Resource pushing method and device, apparatus, and storage medium

Abstract

This application provides a method, apparatus, device, and storage medium for resource push, relating to the field of big data resource scheduling technology. Based on the utilization rate of a first historical push resource used by the target object and the relevance between the first historical push resource and each resource tag, the method obtains the target object's preference value for each resource tag; it determines the term frequency (TF) value of each resource tag based on a first resource tag library corresponding to the first historical push resource, and determines the inverse text frequency (IDF) value of each resource tag based on a second resource tag library corresponding to all historical push resources used by the target object; it obtains the target object's target preference value for each resource tag based on the preference value, TF value, and IDF value; and it determines the target push resource from candidate push resources based on the target preference value and pushes it to the target object's terminal device. This application accurately pushes resources to users based on user profiles, improving the utilization rate of target push resources and reducing push costs.

Description

Technical Field

[0001] This application relates to the field of big data resource scheduling technology, and in particular to a method, apparatus, device and storage medium for pushing resources. Background Technology

[0002] With the rapid development of internet technology, resource scheduling applications are expanding, including news recommendations, business recommendations, learning recommendations, and lifestyle recommendations. Resource delivery via the internet is gradually becoming a mainstream method, and the accuracy of the delivery strategy directly impacts user engagement with those resources.

[0003] In many scenarios, users receive push notifications, such as coupons or vouchers from merchants. However, since different users generally have different preferences, if the push notifications are not suitable for the users, it can lead to increased costs for merchants without achieving the expected benefits. Summary of the Invention

[0004] This application provides a method, apparatus, device, and storage medium for pushing resources, in order to solve the problem in the prior art that if the pushed resources are not compatible with the user, it will cause the merchant's costs to increase without achieving the expected revenue.

[0005] In a first aspect, this application provides a method for pushing resources, comprising: obtaining a preference value of the target object for each resource tag based on the usage rate of a first historical push resource already used by the target object and the relevance of the first historical push resource to each resource tag; determining the term frequency (TF) value of each resource tag based on a first resource tag library corresponding to the first historical push resource, and determining the inverse text frequency index (IDF) value of each resource tag based on a second resource tag library corresponding to a second historical push resource already used by all historical objects; obtaining a target preference value of the target object for each resource tag based on the preference value, TF value, and IDF value of each resource tag; and determining a target push resource from candidate push resources based on the target preference value, and pushing it to the target object's terminal device.

[0006] Secondly, this application provides a resource push device, comprising: a first acquisition module, configured to acquire a target object's preference value for each resource tag based on the usage rate of a first historical push resource already used by the target object and the relevance between the first historical push resource and each resource tag; a second acquisition module, configured to determine the term frequency (TF) value of each resource tag based on a first resource tag library corresponding to the first historical push resource, and to determine the inverse text frequency index (IDF) value of each resource tag based on a second resource tag library corresponding to all historical push resources already used by the target object; a third acquisition module, configured to acquire a target preference value of the target object for each resource tag based on the preference value, TF value, and IDF value of each resource tag; and a push module, configured to determine a target push resource from candidate push resources based on the target preference value, and push it to the target object's terminal device.

[0007] Thirdly, this application provides an electronic device, including: a processor, and a memory communicatively connected to the processor; the memory stores computer-executable instructions; the processor executes the computer-executable instructions stored in the memory to implement the resource push method as described above.

[0008] Fourthly, a computer-readable storage medium is provided, wherein computer-executable instructions are stored in the computer-readable storage medium, and the computer-executable instructions are executed by a processor to implement the resource push method as described above.

[0009] Fifthly, a computer program product is provided, including a computer program / instruction, characterized in that, when the computer program / instruction is executed by a processor, it implements the resource push method described above.

[0010] The resource push method, apparatus, device, and storage medium provided in this application analyze the historical push resources used by the user before the current moment to better depict the user profile in real time, so as to determine the target push resources suitable for the user from the candidate push resources, accurately identify the user's preferences, so as to achieve accurate push of push resources, improve the user experience, increase the utilization rate of target push resources, and reduce push costs. Attached Figure Description

[0011] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0012] Figure 1 This is an exemplary implementation of a resource push method proposed in this application.

[0013] Figure 2This is a schematic diagram provided in an embodiment of the present application, representing the usage rate of a first historical push resource corresponding to a certain user and the relevance of the first historical push resource to each resource tag.

[0014] Figure 3 This is an exemplary implementation of a resource push method proposed in this application.

[0015] Figure 4 This is a schematic diagram illustrating how to determine the target push resource corresponding to a new user when the target object is a new user, as provided in an embodiment of this application.

[0016] Figure 5 This application provides an exemplary implementation of determining a target push resource from candidate push resources based on a target preference value.

[0017] Figure 6 This application provides an exemplary implementation method for determining a target push resource from candidate push resources based on a target preference value.

