Information processing method and device for advertising participation
By filtering effective user data and utilizing conversion rate and user profile models, combined with a bid prediction model, the problem of automated differentiated bidding in advertising competition was solved, improving conversion efficiency and reducing costs.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- SHANGHAI SHUHE INFORMATION TECH CO LTD
- Filing Date
- 2022-06-28
- Publication Date
- 2026-06-05
Smart Images
Figure CN114997932B_ABST
Abstract
Description
Technical Field
[0001] This disclosure relates to the field of data processing technology, specifically to an information processing method and apparatus for advertising competition. Background Technology
[0002] RTA, or Real-Time API, is a real-time advertising bidding process from the user's perspective. Users model their target audience and then bid based on that audience when ad requests reach the media.
[0003] In related technologies, it is not possible to automatically perform differentiated bidding. Summary of the Invention
[0004] The main purpose of this disclosure is to provide an information processing method and apparatus for advertising competition.
[0005] To achieve the above objectives, according to a first aspect of this disclosure, an information processing method for advertising bidding is provided, comprising: after acquiring user data in the current traffic, filtering the user data to determine valid user data; predicting the conversion rate corresponding to the valid user data after advertising placement based on a pre-established conversion rate prediction model; predicting the user profile of the valid user data through a user profile prediction model; and determining the bidding price corresponding to the current user data using a bid estimation model based on the predicted conversion rate and / or the user profile.
[0006] Optionally, filtering the user data to determine valid user data includes: removing user data that does not meet preset qualifications; and removing user data that does not meet conversion rules at preset nodes in the advertising delivery business.
[0007] Optionally, the method may further include, before determining the bidding price, determining whether to bid on user data within the current traffic.
[0008] Optionally, determining whether to participate in the bidding for user data in the current traffic includes: using a pre-established prediction model to predict the risk level of a user having multiple businesses for each valid user data; and based on the results of the above prediction, determining whether to participate in the advertising campaign for the current traffic.
[0009] Optionally, determining whether to run ads for the current traffic based on the above prediction results includes: determining whether to run ads for the current traffic based on the above prediction results and the business details corresponding to each user provided by a third party.
[0010] Optionally, determining the bidding price corresponding to the current user data based on the predicted conversion rate and / or the user profile, using a bidding prediction model, includes: determining a bidding strategy based on the predicted conversion rate and / or the user profile; determining the value of the effective user data using a bidding prediction model; and determining the bidding bid corresponding to the effective user data based on the bidding strategy and the user value.
[0011] According to a second aspect of this disclosure, an information processing apparatus for advertising bidding is provided, comprising: a data filtering unit configured to filter user data in current traffic after acquiring user data to determine valid user data; a conversion rate prediction unit configured to predict the conversion rate corresponding to the valid user data after advertising placement based on a pre-established conversion rate prediction model; a profile prediction unit configured to predict the user profile of the valid user data using a user qualification profile prediction model; and a bidding unit configured to determine the bidding price corresponding to the current user data based on the predicted conversion rate and / or the user profile using a bidding estimation model.
[0012] Optionally, filtering the user data to determine valid user data includes: removing user data that does not meet preset qualifications; and removing user data that does not meet conversion rules at preset nodes in the advertising delivery business.
[0013] According to a third aspect of this disclosure, a computer-readable storage medium is provided storing computer instructions for causing the computer to perform the information processing method for advertising competition as described in any implementation of the first aspect.
[0014] According to a fourth aspect of this disclosure, an electronic device is provided, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores a computer program executable by the at least one processor, the computer program being executed by the at least one processor to cause the at least one processor to perform the information processing method for advertising competition as described in any implementation of the first aspect.
[0015] The information processing method and apparatus for advertising bidding in this disclosure includes: after acquiring user data from the current traffic, filtering the user data to determine valid user data; predicting the conversion rate corresponding to the valid user data after advertising placement based on a pre-established conversion rate prediction model; predicting the user profile of the valid user data using a user profile prediction model; and determining the bidding price corresponding to the current user data using a bid estimation model based on the predicted conversion rate and / or the user profile. This achieves real-time automated bidding and pricing for different user data based on their corresponding conversion rates and user profiles, overcoming the technical problem of not being able to achieve automated RTA bidding in related technologies. Attached Figure Description
[0016] To more clearly illustrate the technical solutions in the specific embodiments of this disclosure or the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this disclosure. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.
