of each media resource in the merged candidate resource set, at least one media resource with the highest matching degree in the merged candidate resource set is selected to form a resource set. Here, multiple resource types refer to the type of content contained in the media resource or the type of format of the media resource itself. For example, multiple resource types may be text, image, or video; or multiple resource types may be housing listings or information, such as housing listings including new housing listings, used housing listings, rental housing listings, etc.

[0047] In this embodiment of the disclosure, considering that there are media resources of various resource types in the media resource library, based on the object information of the target object, media resources that the target object may be interested in are first selected from the media resources under each resource type, and then media resources that the target object is more interested in are further selected from the media resources selected under multiple resource types. In this way, the recommendation of media resources under multiple resource types can be taken into account, so as to enrich the resource types of media resources recommended to the target object. At the same time, it can ensure that the recommended media resources match the interests of the target object and ensure the accuracy of subsequent resource recommendations.

[0048] In some embodiments, step S201 includes: receiving a resource recommendation request sent by a terminal corresponding to the target object; and in response to the resource recommendation request, determining a resource set from a media resource library based on the object information of the target object. The resource recommendation request carries an object identifier or object information of the target object. The object identifier is a unique identifier used to represent the target object and can be represented in any form, such as an account, string, or device identifier. The terminal corresponding to the target object is the terminal used by the user represented by the target object, that is, the terminal logged in based on the object identifier. If the resource recommendation request carries an object identifier, the electronic device can query the object information corresponding to the object identifier from the stored object information based on the object identifier in the resource recommendation request, and then recommend media resources for the object; if the resource recommendation request carries object information, the electronic device can directly recommend media resources for the object based on the object information in the resource recommendation request.

[0049] S202. Based on the matching degree between the target type media resources and the target object and the recommendation parameters of the target type media resources, obtain the ranking parameters of the target type media resources. The recommendation parameters are used to indicate the urgency of recommending the target type media resources.

[0050] In this embodiment of the disclosure, considering that there are many target type media resources in the media resource library, not only are the recommendation parameters of the target type media resources used, but also the matching degree between the target type media resources and the target object is combined to determine the sorting parameters for each target type media resource for the target object. In order to select at least one first media resource for the target object by combining the sorting parameters, this can take into account the recommendation urgency of the target type media resources and the matching with the target object, ensuring that the target type media resources with high urgency are recommended to the target object as much as possible, while also ensuring that the recommended target type media resources are as consistent as possible with the target object's interests, thus ensuring the subsequent recommendation effect.

[0051] The matching degree reflects the target audience's interest in media resources of a specific type. A higher matching degree indicates greater interest, while a lower matching degree indicates less interest. The matching degree can be represented by any numerical value; for example, it can be represented by a value within the range (0,1). A matching degree closer to 1 indicates greater interest. The ranking parameter is equivalent to a recommendation score calculated for the target audience, taking into account both the matching degree and the recommendation parameters. Subsequently, media resources of the same type with higher ranking parameters are recommended to the target audience first.

[0052] In some embodiments, step S202 includes: for any target type media resource, performing a weighted sum of the matching degree corresponding to the target type media resource and the recommendation parameters of the target type media resource to obtain the ranking parameters of the target type media resource. In this embodiment of the disclosure, for any target type media resource, combining the weight of the matching degree and the weight of the recommendation parameters, performing a weighted sum of the matching degree between the target type media resource and the target object and the recommendation parameters of the target type media resource to obtain the ranking parameters of the target type media resource, can ensure that the ranking parameters meet the requirements, balancing the goal of exposing the target type media resource and recommending media resources of interest to the target object, so that a very urgent but not very relevant media resource to the target object will be ranked higher in the final ranking, but will not be absolutely at the top, so as to ensure the accuracy of the ranking parameters, and thus ensure the accuracy of subsequent media resource recommendations based on this.

[0053] In some embodiments, the matching degree between the target type media resource and the target object, the recommendation parameters of the target type media resource, and the ranking parameters of the target type media resource satisfy the following formula (1): Score = α·boost Score +(1-α)·Rank Score (1) Where Score represents the sorting parameter for the target type of media resources, boost... Score Rank is a recommended parameter used to represent the target type of media resources. Score The term is used to represent the matching degree between target type media resources and target objects. α is used to represent the weight of the recommendation parameters, and 1-α is used to represent the weight of the matching degree.

[0054] In some embodiments, the process of obtaining the matching degree includes: obtaining the matching degree between the target type media resource and the target object based on the object information of the target object and the resource information of the target type media resource. In this embodiment, since object information can characterize the target object and resource information can characterize the target type media resource, combining the two can calculate the matching degree between the target type media resource and the target object, thereby ensuring the accuracy of the matching degree. Furthermore, the matching degree can be determined in any way, for example, by using a similarity calculation method, and using the calculated similarity as the matching degree. The similarity calculation method can be cosine similarity, Euclidean distance, etc.

[0055] In some embodiments, the process of determining a target type media resource from a media resource library includes: identifying media resources in the media resource library that meet recommended criteria as target type media resources. The recommended criteria indicate that the publication duration of the media resource is below a threshold, or that the media resource is a specified media resource. The publication duration of a media resource refers to the duration for which the media resource has been published in the media resource library. In this embodiment, after a media resource is published, it can be accessed by an object. The threshold can be any duration, for example, 30 days. The specified media resource refers to a media resource specified by the administrator.

[0056] In this embodiment of the disclosure, the media resource library includes a large number of media resources. Only media resources in the media resource library that meet the recommendation criteria are referred to as target type media resources, while media resources that do not meet the recommendation criteria are referred to as non-target type media resources.

[0057] In some embodiments, by obtaining the attribute information of each media resource in the media resource library, it is possible to obtain the publication duration or whether the media resource has been specified, thereby determining whether the media resource meets the recommendation criteria. The attribute information records various dimensions of the media resource's attributes, such as publication duration and specified dimensions.

[0058] In some embodiments, the process of obtaining recommendation parameters includes: determining recommendation parameters for the target type media resource based on its exposure parameters. The exposure parameters indicate the exposure volume or exposure progress of the target type media resource, and the exposure progress indicates the ratio of the exposure volume to the target exposure volume. Specifically, the exposure parameters indicate the exposure status of the target type media resource after its publication, such as the exposure volume or exposure progress; the exposure volume refers to the number of times the target type media resource is exposed, that is, the number of times the target type media resource is recommended. The target exposure volume is a preset exposure volume and can be any value, for example, a target exposure volume of 50 times.

