Optimization method and device for improving display revenue per unit area of recommendation system
By calculating the unit area income in the recommendation system and introducing the unit area income factor, the problem of profit deviation caused by the area bias in the recommendation system is solved, the accuracy and efficiency of the recommendation system are improved, and the average time of the resource page is improved.
Patent Information
- Application Number
- CN202310528471.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2023-05-09
- Publication Date
- 2024-12-10
- Estimated Expiration
- 2043-05-09
AI Technical Summary
The existing recommendation system does not consider the cost of resource display area when calculating click-through rate and duration estimates, resulting in a deviation in measuring returns at the same cost.
The click-through rate estimate module, duration estimate module and presentation page pre-generating module obtain the click-through rate, duration and presentation height of the resource, calculate the unit area income, and obtain the unit area income factor based on this, which acts on the click-through rate and duration estimates Module to reduce the bias caused by the display height.
It improves the revenue accuracy of the recommendation system, improves the display efficiency of the recommendation result list page and the average time of the resource page, and thus improves the total media time of the recommendation system.
Smart Images
Figure CN116562920B_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the field of big data technology, and specifically relates to an optimization method and device for improving the display revenue per unit area of a recommendation system. Background Art
[0002] Recommended products are very popular in the market, such as Douyin, Xiaohongshu, Kuaishou, Toutiao, etc. In the list page of recommended results, resources with larger display areas are more likely to attract users' attention, resulting in higher post-click rate and duration. Generally, the display width on the same mobile terminal is basically the same, so the display height of the resource can be used to refer to the display area.
[0003] In the related technologies, the click-through rate estimation and duration estimation models currently used do not take into account the cost required for the resource display area. If the cost required for the resource display area is not taken into account, it will bring display area bias to the click-through rate estimation, duration estimation and other modules of the recommendation system, resulting in deviations in measuring benefits at the same cost. Summary of the invention
[0004] In view of this, the purpose of the present invention is to overcome the shortcomings of the prior art and provide an optimization method and device for improving the display revenue per unit area of a recommendation system, so as to solve the problem in the prior art that the display area bias leads to a deviation in measuring revenue at the same cost.
[0005] To achieve the above objectives, the present invention adopts the following technical solution: an optimization method for improving the display revenue per unit area of a recommendation system, applied to the recommendation system, the system comprising a click rate estimation module, a duration estimation module and a display page pre-generation module; the method comprises:
[0006] Obtaining a click rate estimate of a resource through the click rate estimate module, obtaining a duration estimate of a resource through the duration estimate module, and obtaining a resource display height through the display page pre-generation module;
[0007] Calculate the unit area revenue of the resource based on the estimated click rate, estimated duration, and resource display height;
[0008] A unit area revenue factor is obtained based on the unit area revenue, and the unit area revenue factor is applied to a click rate estimation module and a duration estimation module.
[0009] Furthermore, the resource unit area revenue is calculated according to the click rate estimate, duration estimate and resource display height in the following manner:
[0010]
[0011] in, ctrrepresents the estimated click-through rate. dur It represents the estimated time after a click occurs. template_ ht Indicates the resource display height.
[0012] Furthermore, the unit area benefit factor is obtained based on the unit area benefit of the resource in the following manner:
[0013]
[0014] in, area represents the revenue per unit area, area avg represents the mean income per unit area, k , w Both represent coefficients used to adjust the intensity of unit area revenue.
[0015] Furthermore, a multiplication fusion algorithm is used to apply the unit area revenue factor to the click rate estimation module and the duration estimation module.
[0016] Furthermore, the unit area benefit factor has a value range of [0.5, 1.5].
[0017] The embodiment of the present application provides an optimization device for improving the display revenue per unit area of a recommendation system, which is applied to the recommendation system. The system includes a click rate estimation module, a duration estimation module, and a display page pre-generation module; the device includes:
[0018] An acquisition module, used to acquire a click rate estimate of a resource through the click rate estimation module, acquire a duration estimate of a resource through the duration estimation module, and acquire a resource display height through a display page pre-generation module;
[0019] A calculation module, used to calculate the unit area revenue of the resource according to the estimated click rate, the estimated duration and the resource display height;
[0020] The action module is used to obtain a unit area revenue factor based on the unit area revenue, and to apply the unit area revenue factor to a click rate estimation module and a duration estimation module.
[0021] An embodiment of the present application provides a computer device, including: a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor executes any step of the above-mentioned optimization method for improving the display revenue per unit area of the recommendation system.
[0022] An embodiment of the present application also provides a computer storage medium storing a computer program. When the computer program is executed by a processor, the processor executes any step of the above-mentioned optimization method for improving the display revenue per unit area of the recommendation system.