[0018] Figure 7 This is a flowchart illustrating a resource push method provided in an embodiment of this application.

[0019] Figure 8 This is a schematic diagram of a resource push device provided in an embodiment of this application.

[0020] Figure 9 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application.

[0021] The accompanying drawings illustrate specific embodiments of this application, which will be described in more detail below. These drawings and descriptions are not intended to limit the scope of the concept in any way, but rather to illustrate the concept of this application to those skilled in the art through reference to particular embodiments. Detailed Implementation

[0022] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.

[0023] The acquisition, storage, and application of user personal information involved in the technical solution of this application all comply with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0024] Figure 1This is an exemplary implementation of a resource push method proposed in this application, such as... Figure 1 As shown, the method for pushing this resource includes the following steps:

[0025] S101, based on the usage rate of the first historical push resource used by the target object and the relevance between the first historical push resource and each resource tag, obtain the target object's preference value for each resource tag.

[0026] The target object is identified from the candidate objects, and the historical push resources used by the target object are obtained as the first historical push resources. Based on the usage data of the first historical push resources, the usage rate of the first historical push resources corresponding to the target object and the relevance of the first historical push resources to each resource tag are obtained. Based on the target object's usage rate of each first historical push resource and the relevance of the first historical push resources to each resource tag, the preference value of the target object for each resource tag is obtained. Here, the target object can be the user to whom resources are to be pushed, and the first historical push resources are push resources that have been pushed to the target object and used by the target object. Each first historical push resource can correspond to multiple resource tags.

[0027] The utilization rate of any first historical push resource refers to the quotient of the number of times the same first historical push resource was used to the total number of times the first historical push resource was pushed to the target object. For example, taking a consumer voucher as the first historical push resource, if the first historical push resource is a discount coupon of 30 yuan off a purchase of 100 yuan on Platform A, and a user is pushed 10 such coupons, and the user uses 7 of them, then the utilization rate of the discount coupons for that user is 70%.

[0028] Let u represent the target object and t represent the resource tag. Then, the formula for calculating the preference value of the target object for each resource tag is:

[0029]

[0030] In the above formula: level(u,t) represents the preference value of target object u for resource tag t; p(u,i) represents the usage rate of target object u for the first historical push resource i; rel(i,t) represents the relevance between the first historical push resource i and resource tag t.

[0031] For example, taking the first historical push resource as a consumer voucher, the resource tags corresponding to the consumer voucher may include discount coupons, cash coupons, experience coupons, gift coupons, discount coupons, special offer coupons, exchange coupons, and general coupons. Among them, any consumer voucher may correspond to multiple resource tags. For example, the resource tags for a discount coupon of 30 yuan off purchases of 100 yuan or more, which is universal on platform A, may include discount coupons and general coupons.

[0032] For example, Figure 2 This is a diagram illustrating the usage rate of a user's first historical push resource and the relevance of that resource to each resource tag, such as... Figure 2 As shown, a user's first historical push resources include first historical push resource A, first historical push resource B, and first historical push resource C. Each first historical push resource for this user corresponds to three resource tags: resource tag 1, resource tag 2, and resource tag 3. The relevance of first historical push resource A to resource tags 1, 2, and 3 is 0.5, 0.8, and 0.6, respectively; the relevance of first historical push resource B to resource tags 1, 2, and 3 is 0.4, 0.3, and 0.7; and the relevance of first historical push resource C to resource tags 1, 2, and 3 is 0.2, 0.5, and 0.1. If the user's usage rate for first historical push resource A is 90%, for first historical push resource B is 75%, and for first historical push resource C is 30%, then according to the formula, the preference value of target object u for resource tag 1 can be calculated as follows:

[0033]

[0034] Similarly, the preference value of the target object u for resource tag 2 or resource tag 3 can be obtained, which will not be elaborated here.

[0035] S102, determine the term frequency (TF) value of each resource tag based on the first resource tag library corresponding to the first historical push resource, and determine the inverse text frequency index (IDF) value of each resource tag based on the second resource tag library corresponding to the second historical push resources used by all historical objects.

[0036] Obtain the first resource tag library corresponding to all first historical push resources used by the target object. The first resource tag library includes all resource tags corresponding to all first historical push resources used by the target object, as well as the cumulative count of each resource tag. Obtain the cumulative count of all resource tags corresponding to all first historical push resources used by the target object based on the cumulative count of each resource tag. Determine the term frequency (TF) value of each resource tag based on the cumulative count of each resource tag in the first resource tag library and the cumulative count of all resource tags. Let TF(u,t) represent the TF value of resource tag t for target object u.