[0059] In this embodiment of the disclosure, for a target type of media resource, it is necessary to achieve a certain exposure (such as a target exposure). Therefore, by combining the actual exposure parameters of the target type of media resource, the difference between the actual exposure parameters and the target can be determined, which can reflect the urgency of recommending the target type of media resource. In this way, the recommendation parameters of the target type of media resource are determined to ensure that the recommendation parameters match the actual exposure parameters of the target type of media resource and ensure the accuracy of the recommendation parameters.

[0060] In some embodiments, the process of determining recommended parameters using exposure parameters includes: determining recommended parameters for the target type media resource based on the exposure parameters and time parameters of the target type media resource. The time parameter indicates the release duration or time schedule of the target type media resource, and the time schedule indicates the ratio of the release duration to the target duration. The release duration of the target type media resource refers to the duration after the target type media resource is released in the media resource library, and the target duration can be any duration, for example, 30 days.

[0061] In this embodiment of the disclosure, for a target type of media resource, it is necessary to achieve a certain exposure (such as target exposure) within a certain period of time (such as target duration). Therefore, by combining the actual exposure parameters and time progress of the target type of media resource, the difference between the actual exposure parameters and the target and the remaining time can be determined, thereby determining the urgency of recommending the target type of media resource, that is, determining the recommendation parameters of the target type of media resource, so as to ensure that the recommendation parameters match the actual exposure and release of the target type of media resource and ensure the accuracy of the recommendation parameters.

[0062] In some embodiments, the exposure parameter indicates the exposure progress of the target type media resource, and the time parameter indicates the time progress of the target type media resource. The process of determining recommended parameters based on the exposure parameter and the time parameter includes: determining a first parameter and a second parameter based on the exposure progress and time progress of the target type media resource. The first parameter is negatively correlated with the exposure progress and positively correlated with the time progress. The difference between the exposure progress and the target exposure progress is negatively correlated with the second parameter, and the second parameter is positively correlated with the time progress. The sum of the first parameter and the second parameter is determined as the recommended parameter for the target type media resource. The target exposure progress is a fixed target value, for example, a target exposure progress of 100%. In this embodiment, the target exposure progress is used to characterize whether the current exposure of the target type media resource has reached the preset target exposure, i.e., to measure the degree to which the exposure target has been achieved.

[0063] In this embodiment, by combining the time progress and exposure progress of the target type media resources, a first parameter that is positively correlated with the time progress and negatively correlated with the amount of exposure is constructed, as well as a second parameter that is positively correlated with the target exposure gap and positively correlated with the time progress. The two parameters are superimposed in real time as the recommendation parameter. This achieves rapid exposure through time incentives when the resources are first launched, and increases the intensity of the campaign through gap compensation when the target exposure is not achieved, avoiding exposure lag and over-campaigning issues. Ultimately, this simultaneously improves the utilization rate of inventory resources, the campaign achievement rate, and the user's fresh experience.

[0064] In some embodiments, the recommended parameters for target type media resources satisfy the following formula (2): (2) in, Recommended parameters used to represent the target type of media resources. Used to indicate the exposure progress of media resources of a target type. Used to indicate the time progress of a target type of media resource; , , , These are used to represent the weighting coefficients, both of which are constants. In the above formula (2), This ensures that the lower the exposure progress and the longer the release time of the target type of media resources, the higher the recommendation urgency. It is a variant of the Gaussian function that can play the role of "tail lifting". That is, when the exposure progress is close to 100% (for example, more than 90% completed), this part will increase sharply, thereby greatly improving the recommendation parameters and ensuring that the target type of media resources can "just in time" achieve the exposure target, that is, the exposure amount reaches the target exposure amount.

[0065] It should be noted that the above embodiment uses the combination of exposure parameters and time parameters of the target type media resource to determine the recommended parameters as an example. In another embodiment, the recommended parameters can also be determined based solely on the exposure parameters. For example, based on the exposure progress of the target type media resource, a third parameter and a fourth parameter are determined. The third parameter is negatively correlated with the exposure progress, and the difference between the exposure progress and the target exposure progress is negatively correlated with the fourth parameter. The sum of the third parameter and the fourth parameter is determined as the recommended parameters for the target type media resource.

[0066] For example, the third parameter is equivalent to the one in formula (2) above. The fourth parameter is equivalent to the one in formula (2) above. .

[0067] It should be noted that the above method for determining recommendation parameters for target type media resources is only one example. In other embodiments, other methods can be used to determine recommendation parameters for target type media resources. In some embodiments, multiple emergency recommendation stages are preset, and the method for calculating recommendation parameters differs in different emergency recommendation stages. For example, there are two preset emergency recommendation stages, namely, a first emergency recommendation stage and a second emergency recommendation stage; the above method for determining recommendation parameters for target type media resources is used as the calculation method corresponding to the first emergency recommendation stage; for the second emergency recommendation stage, multiple exposure intervals and recommendation parameters corresponding to each exposure interval are preset, the exposure interval to which the current exposure of the target type media resource belongs is determined from the multiple exposure intervals, and the recommendation parameters corresponding to the exposure interval to which it belongs are determined as the recommendation parameters for the target type media resource. The recommendation parameters corresponding to different exposure intervals are different. For example, if there are multiple exposure ranges (50,60], (60,70], and (70,80], and the exposure of the target type media resource is 55, which falls within the exposure range (50,60], then the recommended parameter corresponding to the exposure range (50,60] will be determined as the recommended parameter for the target type media resource.

[0068] In this embodiment, the first and second emergency recommendation stages are divided based on whether the exposure of the target type media resource is less than the target exposure. If the exposure of the target type media resource is less than the target exposure, the target type media resource is in the first emergency recommendation stage, i.e., the uniform quantity maintenance stage; if the exposure of the target type media resource is not less than the target exposure, the target type media resource is in the second emergency recommendation stage, i.e., the tiered boosting stage. For example, in the uniform quantity maintenance stage, the relationship between the boost score (i.e., recommendation parameter), exposure completion rate, and boost time progress of the target type media resource is as follows: Figure 3 As shown.

[0069] In some embodiments, when the recommendation system recommends media resources to various objects using a media resource library, the recommendation system can periodically (e.g., hourly) collect online recommendation logs through offline tasks. The online recommendation logs include the exposure and interaction status of each media resource in the media resource library; based on the online recommendation logs, the exposure or exposure progress of the target type of media resource can be obtained.