[0017] Figure 1 This is a flowchart of an information processing method for advertising competition according to an embodiment of the present disclosure;
[0018] Figure 2 This is an application diagram of an information processing method for advertising competition according to an embodiment of the present disclosure;
[0019] Figure 3 This is a schematic diagram illustrating another application of the information processing method for advertising competition according to embodiments of the present disclosure.
[0020] Figure 4 This is a schematic diagram of an electronic device according to an embodiment of the present disclosure. Detailed Implementation
[0021] To enable those skilled in the art to better understand the present disclosure, the technical solutions of the present disclosure will be clearly and completely described below with reference to the accompanying drawings of the embodiments. Obviously, the described embodiments are only some embodiments of the present disclosure, and not all embodiments. Based on the embodiments of the present disclosure, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present disclosure.
[0022] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this disclosure are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate for the embodiments of this disclosure described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.
[0023] It should be noted that, unless otherwise specified, the embodiments and features described in this disclosure can be combined with each other. This disclosure will now be described in detail with reference to the accompanying drawings and embodiments.
[0024] According to embodiments of this disclosure, an information processing method for advertising competition is provided, such as... Figure 1 As shown, the method includes the following steps 101 to 104:
[0025] Step 101: After obtaining the user data in the current traffic, filter the user data to determine the valid user data.
[0026] As an optional implementation of this embodiment, filtering the user data to determine valid user data includes: removing user data that does not meet preset qualifications; and removing user data that does not meet conversion rules at preset nodes in the advertising delivery business.
[0027] In this embodiment, after acquiring traffic, user data can be filtered. The filtering method may include removing user data that does not match preset qualification information. This preset qualification information may include the business scope based on the product outline indicated in the targeted advertisement. Based on this, data from non-target customer groups, user data from dynamic blacklists, registered users, and users from the overdue list shared by the Internet Finance Association can be excluded. In addition, user data that does not meet preset conversion rules at preset nodes of the targeted business is also filtered out. This rule could be N exposures but M non-conversions, where M / N does not meet the preset value. This type of user data is filtered to exclude invalid multiple exposures. It is understood that the preset node can be an exposure node. Through this embodiment, a series of users not acquired in this instance can be directly filtered out, preventing them from entering the information flow customer acquisition process.
[0028] Step 102: Based on the pre-established conversion rate prediction model, predict the conversion rate corresponding to the effective user data after the ad campaign.
[0029] In this embodiment, the advertising delivery business based on RTA can be divided into multiple nodes, including but not limited to click nodes, registration nodes, first login nodes, business application nodes (e.g., loan application nodes), nodes where the application is implemented (e.g., the node where the loan is processed after the loan application is completed), evaluation nodes (e.g., credit limit evaluation nodes), usage nodes (e.g., credit limit utilization rate), and charging nodes (e.g., CPS, CPA, CPCL, CPC, or CPM, etc.). This embodiment can pre-establish a conversion rate prediction model, which can predict the conversion rate of each node based on user data.
[0030] A further conversion rate prediction model can be pre-trained. During training, the actual results of each conversion node are used as the target variable, and the relevant information of the user ID (user data) sent in the information flow is used as the dependent variable. A multi-objective machine learning model is used to learn the characteristics of users with high conversion efficiency across the entire process. This multi-objective model is then invoked to make predictions during RTA (Real-Time Acquisition) bidding requests, obtaining the conversion probability of the sent user across the entire process. This embodiment can optimize the conversion efficiency of the entire user journey by applying gradient machine learning algorithms to construct a multi-objective optimization machine learning model.
[0031] In related technologies, since everyone wants to acquire high-quality users, they increase their bids in information flow advertising to enhance their bargaining power and thus obtain better users. However, the cost increases significantly. This embodiment improves the conversion efficiency of the entire link through a conversion rate prediction model, thereby achieving the goal of reducing CPA and CPS.