[0037] The process involves obtaining a second resource tag library corresponding to all second historical push resources used by a historical object. A historical object refers to any object that has received push resources; for example, if a company pushed consumer vouchers to 10 million people, these 10 million people are the historical objects. All historical push resources used by these historical objects are considered second historical push resources. The second resource tag library includes all resource tags corresponding to all second historical push resources used by the historical object, as well as the cumulative count of each resource tag. Based on the cumulative count of each resource tag in the second resource tag library, the cumulative count of all resource tags corresponding to all second historical push resources used by the historical object is obtained. Then, based on the cumulative count of each resource tag in the second resource tag library and the cumulative count of all resource tags, the Inverse Document Frequency (IDF) value of each resource tag is determined. IDF(u,t) represents the IDF value of resource tag t for target object u.

[0038] S103, based on the preference value, TF value and IDF value of each resource tag, obtain the target preference value of the target object for each resource tag.

[0039] Based on the preference value, TF value, and IDF value of each resource tag determined above, the target preference value of the target object for each resource tag is obtained. The formula for calculating the target preference value of target object u for resource tag t is as follows:

[0040] Pre(u,t)=level(u,t)×TF(u,t)×IDF(u,t)

[0041] In the above formula, Pre(u,t) represents the target preference value of target object u for resource label t; level(u,t) represents the preference value of user u for resource label t; TF(u,t) represents the TF value of resource label t for target object u; and IDF(u,t) represents the IDF value of resource label t for target object u.

[0042] S104, Based on the target preference value, determine the target push resource from the candidate push resources and push it to the target object's terminal device.

[0043] Based on the target preference value of each resource tag for the target object as determined above, the target push resource is determined from the candidate push resources and pushed to the target object's terminal device.

[0044] As one feasible approach, the target object's target preference value for each resource tag can be sorted from largest to smallest to determine the largest target preference value among all target preference values. After determining the largest target preference value, the target resource tag corresponding to the largest target preference value is obtained. The relevance of each candidate push resource to the target resource tag is then sorted from largest to smallest among all candidate push resources. The top N candidate push resources after sorting are selected as the target push resources.

[0045] This application proposes a resource push method. It obtains the target object's preference value for each resource tag based on the usage rate of a first historical push resource and the relevance between the first historical push resource and each resource tag. It then determines the term frequency (TF) value of each resource tag based on a first resource tag library corresponding to the first historical push resource, and the inverse text frequency index (IDF) value of each resource tag based on a second resource tag library corresponding to all historical push resources used by the target object. Based on the preference value, TF value, and IDF value of each resource tag, it obtains the target object's target preference value for each resource tag. Finally, based on the target preference value, it determines the target push resource from the candidate push resources and pushes it to the target object's terminal device. This application analyzes the historical push resources used by the user before the current moment to better depict the user profile in real time, thereby determining the target push resource suitable for the user from the candidate push resources, accurately identifying the user's preferences, achieving precise push of push resources, improving the user experience, increasing the usage rate of the target push resource, and reducing push costs.

[0046] Specifically, when determining the term frequency (TF) value of each resource tag based on the first resource tag library corresponding to the first historical push resources, the first resource tag library corresponding to the first historical push resources used by the target object can be obtained. Based on the first resource tag library, the first cumulative count corresponding to each resource tag and the second cumulative count corresponding to all resource tags can be obtained. For any resource tag, the quotient of the first cumulative count and the second cumulative count is used as the TF value of that resource tag.

[0047]

[0048] In the above formula, TF(u,t) represents the TF value of resource tag t for target object u; n(u,t) i This indicates that the target object u in the first resource tag library uses the resource tag t. i The first cumulative count; This represents the second cumulative number of times the target object u in the first resource tag library has used all resource tags.

[0049] Specifically, when determining the Inverse Text Frequency Index (IDF) value of each resource tag based on the second resource tag library corresponding to the second historical push resources used by all historical objects, the second resource tag library corresponding to the second historical push resources used by all historical objects can be obtained. Based on the second resource tag library, the third cumulative count corresponding to each resource tag and the fourth cumulative count corresponding to all resource tags can be obtained. For any resource tag, the IDF value corresponding to that resource tag can be obtained based on the third and fourth cumulative counts.

[0050]

[0051] In the above formula:

[0052] IDF(u,t) represents the IDF value of resource label t for target object u; This represents the fourth cumulative count of all resource tags in the second resource tag library; This indicates the third cumulative count of resource tag t in the second resource tag library.

[0053] Figure 3 This is an exemplary implementation of a resource push method proposed in this application, such as... Figure 3 As shown, based on the above embodiments, the resource push method further includes the following steps:

[0054] S301, obtain the identification information of the target object, and determine whether the target object is a new object based on the identification information.

[0055] When pushing resources to a target audience, the first step is to determine whether the target audience is a new user. Specifically, this can be done by obtaining the target audience's identification information, such as their ID number or registration number, and then using this information to determine if the target audience is a new user.