[0070] It should be noted that before recommending media resources to each object, target type media resources that meet the recommendation criteria are pre-selected from the media resource library, and recommendation parameters are calculated for each target type media resource. Then, the recommendation parameters of each target type media resource are cached so that when recommending media resources to any object, the recommendation parameters of each target type media resource can be directly obtained from the cache.

[0071] For example, after determining the recommendation parameters for each target type of media resource, these parameters are stored in Elasticsearch (ES). ES can also store information about each media resource in the media resource library, including recommendation parameters, whether it requires urgent recommendation, and its current urgent recommendation stage. The recommendation parameters for non-target type media resources may be 0 or a preset fixed value. The recommendation parameters for target type media resources are calculated as shown in the above embodiment. The "whether it requires urgent recommendation" field indicates whether the corresponding media resource is a target type resource; those requiring urgent recommendation are target type media resources, and those not requiring urgent recommendation are non-target type media resources. The "current urgent recommendation stage" indicates which urgent recommendation stage the target type media resource currently belongs to.

[0072] Furthermore, while caching media resources of various target types, the system periodically recalculates the recommendation parameters of each target type of media resource based on its actual recommendation performance, and updates the cache accordingly to ensure the accuracy of the cached recommendation parameters. The update cycle for recommendation parameters is hourly, daily, etc. For example, when a target type of media resource is recommended to a user, meaning it has been successfully exposed to the user, its exposure log is recorded. In the next offline calculation cycle (e.g., after 1 hour), the system collects these recommendation logs and updates the exposure volume of the target type of media resource by incrementing the exposure volume by 1. Subsequently, the offline system recalculates the recommendation parameters of the target type of media resource. As the exposure volume of the target type of media resource increases and the exposure progress rises, the recommendation parameters of the target type of media resource automatically decrease. When the exposure volume of a target type of media resource reaches the target exposure volume (e.g., 50 times), the target type of media resource will automatically exit the uniform volume maintenance phase and no longer occupy this portion of the media resource traffic. It then enters the tiered recommendation phase to obtain more efficiency-based traffic amplification. This achieves an automatic, closed-loop, and efficient media resource-assisted recommendation mechanism.

[0073] S203. Based on the sorting parameters of the target type media resources, determine at least one first media resource from the target type media resources, wherein the sorting parameters of the first media resource are higher than the sorting parameters of other target type media resources.

[0074] In this embodiment, since the sorting parameter takes into account both the matching degree between the target type media resource and the target object and the urgency of recommending the target type media resource, the target type media resources with larger sorting parameters are preferentially selected as the media resources to be recommended to the target object. For example, the target type media resources are sorted in descending order of sorting parameters, and at least one target type media resource at the top of the sort is selected as the first media resource, such as selecting the top 10 target type media resources as the first media resources.

[0075] It should be noted that the embodiments disclosed herein are illustrated by taking the combination of recommended parameters of target type media resources and matching degree with target object to determine the first media resource. In another embodiment, it is not necessary to perform the above steps S202-S203. Instead, other methods are adopted to determine at least one first media resource from the target type media resources based on the recommended parameters of target type media resources in the media resource library.

[0076] S204. Obtain the traffic ratio of the target type of media resources.

[0077] In this embodiment of the disclosure, the traffic ratio is used to limit the proportion of target type media resources among the recommended media resources when recommending media resources to various objects based on media resources in the media resource library, so as to ensure that the exposure of target type media resources meets expectations. The traffic ratio can be any ratio, for example, a traffic ratio of 20%.

[0078] In some embodiments, the traffic ratio is pre-stored in a database, making it possible to retrieve the traffic ratio from the database. Alternatively, the traffic ratio of the current target type media resource can be determined based on the current predicted click-through rate (pCTR) of the target type media resource, the current average click-through rate of all media resources in the media resource library, and the overall exposure progress of the target type media resource.

[0079] The current average click-through rate (CTR) of the target type media resource refers to the average number of clicks for that media resource within the current statistical period (e.g., the last hour or 30 minutes). It is the ratio of the current cumulative clicks to the current cumulative exposures of the target type media resource. The current average CTR of the target type media resource reflects its real-time efficiency in attracting user clicks. The current average CTR of all media resources in the media resource library refers to the overall average number of clicks for all media resources in the library within the current statistical period. It is the ratio of the current cumulative total clicks to the current cumulative total exposures of all media resources in the media resource library. The current average CTR of all media resources in the media resource library reflects the real-time overall click efficiency of the media resource library. The overall exposure progress of the target type media resource refers to the ratio of the current exposures achieved by the target type media resource to the preset total exposure target for the target type media resource. The overall exposure progress of a target type of media resource is the ratio of the current cumulative exposure of the target type of media resource to the preset total exposure target of the target type of media resource. It can reflect the progress of the exposure task of the target type of media resource. The lower the progress, the greater the gap from the preset total exposure target.

[0080] In some embodiments, the process of determining the traffic ratio includes: determining the ratio of the current average click-through rate of the target type media resource to the current average click-through rate of all media resources in the media resource library, and weighting and integrating the ratio with the overall exposure progress of the target type media resource to obtain the current traffic ratio of the target type media resource.

[0081] In this embodiment of the disclosure, before updating the resource set, a dynamic formula is used to calculate the proportion of traffic that needs to be allocated to the target type of media resources. This calculation takes into account the comparison between the average click-through rate (CTR) of the target type of media resources and the average CTR of all media resources in the media resource library, as well as the progress towards achieving the overall exposure target. If the exposure efficiency of the target type of media resources is high or the exposure target is not achieved, the traffic proportion will be increased accordingly.

[0082] S205. Randomly generate the probability of updating the resource set for at least one first media resource.

[0083] In this embodiment of the disclosure, after determining the first media resource from the target type media resources, a probability is randomly generated for the first media resource. This probability is used to represent the possibility of updating the resource set with the first media resource.

[0084] S206. When the probability is less than the traffic ratio, update the resource set based on at least one first media resource to obtain the updated resource set.

[0085] In this embodiment, by first obtaining the traffic ratio of the target type media resources, then randomly generating the probability of updating the resource set with the first media resource, and using the probability being less than the traffic ratio as the update trigger condition, the traffic allocation of the target type media resources is realized. In this way, the distribution ratio of the target type media resources is clearly defined through the traffic ratio, ensuring that its exposure meets expectations. Furthermore, the random probability mechanism avoids rigid traffic allocation, making resource set updates more flexible, improving the overall diversity and delivery effect of the resource set, ensuring the controllability of recommending target type media resources to the target audience, and balancing the achievement rate of exposure goals with user experience.