[0023] The present invention adopts the above technical solution, and the beneficial effects that can be achieved include:
[0024] The present invention provides an optimization method and device for improving the display revenue per unit area of a recommendation system. The application obtains the click rate estimate, duration estimate and resource display height of the resource through the recommendation system, thereby calculating the unit area revenue of the resource. The recommendation system revenue factor can be obtained through the unit area revenue. Applying the unit area revenue factor to the recommendation system can help reduce the bias caused by the display height and improve the accuracy of measuring the revenue. In addition, the unit area revenue factor can also improve the display efficiency of the recommendation result list page, increase the average duration of the resource page, increase the average reading time per person, and thus increase the total media time of the recommendation system. BRIEF DESCRIPTION OF THE DRAWINGS
[0025] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0026] Figure 1 A schematic diagram of the steps of the optimization method for improving the display revenue per unit area of the recommendation system of the present invention;
[0027] Figure 2 A schematic diagram of the structure of the recommendation system provided by the present invention;
[0028] Figure 3 A schematic diagram of the structure of an optimization device for improving the display revenue per unit area of a recommendation system according to the present invention;
[0029] Figure 4 A schematic diagram of the structure of a computer device involved in the optimization method for improving the display revenue per unit area of a recommendation system of the present invention. DETAILED DESCRIPTION
[0030] To make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be described in detail below. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other implementation methods obtained by ordinary technicians in this field without creative work belong to the scope of protection of the present invention.
[0031] First, this application conducted a survey on the popular product forms in the current market and found that in the list page of recommendation results, resources with larger display areas are more likely to attract users' attention, resulting in relatively higher a posteriori click-through rate and duration. The relevant results are shown in the posteriori statistics of the recommendation verification system in Table 1.
[0032] Table 1. Post-hoc click rate and duration of resources at different display heights of the recommendation verification system
[0033]
[0034] Table 1 proves that resources with larger display areas are more attractive to users, resulting in higher a posteriori click-through rate and duration.
[0035] A specific optimization method and device for improving the display revenue per unit area of a recommendation system provided in an embodiment of the present application is described below in conjunction with the accompanying drawings.
[0036] like Figure 1 As shown, the optimization method for improving the display revenue per unit area of the recommendation system provided in the embodiment of the present application is applied to the recommendation system, and the system includes a click rate estimation module, a duration estimation module and a display page pre-generation module;
[0037] First of all, it should be noted that Figure 2 As shown, the recommendation system provided by the present application includes a content model, a recommendation model, a scenario model, a user model and a display page pre-generation module; the technical solution provided by the present application is carried out for the modules in the recommendation model, and the recommendation model includes a click-through rate prediction module and a duration prediction module.
[0038] The method comprises:
[0039] S101, obtaining a click rate estimation value of a resource through the click rate estimation module, obtaining a duration estimation value of a resource through the duration estimation module, and obtaining a resource display height through a display page pre-generation module;
[0040] It should be noted that in this application, the template height of the resource on the list page is used to approximate the resource display height. The above-mentioned click rate estimate, duration estimate and resource display height can all be directly obtained through the recommendation system.
[0041] It should be noted that in this application, the resource display height is measured by the template height of the resource on the list page. The above-mentioned click rate estimate, duration estimate and resource display height can all be directly obtained through the recommendation system.
[0042] S102, calculating the unit area revenue of the resource according to the estimated click rate, the estimated duration, and the resource display height;
[0043] It can be understood that the click rate estimate and duration estimate of the resources defined in this application simulate the unit area revenue, and the unit area revenue of each resource is defined as the ratio of the landing page duration obtained by the resource to the display height of its list page.
[0044] In some embodiments, the resource unit area revenue is calculated according to the click rate estimate, duration estimate and resource display height in the following manner:
[0045]
[0046] in, ctr represents the estimated click-through rate. dur It represents the estimated time after a click occurs. template_ ht Indicates the resource display height.
[0047] S103, obtaining a unit area revenue factor based on the unit area revenue, and applying the unit area revenue factor to a click rate estimation module and a duration estimation module.
[0048] After obtaining the unit area benefit, the unit area benefit factor is obtained based on the resource unit area benefit in the following manner:
[0049]
[0050] in, area represents the revenue per unit area, area avg represents the mean income per unit area, k , w Both represent coefficients used to adjust the intensity of unit area revenue.
[0051] It should be noted that the k , w The value of is obtained by performing multiple groups of small traffic experiments in the simulation online environment and taking the group of values with the largest total media duration of the recommendation system. The unit area revenue factor has a value range of [0.5, 1.5].