[0056] S302, in response to the target object being a new object, obtain the target group to which the target object belongs.

[0057] If the target object is determined to be a new object based on the identification information, in order to push push resources suitable for the new object, it is necessary to determine the target group to which the target object belongs.

[0058] For example, when obtaining the target group to which the target object belongs, the user characteristics of the target object can be obtained as the first user characteristics, and the user characteristics of each old user can be obtained as the second user characteristics. The similarity between the first user characteristics of the target object and the second user characteristics of each old user can be calculated. The second user characteristics with the highest similarity to the first user characteristics can be determined. The old user corresponding to the second user characteristics can be taken as the target old user, the user group to which the target old user belongs can be obtained, and the user group to which the target old user belongs can be determined as the target group.

[0059] Optionally, when obtaining user characteristics of a target object or an existing user, the first user characteristic of the target object or the second user characteristic of the existing user can be generated based on the target object's or existing user's gender, age, education, occupation, region, movement trajectory, and usage of push resources.

[0060] S303: Obtain historical push resource usage data of the target group, determine the target push resource corresponding to the target object based on the historical push resource usage data, and push it to the target object's terminal device.

[0061] After identifying the target group to which the target object belongs, the system obtains the historical push resource usage data of that target group. Based on this data, it determines the target push resource corresponding to the target object and pushes it to the target object's terminal device. For example, taking consumer vouchers as the historical push resource, if the historical data indicates that the target group has used m types of consumer vouchers, the system obtains the usage count for each type of voucher and pushes the top 5 vouchers by usage count to the target object's terminal device.

[0062] Figure 4 This is a diagram illustrating how to determine the target push resource for a new user, as shown below. Figure 4 As shown, after obtaining the first user feature of a new user, the second user feature of each old user is determined, and the similarity between the first user feature of the new user and the second user feature of each old user is calculated. The second user feature with the highest similarity to the first user feature is determined, and the old user corresponding to the second user feature is taken as the target old user. For example, if the target old user corresponding to the new user is user group 2, the historical push resource usage data 2 corresponding to user group 2 is obtained, and the target push resource 2 corresponding to the target object is determined based on the historical push resource usage data 2, and the target push resource 2 is pushed to the terminal device of the new user.

[0063] In this embodiment of the application, when the target object is a new object, the target group to which the target object belongs is obtained, and the target push resource corresponding to the target object is determined according to the push resource most frequently used by the target group. This avoids the randomness of push to new users and improves the adaptability of new users to target push resources.

[0064] Figure 5 This application presents an exemplary implementation method for determining a target push resource from candidate push resources based on a target preference value, such as... Figure 5 As shown, it includes the following steps:

[0065] S501, obtain the maximum target preference value among all target preference values, and use the resource tag corresponding to the maximum target preference value as the first target resource tag.

[0066] Sort all target preference values ​​corresponding to all resource tags in descending order, obtain the maximum target preference value among all target preference values, and use the resource tag corresponding to the maximum target preference value as the first target resource tag. For example, if the target preference value corresponding to resource tag 1 is the maximum value among all target preference values, then resource tag 1 will be used as the first target resource tag.

[0067] S502, obtain the first target relevance between each candidate push resource and the first target resource tag.

[0068] The relevance of each candidate push resource to the first target resource tag is obtained as the first target relevance. For example, if the first target resource tag is resource tag 1 and there are 100 candidate push resources, the relevance of each of these 100 candidate push resources to resource tag 1 is obtained as the first relevance of each of these 100 candidate push resources to resource tag 1.

[0069] S503, sort the candidate push resources according to the relevance of the first target, and obtain the candidate push resource queue generated after sorting.

[0070] Candidate push resources are sorted according to the magnitude of the first target relevance, and a queue of candidate push resources is obtained after sorting. For example, if the first target resource tag is resource tag 1, and there are 100 candidate push resources, then after obtaining the first relevance of each of these 100 candidate push resources to resource tag 1, the candidate push resources are sorted according to the magnitude of the first target relevance, and a queue of candidate push resources is obtained after sorting.

[0071] S504, determine the target push resource based on the candidate push resource queue.

[0072] The target push resource is determined based on the candidate push resource queue. Optionally, the top 10 candidate push resources in the candidate push resource queue can be selected as the target push resource.

[0073] In this embodiment, candidate push resources are filtered based on the target object's target preference value for each resource tag, so as to determine the target push resources suitable for the user from the candidate push resources, accurately identify the user's preferences, so as to achieve accurate push of push resources, improve the user's experience, increase the utilization rate of target push resources, and reduce push costs.

[0074] Figure 6 This application presents an exemplary implementation method for determining a target push resource from candidate push resources based on a target preference value, such as... Figure 6 As shown, it includes the following steps:

[0075] S601, sort all target preference values ​​in descending order and obtain the target preference value queue generated after sorting.