[0086] In some embodiments, the process of updating the resource set based on the first media resource includes any of the following.

[0087] The first step: Randomly replace media resources in the resource set with at least one first media resource to obtain an updated resource set. In this embodiment of the disclosure, a random replacement method is adopted, where first media resources are used to randomly replace media resources in the resource set. This can retain some of the original media resources in the resource set while balancing the recommendation needs of target type media resources with the user's experience of media resource diversity. For example, the resource set includes 10 media resources, and the first media resources include 5; after triggering an update according to the traffic ratio R=30%, 3 media resources are randomly selected from the resource set, and 3 first media resources are randomly selected from the 5 first media resources; the 3 randomly selected first media resources are replaced with the 3 randomly selected media resources from the resource set, that is, the updated resource set includes the original 7 media resources and the 3 randomly selected first media resources.

[0088] The second step: Replace all media resources in the resource set with at least one first media resource to obtain an updated resource set. In this embodiment, replacing all media resources in the resource set with first media resources can maximize the concentration of traffic towards the target type of media resources, quickly make up for the exposure gap of the target type of media resources, and ensure that the target type of media resource recommendation task is completed on time. For example, the resource set includes 10 media resources, and the first media resources include 6; after triggering the update according to the traffic ratio R=30%, all 10 media resources in the resource set are replaced with 6 first media resources, that is, the updated resource set includes 6 first media resources.

[0089] The third step: Add at least one primary media resource to the resource set to obtain an updated resource set. In this embodiment, directly adding all primary media resources to the resource set can quickly expand the display base of the target type of media resources without affecting the exposure of the original media resources in the resource set. This method can improve the richness of the resource set and give the target type of media resources more display opportunities, ensuring the recommendation effect of the target type of media resources. For example, if the resource set includes 10 media resources and 6 primary media resources; after triggering an update according to the traffic ratio R=30%, adding all 6 primary media resources to the resource set will result in an updated resource set including the original 10 media resources and 6 primary media resources.

[0090] Fourthly: Randomly add at least one first media resource to the resource set to obtain an updated resource set. In this embodiment, randomly selecting a portion of the first media resources to add to the resource set balances the exposure needs of the target type of media resources with the dynamic balance of the resource set. This avoids the resource set becoming too large and diluting media resources that match the target audience's interests due to adding all of them. At the same time, the random mechanism ensures that different first media resources have a fair exposure opportunity, guaranteeing recommendation effectiveness and user experience. For example, the resource set includes 10 media resources, and 6 of them are first media resources. After triggering an update based on a traffic ratio R=30%, 3 first media resources are randomly selected from the 6 first media resources and added to the resource set. The updated resource set then includes the original 10 media resources and 3 randomly selected first media resources.

[0091] It should be noted that the embodiments disclosed herein are illustrated by taking the updating of the resource set in combination with the traffic ratio as an example. In another embodiment, it is not necessary to perform the above steps S204-S206. Instead, other methods are adopted to update the resource set based on at least one first media resource to obtain the updated resource set.

[0092] S207. Recommend media resources from the updated resource set to the target object.

[0093] In some embodiments, the updated resource link of the media resource in the resource set is sent to the device corresponding to the target object, or the media resource is sent directly. In this embodiment of the disclosure, when recommending media resources in the updated resource set to the target object, the media resource can be sent directly to the device corresponding to the target object so that the device can display the media resource; or, the resource link of the media resource is sent to the device so that the device can access the media resource through the resource link.

[0094] In the solution provided by this embodiment, in response to a resource recommendation request for a target object, media resources of interest to the object are first filtered based on the object information to form an initial recommendation result; then, in combination with the urgency of the target type media resources that need urgent recommendation, some resources are selected from the target type media resources to update the initial recommendation result, and then the updated recommendation result is recommended to the target object; in this way, by collaboratively integrating media resources that are interest-oriented to the object and target type media resources with high urgency, the recommendation experience of the target object can be guaranteed, the cold start problem of target type media resources that need urgent recommendation can be alleviated, and the exposure rate of similar target type media resources can be improved.

[0095] In the above Figure 2 Based on the embodiments shown, this disclosure first filters from target type media resources based on recommendation parameters, and then filters from the filtering results again based on sorting parameters. For details of the process, please refer to the following embodiments. Figure 4 The diagram illustrates a flowchart of another media resource recommendation method provided in this disclosure, which can be executed by an electronic device, such as... Figure 4 As shown, this media resource recommendation method may include the following steps: S401. In response to a resource recommendation request for a target object, a resource set is determined from a media resource library based on the object information of the target object. The resource set includes at least one media resource.

[0096] It should be noted that step S401 is the same as step S201 above, and will not be repeated here.

[0097] S402. Based on the recommended parameters of the target type media resources, determine the second media resource from the target type media resources, wherein the recommended parameters of the second media resource are higher than the recommended parameters of other target type media resources.

[0098] In this embodiment of the disclosure, considering that there are many target type media resources in the media resource library and that the recommendation urgency of different target type media resources is different, the recommendation parameters of each target type media resource in the media resource library are first combined to filter out target type media resources with high recommendation urgency. Then, for the selected target type media resources, the ranking parameters calculated based on the recommendation parameters and matching degree are used for further filtering. This can take into account both the recommendation urgency of the target type media resources and the matching situation with the target object, ensuring that the target type media resources with high urgency are recommended to the target object as much as possible, while also ensuring that the recommended target type media resources are as consistent as possible with the target object's interests, thus ensuring the subsequent recommendation effect.

[0099] For example, target type media resources are arranged in descending order of recommendation parameters, and the first target type media resource is identified as the second media resource.

[0100] In some embodiments, target type media resources that satisfy the recommendation strategy are first determined, and then a second media resource is determined from these target type media resources. That is, the process of determining the second media resource includes: determining a third media resource that satisfies the recommendation strategy from the target type media resources based on the recommendation strategy; and determining the second media resource from the third media resource based on the recommendation parameters of the third media resource. The recommendation strategy is used to recommend target type media resources to an audience.

[0101] In this embodiment of the disclosure, considering that there are many target type media resources in the media resource library and that the recommendation urgency of different target type media resources is different, some target type media resources are first screened based on the recommendation strategy, and then screened again from the screened target type media resources according to the recommendation parameters. In this way, screening according to the recommendation strategy can ensure that the third media resources meet the recommendation requirements for the target object, avoid selecting irrelevant media resources, ensure the adaptability of target type media resources and recommendation objects, and prioritize the push of resources with higher recommendation urgency, thus ensuring the accuracy and urgency of the recommendation.