[0032] Furthermore, since each node has a corresponding conversion efficiency, in order to ensure the overall conversion efficiency is improved, a multi-objective optimization approach can be used to optimize the model, thereby ensuring the highest possible conversion rate across the entire chain. (Reference) Figure 2 , Figure 2 A schematic diagram of multi-objective optimization is shown.
[0033] Step 103: Predict the user profile of the valid user data using the user qualification profile prediction model.
[0034] In this implementation, the predictive model can yield user value, which can be represented numerically. This value assessment can be based on user data such as credit limit, total transaction amount, number of transactions, and credit limit utilization rate. The predictive model can be trained. During training, user ID information can be used as the dependent variable (label), and user data such as credit limit, total transaction amount, number of transactions, and credit limit utilization rate can be used as the target variables. The dependent variable can include device information, tags processed from the APP list, internal historical information flow conversion information, and external third-party information.
[0035] This embodiment uses user characteristics learned through a machine learning model to make predictions when submitting an RTA bidding request, obtaining an estimated value for the issued user BID. This estimated value can then be used to guide bidding decisions.
[0036] Step 104: Based on the predicted conversion rate and / or the user profile, use the bid prediction model to determine the bidding price corresponding to the current user data.
[0037] As an optional implementation of this embodiment, determining the bidding price corresponding to the current user data based on the predicted conversion rate and / or the user profile using a bidding prediction model includes: determining a bidding strategy based on the predicted conversion rate and / or the user profile; determining the value of the effective user data using a bidding prediction model; and determining the bidding bid corresponding to the effective user data based on the bidding strategy and the user value.
[0038] In this embodiment, bidding can be based on conversion rate or user profile. Different bidding bases can have different bidding strategies, including strong participation strategy, strong non-participation strategy and other strategies. Different bidding strategies can have different bid coefficient values, which can be used to calculate the bid value.
[0039] In this embodiment, based on the ECPM and OCPC delivery models, the mathematical principles behind the participation and differentiated bidding methods based on RTA end-to-end conversion efficiency improvement, user profiling, and value estimation are as follows:
[0040] max x ECPM = ∑PCTR * PCVR * bid
[0041] max∑x i *PCTR*PCVR
[0042] st∑x i *wp i ≤B; where X is above. i W represents the probability of a successful bid for the i-th exposure; pi B is the winning bid for the i-th exposure; B is the total budget for the advertising campaign.
[0043] Furthermore, the final bid is designed as u(b*), implemented as follows:
[0044]
[0045] refer to Figure 3The application scenario diagram shown illustrates that after determining the participants, the value can be estimated using a value estimation model, and the bidding coefficient can be determined based on the estimation results.
[0046] As an optional implementation of this embodiment, the method further includes, before determining the bidding price, determining whether to participate in the bidding for user data in the current traffic.
[0047] As an optional implementation of this embodiment, determining whether to participate in the bidding for user data in the current traffic includes: predicting the risk level of multiple businesses for each valid user data using a pre-established prediction model; and determining whether to participate in the advertising bidding for the current traffic based on the prediction results.
[0048] In this optional implementation, the prediction model primarily uses the number of multiple borrowing transactions obtained from historically converted users, defining three levels—high, medium, and low—as the target variable. The user ID information from the information stream is used as the dependent variable, which may include device information and tags processed from the app list, such as the number of installed loan apps. A machine learning model is used to learn the characteristics of users with low multiple borrowing rates. This model is then invoked during RTA bidding requests to predict the probability that a user has low multiple borrowing rates. In application, a user's multiple borrowing rate can be the number of loan transactions the user is simultaneously conducting (i.e., the number of channels for borrowing simultaneously; for example, if borrowing from A, B, and C simultaneously, the multiple borrowing rate is 3).
[0049] By estimating the number of users with multiple accounts, it can be determined whether to participate in the bidding for the current traffic.
[0050] As an optional implementation of this embodiment, determining whether to participate in the bidding for advertising based on the above prediction results includes: determining whether to participate in the bidding for advertising based on the above prediction results and the business details corresponding to each user provided by a third party.
[0051] In this optional implementation, refer to Figure 3 To improve the accuracy of eligibility assessment, after determining the number of users with multiple accounts, eligibility can also be assessed based on the multiple account data corresponding to each user provided by a third party. Understandably, eligibility assessment could also be directly based on the multiple account data corresponding to each user provided by a third party; however, relying solely on third-party assessment would increase advertising costs.