[0076] Sort all target preference values ​​corresponding to all resource tags in descending order, and obtain the target preference value queue generated after sorting.

[0077] S602, take the resource tags corresponding to the first M target preference values ​​in the target preference value queue as the second target resource tags.

[0078] The resource tags corresponding to the first M target preference values ​​in the target preference value queue are used as the second target resource tags. Here, M is a preset value.

[0079] For example, if there are 20 resource tags, there are also 20 target preference values ​​in the target preference value queue. When M is 5, the first 5 resource tags in the target preference value queue can be used as the second target resource tags.

[0080] S603, obtain the second target relevance between each candidate push resource and each second target resource tag.

[0081] The relevance of each candidate push resource to each second target resource tag is obtained as the second target relevance. For example, if the second target resource tags are resource tag 1, resource tag 2, resource tag 3, resource tag 4, and resource tag 5, and there are a total of 100 candidate push resources, then the relevance of these 100 candidate push resources to resource tag 1, resource tag 2, resource tag 3, resource tag 4, and resource tag 5 respectively is obtained as the second relevance of these 100 candidate push resources to resource tag 1, resource tag 2, resource tag 3, resource tag 4, and resource tag 5 respectively.

[0082] S604, based on the second target relevance, determine at least one candidate push resource corresponding to each second target resource tag as the target push resource.

[0083] For each second target resource tag, all candidate push resources are sorted in descending order of their relevance to the second target of that tag. The top N candidate push resources after sorting are selected as the target push resources corresponding to that second target resource tag. If there are 5 second target resource tags, then there are a total of 5N target push resources.

[0084] In this embodiment, candidate push resources are filtered based on the target object's target preference value for each resource tag, so as to determine the target push resources suitable for the user from the candidate push resources, accurately identify the user's preferences, so as to achieve accurate push of push resources, improve the user's experience, increase the utilization rate of target push resources, and reduce push costs.

[0085] Figure 7 This is a flowchart illustrating the overall process of a resource delivery method proposed in this application, as follows: Figure 7 As shown, the method for pushing this resource includes the following steps:

[0086] S701, obtain the identification information of the target object, and determine whether the target object is a new object based on the identification information.

[0087] S702, in response to the target object being a new object, obtain the first user characteristics of the target object and the second user characteristics of the old user.

[0088] S703, obtain the similarity between the first user feature and each second user feature, and based on the similarity, determine the old user with the highest similarity to the target object as the target old user.

[0089] S704 identifies the user group to which the target existing users belong as the target group.

[0090] S705: Obtain historical push resource usage data of the target group, determine the target push resource corresponding to the target object based on the historical push resource usage data, and push it to the target object's terminal device.

[0091] The implementation methods for steps S701 to S705 can be referred to the descriptions of the relevant parts in the above embodiments, and will not be repeated here.

[0092] S706, in response to the target being an existing user, obtain the target's preference value for each resource tag based on the usage rate of the first historical push resource already used by the target and the relevance of the first historical push resource to each resource tag.

[0093] S707, determine the term frequency (TF) value of each resource tag based on the first resource tag library corresponding to the first historical push resource, and determine the inverse text frequency index (IDF) value of each resource tag based on the second resource tag library corresponding to the second historical push resources used by all historical objects.

[0094] S708: Based on the preference value, TF value, and IDF value of each resource tag, obtain the target preference value of the target object for each resource tag.

[0095] S709, obtain the maximum target preference value among all target preference values, and use the resource tag corresponding to the maximum target preference value as the first target resource tag.

[0096] S710, obtain the first target relevance between each candidate push resource and the first target resource tag.

[0097] S711, sort the candidate push resources according to the relevance of the first target, and obtain the candidate push resource queue generated after sorting.

[0098] S712 determines the target push resource based on the candidate push resource queue and pushes it to the target object's terminal device.

[0099] The implementation methods for steps S706 to S712 can be referred to the descriptions of the relevant parts in the above embodiments, and will not be repeated here.

[0100] This application proposes a resource push method. It obtains the target object's preference value for each resource tag based on the usage rate of a first historical push resource and the relevance between the first historical push resource and each resource tag. It then determines the term frequency (TF) value of each resource tag based on a first resource tag library corresponding to the first historical push resource, and the inverse text frequency index (IDF) value of each resource tag based on a second resource tag library corresponding to all historical push resources used by the target object. Based on the preference value, TF value, and IDF value of each resource tag, it obtains the target object's target preference value for each resource tag. Finally, based on the target preference value, it determines the target push resource from the candidate push resources and pushes it to the target object's terminal device. This application analyzes the historical push resources used by the user before the current moment to better depict the user profile in real time, thereby determining the target push resource suitable for the user from the candidate push resources, improving the utilization rate of the target push resource, and reducing push costs.