[0102] In some embodiments, the process of determining third-party media resources based on a recommendation strategy includes at least one of the following.

[0103] The first approach: Following the first recommendation strategy, based on the resource information of the target type media resources and the object information of the target object, determine the third media resource from the target type media resources.

[0104] In this embodiment of the disclosure, the first recommendation strategy indicates that target-type media resources of interest to the object are recommended, which is also equivalent to a personalized recommendation strategy. Therefore, according to the first recommendation strategy, based on the resource information of the target-type media resources and the object information of the target object, a third media resource that matches the interests of the target object is determined from the target-type media resources to ensure the accuracy of subsequent recommendations.

[0105] In some embodiments, based on the resource information of the target type media resource and the object information of the target object, the matching degree between the target type media resource and the target object is determined, and at least one target type media resource with the highest matching degree is determined as the third media resource. It should be noted that the process of determining the matching degree based on resource information and object information is the same as the process of determining the matching degree in the above embodiments, and will not be repeated here.

[0106] The second approach: According to the second recommendation strategy, based on the recommendation parameters of the target type media resources, at least one target type media resource with the highest recommendation parameters is identified as the third media resource.

[0107] In this embodiment, the second recommendation strategy prioritizes recommending target type media resources with high recommendation parameters. This second recommendation strategy also acts as a fallback strategy to ensure that target type media resources with high recommendation parameters are recommended, preventing the most urgent media resources from being overlooked. For example, according to the second recommendation strategy, target type media resources are arranged in descending order of recommendation parameters, and at least one media resource at the top is identified as the third media resource. By using one or more of the above recommendation strategies to select third media resources for the target object, the diversity of the selected media resources is enriched, thereby ensuring the effectiveness of subsequent recommendations.

[0108] It should be noted that the two methods mentioned above can be combined. For example, the third media resources identified by the two recommendation strategies can be used as a whole, and then second media resources can be selected from these third media resources based on the recommendation parameters of these third media resources.

[0109] S403. Based on the matching degree between the second media resource and the target object and the recommendation parameters of the second media resource, obtain the ranking parameters of the second media resource.

[0110] It should be noted that step S403 is the same as step S202 above, and will not be repeated here.

[0111] S404. Based on the sorting parameters of the second media resources, determine at least one first media resource from the second media resources, wherein the sorting parameters of the first media resources are higher than the sorting parameters of the media resources other than the first media resources in the second media resources.

[0112] It should be noted that step S404 is the same as step S203 above, and will not be repeated here.

[0113] It should be noted that in this embodiment, the target type media resources are first filtered based on recommendation parameters, and then filtered again based on ranking parameters. In another embodiment, it is not necessary to perform the above steps S402-S404. Instead, other methods are adopted to obtain the ranking parameters of the target type media resources based on the matching degree between the target type media resources and the target object and the recommendation parameters of the target type media resources. Based on the ranking parameters of the target type media resources, at least one first media resource is determined from the target type media resources, and the ranking parameters of the first media resource are higher than the ranking parameters of other target type media resources.

[0114] S405. Obtain the traffic ratio of the target type of media resources.

[0115] S406. Randomly generate the probability of updating the resource set for at least one first media resource.

[0116] S407. When the probability is less than the traffic ratio, update the resource set based on at least one first media resource to obtain the updated resource set.

[0117] S408. Recommend media resources from the updated resource set to the target object.

[0118] It should be noted that steps S405-S408 are the same as steps S204-S207 above, and will not be repeated here.

[0119] In the solution provided by this embodiment, in response to a resource recommendation request for a target object, media resources of interest to the object are first filtered based on the object information to form an initial recommendation result; then, in combination with the urgency of the target type media resources that need urgent recommendation, some resources are selected from the target type media resources to update the initial recommendation result, and then the updated recommendation result is recommended to the target object; in this way, by collaboratively integrating media resources that are interest-oriented to the object and target type media resources with high urgency, the recommendation experience of the target object can be guaranteed, the cold start problem of target type media resources that need urgent recommendation can be alleviated, and the exposure rate of similar target type media resources can be improved.

[0120] It should be noted that the above-mentioned multiple optional embodiments can be combined in any way, and this disclosure does not limit them.

[0121] Based on the embodiments shown above, this disclosure also provides a flowchart for assisting in the recommendation of target type media resources, such as... Figure 5 As shown, the process of recommending target type media resources includes an offline component and an online application component; therefore, the method includes: Step 1: Collect online recommendation logs hourly through offline tasks.

[0122] In this embodiment of the disclosure, the collected online recommendation logs can also be displayed on a dashboard for administrators to view.

[0123] Step 2: Based on the status of each media resource in the media resource library, filter out all media resources that need to be boosted (i.e., target type media resources).

[0124] Step 3: Determine whether the exposure of the boosting media resources is less than the target exposure. If the exposure of the boosting media resources is less than the target exposure, then proceed to the uniform exposure maintenance stage. If the exposure of the boosting media resources is not less than the target exposure, then proceed to the tiered boosting stage.

[0125] Step 4: If the uniform quantity preservation stage is entered, calculate the boost score (i.e., recommended parameter) of the boost media resources according to the above formula (2).

[0126] Step 5: If the tiered boosting stage is entered, the boosting score of the boosting media resources is determined by combining multiple exposure ranges and corresponding boosting scores, as shown in the above embodiment.

[0127] Step 6: Write the boosting information of media resources into an online Elasticsearch database on an hourly basis for real-time querying by the online recommendation system.

[0128] It should be noted that steps 1-6 above are offline parts and are executed by the offline system.

[0129] Step 7: Upon receiving a user recommendation request, for media resources in the media resource library, a recommendation process is performed to obtain a set of resources to be recommended to the user; for boosting media resources in the media resource library, a recommendation process is performed to obtain boosting media resources to be recommended; the boosting media resources to be recommended are used to update the set of resources to be recommended to the user in the manner shown in the above embodiment to obtain the final resource set.

[0130] The recommendation process includes recall, fusion, ranking, and mixed ranking. Recall refers to selecting a subset of media resources according to different recall strategies; fusion refers to merging the media resources selected by different recall strategies and then further filtering them based on boost scores; ranking refers to further filtering the merged media resources according to ranking parameters; mixed ranking refers to merging the recommended resources under each resource type after recall, fusion, and ranking, then ranking and filtering the recommended resources from multiple resource types to obtain a set of recommended resources for the user.