[0052] In the present application, a multiplication fusion algorithm is used to apply the unit area revenue factor to the click rate estimation module and the duration estimation module.
[0053] As a specific implementation method, the effect of the unit area revenue factor on the recommendation system is verified in this application, and the display proportion and posterior duration changes of resources with different display heights are obtained, as shown in Table 2.
[0054] Table 2 Changes in display ratio and posterior duration of resources at different display heights in the recommendation verification system
[0055]
[0056] From Table 2, we can see that compared with not considering the unit area benefit factor, the display ratio of resources with a display height between 100-350 increased by +0.15%, and the a posteriori duration increased by +1.79%; while the display ratio of resources with a display height between 650-800 decreased by -0.34%, and the a posteriori duration remained basically the same (-0.05%). This shows that unit area benefit helps to reduce the bias caused by the display height, promotes resources with a smaller display height and a higher a posteriori duration, and ultimately drives the average duration of resource pages to increase by +0.54%.
[0057] The working principle of the optimization method for improving the unit area display revenue of the recommendation system is as follows: obtain the click rate estimate of the resource through the click rate estimation module in the recommendation system, obtain the duration estimate of the resource through the duration estimation module, and obtain the resource display height through the display page pre-generation module; then calculate the unit area revenue of the resource based on the click rate estimate, duration estimate and resource display height, obtain the unit area revenue factor based on the unit area revenue, and apply the unit area revenue factor to the click rate estimation module and the duration estimation module.
[0058] This application introduces a unit area revenue factor and uses the display height of each resource to normalize the landing page duration it obtains, thereby reducing the bias caused by the resource display height, improving display efficiency, promoting an increase in the average duration of resource pages and the average reading time per person, and thereby driving an increase in the total media duration of the recommendation system.
[0059] like Figure 3 As shown, the embodiment of the present application provides an optimization device for improving the display revenue per unit area of a recommendation system, which is applied to the recommendation system. The system includes a click rate estimation module, a duration estimation module, and a display page pre-generation module; the device includes:
[0060] An acquisition module 201 is used to acquire a click rate estimate of a resource through the click rate estimation module, acquire a duration estimate of a resource through the duration estimation module, and acquire a resource display height through a display page pre-generation module;
[0061] A calculation module 202, configured to calculate a unit area benefit of a resource according to the estimated click rate, the estimated duration, and the resource display height;
[0062] The action module 203 is used to obtain a unit area revenue factor based on the unit area revenue, and apply the unit area revenue factor to the click rate estimation module and the duration estimation module.
[0063] The working principle of the optimization device for improving the unit area display revenue of the recommendation system provided in the embodiment of the present application is as follows: the acquisition module 201 obtains the click rate estimate value of the resource through the click rate estimation module, obtains the duration estimate value of the resource through the duration estimation module, and obtains the resource display height through the display page pre-generation module; the calculation module 202 calculates the unit area revenue of the resource according to the click rate estimate value, the duration estimate value and the resource display height; the action module 203 obtains the unit area revenue factor based on the unit area revenue, and applies the unit area revenue factor to the click rate estimation module and the duration estimation module.
[0064] The present application provides a computer device, including: a memory 1 and a processor 2, and may also include a network interface 3. The memory 1 stores a computer program, and the memory 1 may include a non-permanent memory in a computer-readable medium, a random access memory (RAM) and / or a non-volatile memory, such as a read-only memory (ROM) or a flash memory (flash RAM). The computer device stores an operating system 4, and the memory is an example of a computer-readable medium. When the computer program is executed by the processor, the processor executes an optimization method for improving the display revenue per unit area of the recommendation system. Figure 4 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.
[0065] In one embodiment, the optimization method for improving the display revenue per unit area of the recommendation system provided in the present application can be implemented in the form of a computer program. The computer program can be used in Figure 4 Runs on the computer device shown.
[0066] In some embodiments, when the computer program is executed by the processor, the processor performs the following steps: obtaining a click rate estimate value of the resource through the click rate estimation module, obtaining a duration estimate value of the resource through the duration estimation module, and obtaining a resource display height through the display page pre-generation module; calculating the unit area revenue of the resource based on the click rate estimate value, duration estimate value, and resource display height; obtaining a unit area revenue factor based on the unit area revenue, and applying the unit area revenue factor to the click rate estimation module and the duration estimation module.
[0067] The present application also provides a computer storage medium, examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk-read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassette storage or other magnetic storage devices or any other non-transmission medium that can be used to store information that can be accessed by a computing device.