[0101] Figure 8 This is a schematic diagram of a resource delivery device proposed in this application, such as... Figure 8As shown, the resource push device 800 includes a first acquisition module 801, a second acquisition module 802, a third acquisition module 803, and a push module 804, wherein:

[0102] The first acquisition module 801 is used to acquire the target object's preference value for each resource tag based on the usage rate of the first historical push resources used by the target object and the relevance of the first historical push resources to each resource tag;

[0103] The second acquisition module 802 is used to determine the term frequency (TF) value of each resource tag according to the first resource tag library corresponding to the first historical push resource, and to determine the inverse text frequency index (IDF) value of each resource tag according to the second resource tag library corresponding to the second historical push resource used by all historical objects.

[0104] The third acquisition module 803 is used to acquire the target preference value of the target object for each resource tag based on the preference value, TF value and IDF value of each resource tag;

[0105] The push module 804 is used to determine the target push resource from the candidate push resources according to the target preference value, and push it to the target object's terminal device.

[0106] This application proposes a resource push device, comprising: a first acquisition module, configured to acquire a target object's preference value for each resource tag based on the usage rate of a first historical push resource used by the target object and the relevance between the first historical push resource and each resource tag; a second acquisition module, configured to determine the term frequency (TF) value of each resource tag based on a first resource tag library corresponding to the first historical push resource, and to determine the inverse text frequency index (IDF) value of each resource tag based on a second resource tag library corresponding to all historical push resources used by the target object; a third acquisition module, configured to acquire a target preference value of the target object for each resource tag based on the preference value, TF value, and IDF value of each resource tag; and a push module, configured to determine a target push resource from candidate push resources based on the target preference value and push it to the target object's terminal device. This application analyzes the historical push resources used by the user before the current moment to better depict the user profile in real time, thereby determining the target push resource suitable for the user from candidate push resources, improving the usage rate of the target push resource, and reducing push costs.

[0107] Furthermore, the resource push device 800 also includes: a judgment module 805, used to obtain the identification information of the target object and determine whether the target object is a new object based on the identification information; a fourth acquisition module 806, used to obtain the target group to which the target object belongs in response to the target object being a new object; and a determination module 807, also used to obtain the historical push resource usage data of the target group, and determine the target push resource corresponding to the target object based on the historical push resource usage data, and push it to the target object's terminal device.

[0108] Furthermore, the fourth acquisition module 806 is also used to: acquire the first user characteristics of the target object and the second user characteristics of the old user; acquire the similarity between the first user characteristics and each second user characteristic, and based on the similarity, determine the old user with the highest similarity to the target object as the target old user; and determine the user group to which the target old user belongs as the target group.

[0109] Furthermore, the push module 804 is also used to: obtain the maximum target preference value among all target preference values, and use the resource tag corresponding to the maximum target preference value as the first target resource tag; obtain the first target relevance between each candidate push resource and the first target resource tag; sort the candidate push resources according to the magnitude of the first target relevance, and obtain the candidate push resource queue generated after sorting; and determine the target push resource based on the candidate push resource queue.

[0110] Furthermore, the push module 804 is also used to: sort all target preference values ​​in descending order and obtain the target preference value queue generated after sorting; take the resource tags corresponding to the first M target preference values ​​in the target preference value queue as second target resource tags; obtain the second target relevance between each candidate push resource and each second target resource tag; and determine at least one candidate push resource corresponding to each second target resource tag as a target push resource based on the second target relevance.

[0111] Furthermore, the second acquisition module 802 is also used to: acquire a first resource tag library corresponding to the first historical push resources used by the target object; acquire a first cumulative count corresponding to each resource tag and a second cumulative count corresponding to all resource tags according to the first resource tag library; and for any resource tag, use the quotient of the first cumulative count and the second cumulative count as the TF value of the resource tag.

[0112] Furthermore, the second acquisition module 802 is also used to: acquire the usage data of the third historical push resources corresponding to the old users of the second resource tag library that have been used by all historical objects; acquire the third cumulative number of times corresponding to each resource tag and the fourth cumulative number of times corresponding to all resource tags based on the third historical push resource usage data of the second resource tag library; and acquire the IDF value corresponding to any resource tag based on the third cumulative number of times and the fourth cumulative number of times for any resource tag.

[0113] Figure 9 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Figure 9 As shown, the electronic device may include: a transceiver 91, a processor 92, and a memory 93.

[0114] Processor 92 executes computer execution instructions stored in memory, causing processor 92 to perform the scheme in the above embodiments. Processor 92 can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.

[0115] The memory 93 is connected to the processor 92 via the system bus and completes communication between them. The memory 93 is used to store computer program instructions.

[0116] The transceiver 91 can be used to determine the target push resource from the candidate push resources according to the target preference value, and push it to the terminal device of the target object.