[0131] For example, the media resource library contains various types of media resources, including new home listings, used home listings, rental listings, and news information. For each type of listing, after initial recall, a subset of listings is selected using different recall strategies. Then, in the fusion phase, the listings selected by different recall strategies are merged and further filtered based on the boost score of the merged listings. Next, in the ranking phase, the listings selected in the fusion phase are further filtered based on the predicted click-through rate between the fusion-stage listings and users. This results in new home listings, used home listings, and rental listings after the ranking phase. These three types of listings are then combined, and based on the matching degree between the combined listings and users, a set of listings to be recommended to users (i.e., the new / used / rental listing set) is selected. Furthermore, the media resource library includes various types of resource information (i.e., content materials), which are also filtered in the same way to obtain a set of news information to be recommended to users (i.e., the content material set). Furthermore, following the method described in the above embodiments, for the boosting media resources, after recall, a portion of the boosting media resources are selected according to different recall strategies; then, in the fusion stage, the boosting media resources selected by different recall strategies are merged and then filtered according to the boosting score of the merged boosting media resources; after that, in the sorting stage, the boosting media resources selected in the fusion stage are filtered according to the sorting score of the boosting media resources selected in the fusion stage to obtain the boosting media resources to be recommended (i.e., the boosting material set). Then, the content material set is inserted into a new set according to the admission and exit criteria to obtain a new set, and the boosting material set is then inserted into the new set to obtain the final resource set.

[0132] Step 8: Recommend media resources from the final resource collection to the user.

[0133] It should be noted that steps 7-8 above belong to the online application part and are executed by the recommendation system.

[0134] The solution provided in this disclosure proposes an automated boosting mechanism for all media resources, designing an end-to-end boosting framework that includes "uniform quantity assurance" and "tiered boosting." In the "uniform quantity assurance" stage, a recommendation parameter calculation formula combining "exposure progress" and "time progress" is designed. This not only increases the boosting priority for media resources that have not received exposure for a long time but also introduces a "tail-lifting" mechanism. That is, when the media resource's exposure is about to reach its target, the boosting score increases sharply, ensuring that the media resource can efficiently achieve its exposure target. Furthermore, a dynamic traffic proportion calculation model is introduced, which can adjust the proportion of boosted media resource traffic in the total traffic in real time based on the actual online performance (e.g., pCTR) of boosted media resources and natural traffic media resources, as well as the overall achievement of the exposure target. This ensures the exposure target of boosted media resources while minimizing the negative impact on overall recommendation efficiency. Furthermore, to balance the "certainty" of the boosting mechanism with the "personalization" of the user experience, the "boost score," representing the urgency of exposure for boosted media resources, is combined with the "matching degree," representing user interests. This ensures that high-boost-score media resources receive priority exposure, while their ranking also considers user relevance, avoiding the damage to user experience caused by forcibly inserting irrelevant content. Moreover, through an independent boosted media resource recall, fusion, and ranking chain, and ultimately integrating them into the boosting recommendation stream via insertion, the controllability of the boosting task is guaranteed. Simultaneously, the types of media resources supporting boosting can be dynamically added or deleted online, greatly improving operational efficiency and system scalability. The solution provided in this embodiment can automatically provide all newly added media resources with a fair initial exposure opportunity, mitigating the Matthew effect while maintaining the stability of overall recommendation metrics.

[0135] Figure 6 This is a schematic diagram of a media resource recommendation device provided in an embodiment of this disclosure. This media resource recommendation device can be understood as the aforementioned electronic device or a functional module within the aforementioned electronic device. For example... Figure 6 As shown, the media resource recommendation device includes: The determination module 601 is used to respond to a resource recommendation request for a target object, and determine a resource set from the media resource library based on the object information of the target object, wherein the resource set includes at least one media resource; The determining module 601 is further configured to determine at least one first media resource from the target type media resources based on the recommendation parameters of the target type media resources in the media resource library, wherein the recommendation parameters are used to indicate the recommendation urgency of the target type media resources; Update module 602 is used to update the resource set based on at least one first media resource to obtain the updated resource set; Recommendation module 603 is used to recommend media resources from the updated resource set to the target object.

[0136] In some embodiments, the determining module 601 is used to obtain the ranking parameters of the target type media resources based on the matching degree between the target type media resources and the target object and the recommendation parameters of the target type media resources; and to determine at least one first media resource from the target type media resources based on the ranking parameters of the target type media resources, wherein the ranking parameters of the first media resource are higher than the ranking parameters of other target type media resources.

[0137] In some embodiments, the determining module 601 is further configured to determine a second media resource from the target type media resources based on the recommendation parameters of the target type media resources, wherein the recommendation parameters of the second media resource are higher than the recommendation parameters of other target type media resources; The determining module 601 is used to obtain the ranking parameters of the second media resource based on the matching degree between the second media resource and the target object and the recommendation parameters of the second media resource; based on the ranking parameters of the second media resource, at least one first media resource is determined from the second media resource, wherein the ranking parameters of the first media resource are higher than the ranking parameters of the media resources other than the first media resource in the second media resource.

[0138] In some embodiments, the determining module 601 is further configured to determine a third media resource that satisfies the recommendation strategy from the target type media resources based on the recommendation strategy; The determination module 601 is used to determine the second media resource from the third media resource based on the recommendation parameters of the third media resource.

[0139] In some embodiments, the determining module 601 is configured to perform at least one of the following: According to the first recommendation strategy, based on the resource information of the target type media resources and the object information of the target object, the third media resources are determined from the target type media resources; According to the second recommendation strategy, based on the recommendation parameters of the target type media resources, at least one target type media resource with the highest recommendation parameters is identified as the third media resource.

[0140] In some embodiments, the update module 602 is further configured to obtain the traffic ratio of the target type media resource; and randomly generate the probability of updating the resource set for at least one first media resource; The update module 602 is used to update the resource set based on at least one first media resource when the probability is less than the traffic ratio, so as to obtain the updated resource set.

[0141] In some embodiments, the update module 602 is configured to perform any of the following: The media resources in the resource set are randomly replaced by at least one first media resource to obtain an updated resource set; Replace all media resources in the resource set with at least one primary media resource to obtain the updated resource set; Add at least one primary media resource to the resource set to obtain the updated resource set; At least one primary media resource is randomly added to the resource set to obtain the updated resource set.