[0068] In some embodiments, the present invention further proposes a computer-readable storage medium storing a computer program, which, when executed by a processor, obtains a click rate estimate value of a resource through the click rate estimation module, obtains a duration estimate value of the resource through the duration estimation module, and obtains a resource display height through a display page pre-generation module; calculates the unit area revenue of the resource based on the click rate estimate value, duration estimate value, and resource display height; obtains a unit area revenue factor based on the unit area revenue, and applies the unit area revenue factor to the click rate estimation module and the duration estimation module.
[0069] In summary, the present invention provides an optimization method and device for improving the display revenue per unit area of a recommendation system, the method comprising obtaining a click rate estimate value of a resource through a click rate estimation module, obtaining a duration estimate value of a resource through a duration estimation module, and obtaining a resource display height through a display page pre-generation module; calculating the unit area revenue of the resource according to the click rate estimate value, the duration estimate value, and the resource display height; obtaining a unit area revenue factor based on the unit area revenue, and applying the unit area revenue factor to the click rate estimation module and the duration estimation module. The present invention can effectively reduce and eliminate the display area bias and improve the accuracy of measuring revenue by introducing a unit area revenue factor and applying the unit area revenue factor to the recommendation system; in addition, it can also improve the display efficiency of the recommendation result list page, increase the average duration of the resource page, increase the average reading time per person, and thereby increase the total media time of the recommendation system.
[0070] It can be understood that the method embodiment provided above corresponds to the device embodiment described above, and the corresponding specific contents can be referenced to each other and will not be repeated here.
[0071] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present application may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage and optical storage, etc.) containing computer-usable program codes.
[0072] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.
[0073] These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer-readable memory produce a manufactured product including an instruction method, which is implemented in the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.
[0074] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process in the computer or other programmable device. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.
[0075] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.
Claims
1. An optimization method for improving the display revenue per unit area of a recommendation system, characterized in that: Applied to a recommendation system, the system includes a click rate estimation module, a duration estimation module and a display page pre-generation module; the method includes: Obtaining a click rate estimate of a resource through the click rate estimate module, obtaining a duration estimate of a resource through the duration estimate module, and obtaining a resource display height through the display page pre-generation module; Calculate the unit area revenue of the resource based on the estimated click rate, estimated duration, and resource display height; A unit area revenue factor is obtained based on the unit area revenue, and the unit area revenue factor is applied to a click rate estimation module and a duration estimation module; The resource unit area revenue is calculated according to the estimated click rate, estimated duration and resource display height in the following manner: in, ctr represents the estimated click-through rate. dur It represents the estimated time after a click occurs. template_ht Indicates resource display height; The unit area benefit factor is obtained based on the unit area benefit of the resource in the following manner: in, area represents the revenue per unit area, area avg represents the mean income per unit area, k , w Both represent coefficients used to adjust the intensity of unit area revenue.
2. The method according to claim 1, characterized in that The multiplication fusion algorithm is used to apply the unit area revenue factor to the click rate estimation module and the duration estimation module.
3. The method according to claim 1, characterized in that The resource display height is obtained by the template height of the resource on the list page.
4. The method according to claim 1, characterized in that: The unit area profit factor has a value range of [0.5, 1.5].
5. An optimization device for improving the display revenue per unit area of a recommendation system, characterized in that: Applied to a recommendation system, the system includes a click rate estimation module, a duration estimation module and a display page pre-generation module; the device includes: An acquisition module, used to acquire a click rate estimate of a resource through the click rate estimation module, acquire a duration estimate of a resource through the duration estimation module, and acquire a resource display height through a display page pre-generation module; A calculation module is used to calculate the unit area income of resources according to the click rate estimate, the duration estimate and the resource display height; the unit area income of resources is calculated according to the click rate estimate, the duration estimate and the resource display height in the following manner: in, ctr represents the estimated click-through rate. dur It represents the estimated time after a click occurs. template_ht Indicates resource display height; An action module is used to obtain a unit area profit factor based on the unit area profit, and to apply the unit area profit factor to a click rate estimation module and a duration estimation module; the unit area profit factor is obtained based on the unit area profit of the resource in the following manner: in, area represents the revenue per unit area, area avg represents the mean income per unit area, k , w Both represent coefficients used to adjust the intensity of unit area revenue.
6. A computer device, characterized in that: include: A memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the steps of the optimization method for improving the display benefit per unit area of the recommendation system as described in any one of claims 1 to 4.
7. A computer storage medium, characterized in that: A computer program is stored, and when the computer program is executed by a processor, the processor executes the steps of the optimization method for improving the display revenue per unit area of a recommendation system as described in any one of claims 1 to 4.