[0117] The system bus can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. The system bus can be divided into address bus, data bus, control bus, etc. For ease of representation, only one thick line is used in the diagram, but this does not indicate that there is only one bus or one type of bus. Transceivers are used to enable communication between database access devices and other computers (e.g., clients, read-write libraries, and read-only libraries). Memory may include random access memory (RAM) and may also include non-volatile memory.

[0118] The electronic device provided in this application embodiment can be the terminal device described in the above embodiments.

[0119] This application also provides a chip for executing instructions, which is used to execute the resource push method technical solution described above.

[0120] This application also provides a computer-readable storage medium storing computer instructions. When the computer instructions are executed on a computer, the computer performs the resource push method described in the above embodiments.

[0121] This application also provides a computer program product, which includes a computer program stored in a computer-readable storage medium. At least one processor can read the computer program from the computer-readable storage medium. When the at least one processor executes the computer program, it can implement the resource push method technical solution in the above embodiments.

[0122] Other embodiments of this application will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of this application that follow the general principles of this application and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of this application are indicated by the following claims.

[0123] It should be understood that this application is not limited to the precise structure described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this application is limited only by the appended claims.

Claims

1. A method for pushing resources, characterized in that, include: Based on the usage rate of the first historical push resources used by the target object and the relevance between the first historical push resources and each resource tag, the preference value of the target object for each resource tag is obtained. The usage rate of the first historical push resource refers to the quotient of the number of times the first historical push resource is used and the total number of times the first historical push resource is pushed to the target object. The method for calculating the preference value of the target object for any resource tag is as follows: for any first historical push resource used by the target object, the product of the usage rate of the first historical push resource and the relevance between the first historical push resource and the resource tag is obtained, and the sum of the products corresponding to all the first historical push resources used by the target object is calculated. The sum of the relevance between each first historical push resource used by the target object and the resource tag is obtained, and the sum of the products is divided by the sum of the relevance to obtain the preference value of the target object for the resource tag. The term frequency (TF) value of each resource tag is determined according to the first resource tag library corresponding to the first historical push resource, and the inverse text frequency index (IDF) value of each resource tag is determined according to the second resource tag library corresponding to all the second historical push resources used by the historical object. Based on the preference value, TF value, and IDF value of each resource tag, the target preference value of the target object for each resource tag is obtained. The target preference value of the target object for any resource tag is equal to the product of the target object's preference value for that resource tag, the TF value of that resource tag, and the IDF value of that resource tag. Based on the target preference value, a target push resource is determined from the candidate push resources and pushed to the target object's terminal device.

2. The method according to claim 1, characterized in that, The method further includes: Obtain the identification information of the target object, and determine whether the target object is a new object based on the identification information; In response to the target object being a new object, obtain the target group to which the target object belongs; Obtain historical push resource usage data of the target group, determine the target push resource corresponding to the target object based on the historical push resource usage data, and push it to the target object's terminal device.

3. The method according to claim 2, characterized in that, The step of obtaining the target group to which the target object belongs includes: Obtain the first user characteristics of the target object and the second user characteristics of the existing user; Obtain the similarity between the first user feature and each of the second user features, and based on the similarity, determine the old user with the highest similarity to the target object as the target old user; The user group to which the target old user belongs is identified as the target group.

4. The method according to claim 1, characterized in that, The step of determining the target push resource from the candidate push resources based on the target preference value includes: Obtain the maximum target preference value among all the target preference values, and use the resource tag corresponding to the maximum target preference value as the first target resource tag; Obtain the first target relevance between each of the candidate push resources and the first target resource tag; The candidate push resources are sorted according to the relevance of the first target, and a candidate push resource queue is obtained after sorting. The target push resource is determined based on the candidate push resource queue.

5. The method according to claim 1, characterized in that, The step of determining the target push resource from the candidate push resources based on the target preference value includes: Sort all the target preference values ​​in descending order and obtain the sorted target preference value queue. The resource tags corresponding to the first M target preference values ​​in the target preference value queue are used as the second target resource tags; Obtain the second target relevance between each of the candidate push resources and each of the second target resource tags; Based on the second target relevance, at least one candidate push resource corresponding to each of the second target resource tags is determined as the target push resource.

6. The method according to any one of claims 1-5, characterized in that, The step of determining the term frequency (TF) value of each resource tag based on the first resource tag library corresponding to the first historical push resource includes: Obtain the first resource tag library corresponding to the first historical push resource already used by the target object; Based on the first resource tag library, obtain the first cumulative count corresponding to each resource tag and the second cumulative count corresponding to all resource tags; For any of the resource tags, the quotient of the first cumulative count and the second cumulative count is taken as the TF value of the resource tag.