[0142] In some embodiments, the determining module 601 is further configured to determine media resources in the media resource library that meet the recommendation criteria as target type media resources, wherein the recommendation criteria indicate that the publication duration of the media resource is less than a threshold, or that the media resource is a specified media resource.

[0143] In some embodiments, the determining module 601 is further configured to determine recommended parameters for the target type media resource based on the exposure parameters of the target type media resource, wherein the exposure parameters indicate the exposure amount or exposure progress of the target type media resource, and the exposure progress indicates the ratio of the exposure amount of the target type media resource to the target exposure amount.

[0144] In some embodiments, the determining module 601 is used to determine recommended parameters for the target type media resource based on the exposure parameters and time parameters of the target type media resource. The time parameters indicate the release duration or time progress of the target type media resource, and the time progress indicates the ratio of the release duration to the target duration of the target type media resource.

[0145] In some embodiments, the exposure parameter indicates the exposure progress of the target type media resource, and the time parameter indicates the time progress of the target type media resource; the determining module 601 is used to determine a first parameter and a second parameter based on the exposure progress and the time progress of the target type media resource, wherein the first parameter is negatively correlated with the exposure progress and positively correlated with the time progress, the difference between the exposure progress and the target exposure progress is negatively correlated with the second parameter and the second parameter is positively correlated with the time progress, and the target exposure progress is a fixed target value; the sum of the first parameter and the second parameter is determined as the recommended parameter for the target type media resource.

[0146] In some embodiments, the exposure parameter indicates the exposure amount of the target type media resource; the determining module 601 is used to determine the exposure range to which the exposure amount of the target type media resource belongs from a plurality of preset exposure ranges based on the exposure amount of the target type media resource if the exposure amount of the target type media resource is not less than the target exposure amount, and determine the recommended parameter corresponding to the exposure range to which it belongs as the recommended parameter of the target type media resource.

[0147] It should be noted that the device provided in this embodiment can execute the media resource recommendation method of any of the above embodiments, and its execution method and beneficial effects are similar, so they will not be described again here.

[0148] This disclosure also provides an electronic device, which includes: a memory storing a computer program; and a processor for executing the computer program, wherein when the computer program is executed by the processor, it can implement the media resource recommendation method in the above method embodiments.

[0149] Example, Figure 7 This is a schematic diagram of the structure of an electronic device according to an embodiment of this disclosure. See below for details. Figure 7 The diagram illustrates a structural schematic suitable for implementing the electronic device 700 in the embodiments of this disclosure. The electronic device 700 in the embodiments of this disclosure may include, but is not limited to, mobile terminals such as mobile phones, laptops, digital broadcast receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), in-vehicle terminals (e.g., in-vehicle navigation terminals), and fixed terminals such as digital TVs and desktop computers. Figure 7 The electronic device shown is merely an example and should not be construed as limiting the functionality and scope of the embodiments disclosed herein.

[0150] like Figure 7 As shown, the electronic device 700 may include a processing unit (e.g., a central processing unit, a graphics processor, etc.) 701, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 702 or a program loaded from a storage device 708 into a random access memory (RAM) 703. The RAM 703 also stores various programs and data required for the operation of the electronic device 700. The processing unit 701, ROM 702, and RAM 703 are interconnected via a bus 704. An input / output (I / O) interface 705 is also connected to the bus 704.

[0151] Typically, the following devices can be connected to I / O interface 705: input devices 706 including, for example, touchscreens, touchpads, keyboards, mice, cameras, microphones, accelerometers, gyroscopes, etc.; output devices 707 including, for example, liquid crystal displays (LCDs), speakers, vibrators, etc.; storage devices 708 including, for example, magnetic tapes, hard disks, etc.; and communication devices 709. Communication device 709 allows electronic device 700 to communicate wirelessly or wiredly with other devices to exchange data. Although Figure 7 An electronic device 700 with various devices is shown; however, it should be understood that it is not required to implement or possess all of the devices shown. More or fewer devices may be implemented or possessed alternatively.

[0152] In particular, according to embodiments of this disclosure, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of this disclosure include a computer program product comprising a computer program carried on a non-transitory computer-readable medium, the computer program containing program code for performing the media resource recommendation method shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication device 709, or installed from a storage device 708, or installed from a ROM 702. When the computer program is executed by the processing device 701, it performs the functions defined in the media resource recommendation method of embodiments of this disclosure.

[0153] In some implementations, clients and servers may communicate using any currently known or future-developed network protocol such as HTTP (Hypertext Transfer Protocol) and may interconnect with digital data communication (e.g., communication networks) of any form or medium. Examples of communication networks include local area networks (“LANs”), wide area networks (“WANs”), the Internet (e.g., the Internet of Things), and peer-to-peer networks (e.g., ad hoc peer-to-peer networks), as well as any currently known or future-developed networks.

[0154] The aforementioned computer-readable medium may be included in the aforementioned electronic device; alternatively, it may exist independently and not assembled into the electronic device. The aforementioned computer-readable medium carries one or more programs that, when executed by the electronic device, enable the electronic device to implement the media resource recommendation method in the above embodiments.

[0155] In the context of this disclosure, a machine-readable medium can be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. Machine-readable media can be, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, and portable compact disc read-only memory (CD). The storage medium may be a ROM, an optical storage device, a magnetic storage device, or any suitable combination thereof. Embodiments of this disclosure also provide a computer-readable storage medium storing a computer program that, when executed by a processor, can implement the media resource recommendation method of any of the above embodiments. The execution method and beneficial effects are similar and will not be described again here.

[0156] The above are merely specific embodiments of this disclosure, enabling those skilled in the art to understand or implement this disclosure. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this disclosure. Therefore, this disclosure is not to be limited to these embodiments, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

[0157] The above description is merely a preferred embodiment of this disclosure and an explanation of the technical principles employed. Those skilled in the art should understand that the scope of this disclosure is not limited to technical solutions formed by specific combinations of the above-described technical features, but should also cover other technical solutions formed by arbitrary combinations of the above-described technical features or their equivalents without departing from the above-described concept. For example, technical solutions formed by substituting the above features with (but not limited to) technical features disclosed in this disclosure that have similar functions.

[0158] Furthermore, while the operations are described in a specific order, this should not be construed as requiring these operations to be performed in the specific order shown or in a sequential order. In certain environments, multitasking and parallel processing may be advantageous. Similarly, while several specific implementation details are included in the above discussion, these should not be construed as limiting the scope of this disclosure. Certain features described in the context of individual embodiments may also be implemented in combination in a single embodiment. Conversely, various features described in the context of a single embodiment may also be implemented individually or in any suitable sub-combination in multiple embodiments.