7. The method according to claim 3, characterized in that, The step of determining the inverse text frequency index (IDF) value of each resource tag based on the second resource tag library corresponding to the second historical push resources used by all historical objects includes: Obtain the second resource tag library corresponding to the second historical push resources used by all the aforementioned historical objects; Based on the second resource tag library, obtain the third cumulative count corresponding to each resource tag and the fourth cumulative count corresponding to all resource tags; For any of the resource tags, the IDF value corresponding to the resource tag is obtained based on the third cumulative count and the fourth cumulative count.

8. A resource delivery device, characterized in that, include: The first acquisition module is used to acquire the target object's preference value for each resource tag based on the usage rate of the first historical push resource used by the target object and the relevance between the first historical push resource and each resource tag. The usage rate of the first historical push resource refers to the quotient of the number of times the first historical push resource is used and the total number of times the first historical push resource is pushed to the target object. The method for calculating the target object's preference value for any resource tag is as follows: for any first historical push resource used by the target object, the product of the target object's usage rate of the first historical push resource and the relevance between the first historical push resource and the resource tag is acquired, and the sum of the products corresponding to all the first historical push resources used by the target object is calculated. The sum of the relevance between each first historical push resource used by the target object and the resource tag is acquired, and the sum of the products is divided by the sum of the relevance to obtain the target object's preference value for the resource tag. The second acquisition module is used to determine the term frequency (TF) value of each resource tag according to the first resource tag library corresponding to the first historical push resource, and to determine the inverse text frequency index (IDF) value of each resource tag according to the second resource tag library corresponding to the second historical push resource used by all historical objects. The third acquisition module is used to acquire the target preference value of the target object for each resource tag based on the preference value, TF value and IDF value of each resource tag. The target preference value of the target object for any resource tag is equal to the product of the target object's preference value for that resource tag, the TF value of that resource tag and the IDF value of that resource tag. The push module is used to determine the target push resource from the candidate push resources according to the target preference value, and push it to the terminal device of the target object.

9. The apparatus according to claim 8, characterized in that, The device further includes: The judgment module is used to obtain the identification information of the target object and determine whether the target object is a new object based on the identification information; The fourth acquisition module is used to acquire the target group to which the target object belongs in response to the target object being a new object; The determining module is also used to obtain historical push resource usage data of the target group, and based on the historical push resource usage data, determine the target push resource corresponding to the target object, and push it to the target object's terminal device.

10. The apparatus according to claim 9, characterized in that, The fourth acquisition module is also used for: Obtain the first user characteristics of the target object and the second user characteristics of the existing user; Obtain the similarity between the first user feature and each of the second user features, and based on the similarity, determine the old user with the highest similarity to the target object as the target old user; The user group to which the target old user belongs is identified as the target group.

11. The apparatus according to claim 8, characterized in that, The push module is also used for: Obtain the maximum target preference value among all the target preference values, and use the resource tag corresponding to the maximum target preference value as the first target resource tag; Obtain the first target relevance between each of the candidate push resources and the first target resource tag; The candidate push resources are sorted according to the relevance of the first target, and a candidate push resource queue is obtained after sorting. The target push resource is determined based on the candidate push resource queue.

12. The apparatus according to claim 8, characterized in that, The push module is also used for: Sort all the target preference values ​​in descending order and obtain the sorted target preference value queue. The resource tags corresponding to the first M target preference values ​​in the target preference value queue are used as the second target resource tags; Obtain the second target relevance between each of the candidate push resources and each of the second target resource tags; Based on the second target relevance, at least one candidate push resource corresponding to each of the second target resource tags is determined as the target push resource.

13. The apparatus according to any one of claims 8-12, characterized in that, The second acquisition module is further configured to: Obtain the first resource tag library corresponding to the first historical push resource already used by the target object; Based on the first resource tag library, obtain the first cumulative count corresponding to each resource tag and the second cumulative count corresponding to all resource tags; For any of the resource tags, the quotient of the first cumulative count and the second cumulative count is taken as the TF value of the resource tag.

14. The apparatus according to claim 10, characterized in that, The second acquisition module is further configured to: Obtain the second resource tag library corresponding to the second historical push resources used by all the aforementioned historical objects; Based on the second resource tag library, obtain the third cumulative count corresponding to each resource tag and the fourth cumulative count corresponding to all resource tags; For any of the resource tags, the IDF value corresponding to the resource tag is obtained based on the third cumulative count and the fourth cumulative count.

15. An electronic device, characterized in that, include: A processor, and a memory communicatively connected to the processor; The memory stores computer-executed instructions; The processor executes computer execution instructions stored in the memory to implement the method as described in any one of claims 1-7.

16. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions, which, when executed by a processor, are used to implement the method as described in any one of claims 1-7.

17. A computer program product, characterized in that, Includes a computer program that, when executed by a processor, implements the method of any one of claims 1-7.

Images