[0159] Although the subject matter has been described using language specific to structural features and / or methodological logic, it should be understood that the subject matter defined in the appended claims is not necessarily limited to the specific features or actions described above. Rather, the specific features and actions described above are merely illustrative examples of implementing the claims.

Claims

1. A media resource recommendation method, characterized in that, include: In response to a resource recommendation request for a target object, a resource set is determined from a media resource library based on the object information of the target object, the resource set including at least one media resource; Based on the recommendation parameters of the target type media resources in the media resource library, at least one first media resource is determined from the target type media resources, wherein the recommendation parameters are used to indicate the recommendation urgency of the target type media resource; The resource set is updated based on the at least one first media resource to obtain the updated resource set; The updated resource set of media resources is recommended to the target object.

2. The method according to claim 1, characterized in that, The step of determining at least one first media resource from the target type media resources based on recommendation parameters in the media resource library includes: Based on the matching degree between the target type media resource and the target object and the recommendation parameters of the target type media resource, the sorting parameters of the target type media resource are obtained; Based on the sorting parameters of the target type media resources, at least one first media resource is determined from the target type media resources, wherein the sorting parameters of the first media resource are higher than the sorting parameters of other target type media resources.

3. The method according to claim 2, characterized in that, Before obtaining the ranking parameters of the target type media resources based on the matching degree between the target type media resources and the target object and the recommendation parameters of the target type media resources, the method further includes: Based on the recommendation parameters of the target type media resources, a second media resource is determined from the target type media resources, wherein the recommendation parameters of the second media resource are higher than the recommendation parameters of other target type media resources; The step of obtaining the ranking parameters of the target type media resources based on the matching degree between the target type media resources and the target object and the recommendation parameters of the target type media resources includes: Based on the matching degree between the second media resource and the target object and the recommendation parameters of the second media resource, the sorting parameters of the second media resource are obtained; The step of determining the at least one first media resource from the target type media resources based on the sorting parameters of the target type media resources includes: Based on the sorting parameters of the second media resource, at least one first media resource is determined from the second media resource, wherein the sorting parameters of the first media resource are higher than the sorting parameters of the media resources other than the first media resource in the second media resource.

4. The method according to claim 3, characterized in that, Before determining the second media resource from the target type media resources based on the recommendation parameters of the target type media resources, the method further includes: Based on the recommendation strategy, a third media resource that satisfies the recommendation strategy is determined from the media resources of the target type; The step of determining a second media resource from the target type media resources based on the recommendation parameters of the target type media resources includes: Based on the recommendation parameters of the third media resource, the second media resource is determined from the third media resource.

5. The method according to claim 4, characterized in that, The step of determining a third media resource that satisfies the recommendation strategy from the target type of media resources includes at least one of the following: According to the first recommendation strategy, based on the resource information of the target type media resources and the object information of the target object, the third media resource is determined from the target type media resources; According to the second recommendation strategy, based on the recommendation parameters of the target type media resources, at least one target type media resource with the highest recommendation parameters is determined as the third media resource.

6. The method according to any one of claims 1 to 5, characterized in that, Before updating the resource set based on the at least one first media resource to obtain the updated resource set, the method further includes: Obtain the traffic ratio of the target type of media resource; Randomly generate the probability of updating the resource set for the at least one first media resource; The step of updating the resource set based on the at least one first media resource to obtain the updated resource set includes: If the probability is less than the traffic ratio, the resource set is updated based on the at least one first media resource to obtain the updated resource set.

7. The method according to any one of claims 1 to 5, characterized in that, The step of updating the resource set based on the at least one first media resource to obtain an updated resource set includes any one of the following: The updated resource set is obtained by randomly replacing the media resources in the resource set with at least one first media resource. The updated resource set is obtained by replacing all media resources in the resource set with at least one first media resource. All of the at least one first media resource are added to the resource set to obtain the updated resource set; The at least one first media resource is randomly added to the resource set to obtain the updated resource set.

8. The method according to any one of claims 1 to 5, characterized in that, Before determining at least one first media resource from the target type media resources based on the recommendation parameters of the media resource library, the method further includes: Media resources in the media resource library that meet the recommendation criteria are identified as the target type of media resources. The recommendation criteria indicate that the publication duration of the media resource is less than a threshold, or that the media resource is a specified media resource.

9. The method according to any one of claims 1 to 5, characterized in that, Before determining at least one first media resource from the target type media resources based on the recommendation parameters of the media resource library, the method further includes: Based on the exposure parameters of the target type media resource, recommended parameters for the target type media resource are determined. The exposure parameters indicate the exposure volume or exposure progress of the target type media resource, and the exposure progress indicates the ratio of the exposure volume of the target type media resource to the target exposure volume.

10. The method according to claim 9, characterized in that, The step of determining the recommended parameters for the target type of media resource based on the exposure parameters of the target type of media resource includes: Based on the exposure parameters and time parameters of the target type media resource, recommended parameters for the target type media resource are determined. The time parameters indicate the release duration or time progress of the target type media resource, and the time progress indicates the ratio of the release duration to the target duration of the target type media resource.

11. The method according to claim 10, characterized in that, The exposure parameter indicates the exposure progress of the target type media resource, and the time parameter indicates the time progress of the target type media resource; determining the recommended parameters of the target type media resource based on the exposure parameter and the time parameter includes: Based on the exposure progress and time progress of the target type media resource, a first parameter and a second parameter are determined. The first parameter is negatively correlated with the exposure progress and positively correlated with the time progress. The difference between the exposure progress and the target exposure progress is negatively correlated with the second parameter and positively correlated with the time progress. The target exposure progress is a fixed target value. The sum of the first parameter and the second parameter is determined as the recommended parameters for the target type of media resource.

12. The method according to claim 9, characterized in that, The exposure parameter indicates the amount of exposure for the target type of media resource; The step of determining the recommended parameters for the target type of media resource based on the exposure parameters of the target type of media resource includes: If the exposure of the target type media resource is not less than the target exposure, based on the exposure of the target type media resource, the exposure range to which the exposure of the target type media resource belongs is determined from a set of preset exposure ranges, and the recommended parameters corresponding to the exposure range to which it belongs are determined as the recommended parameters of the target type media resource.

13. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program for performing the media resource recommendation method as described in any one of claims 1 to 11.

14. A computer program product, characterized in that, It includes a computer program that, when executed by a processor, implements the media resource recommendation method as described in any one of claims 1 to 11.

Images