[0052] This optional implementation method can ensure that user data can be converted with a high probability, and can also ensure that the converted data has reduced risk based on business assumptions. For example, the risk of users who have taken out loans being unable to repay them is reduced.
[0053] This embodiment uses technologies such as RTA to automatically screen potential high-quality users, determine whether to participate in the bidding in real time, and make a series of automated decisions, enabling differentiated pricing for different users.
[0054] This embodiment achieves automated screening of valid user data, applies gradient machine learning algorithms to construct a multi-objective optimization machine learning model, optimizes the conversion efficiency of the entire user journey, and realizes differentiated bidding for different users through a Bid value prediction model, overcoming the problem that related technologies cannot automatically perform differentiated bidding.
[0055] It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although a logical order is shown in the flowchart, in some cases the steps shown or described may be executed in a different order than that shown here.
[0056] According to embodiments of this disclosure, an apparatus for implementing the above-described information processing method for advertising bidding is also provided. The apparatus includes: a data filtering unit configured to filter user data in current traffic to determine valid user data; a conversion rate prediction unit configured to predict the conversion rate corresponding to the valid user data after advertising placement based on a pre-established conversion rate prediction model; a profile prediction unit configured to predict the user profile of the valid user data using a user qualification profile prediction model; and a bidding unit configured to determine the bidding price corresponding to the current user data based on the predicted conversion rate and / or the user profile, using a bidding estimation model.
[0057] As an optional implementation of this embodiment, filtering the user data to determine valid user data includes: removing user data that does not meet preset qualifications; and removing user data that does not meet conversion rules at preset nodes in the advertising delivery business.
[0058] As an optional implementation of this embodiment, the method further includes, before determining the bidding price, determining whether to participate in the bidding for user data in the current traffic.
[0059] As an optional implementation of this embodiment, determining whether to participate in the bidding for user data in the current traffic includes: predicting the risk level of multiple businesses for each valid user data using a pre-established prediction model; and determining whether to participate in the advertising bidding for the current traffic based on the prediction results.
[0060] As an optional implementation of this embodiment, determining whether to participate in the bidding for advertising based on the above prediction results includes: determining whether to participate in the bidding for advertising based on the above prediction results and the business details corresponding to each user provided by a third party.
[0061] As an optional implementation of this embodiment, determining the bidding price corresponding to the current user data based on the predicted conversion rate and / or the user profile using a bidding prediction model includes: determining a bidding strategy based on the predicted conversion rate and / or the user profile; determining the value of the effective user data using a bidding prediction model; and determining the bidding bid corresponding to the effective user data based on the bidding strategy and the user value.
[0062] This disclosure provides an electronic device, such as... Figure 4 As shown, the electronic device includes one or more processors 41 and a memory 42. Figure 4 Take a processor 41 as an example.
[0063] The controller may also include an input device 43 and an output device 44.
[0064] The processor 41, memory 42, input device 43, and output device 44 can be connected via a bus or other means. Figure 4 Taking the example of a connection between China and Israel via a bus.
[0065] Processor 41 can be a Central Processing Unit (CPU). Processor 41 can also be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, or combinations thereof. The general-purpose processor can be a microprocessor or any conventional processor.
[0066] The memory 42, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs, non-transitory computer-executable programs, and modules, such as the program instructions / modules corresponding to the control method in this embodiment. The processor 41 executes various functional applications and data processing of the server by running the non-transitory software programs, instructions, and modules stored in the memory 42, thereby implementing the information processing method for advertising competition described in the above method embodiment.
[0067] The memory 42 may include a program storage area and a data storage area. The program storage area may store the operating system and applications required for at least one function; the data storage area may store data created by the use of the processing device operated by the server. Furthermore, the memory 42 may include high-speed random access memory and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, the memory 42 may optionally include memory remotely located relative to the processor 41, and these remote memories can be connected to a network connection device via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.
[0068] Input device 43 can receive input numerical or character information, and generate key signal inputs related to user settings and function control of the server's processing device. Output device 44 may include display devices such as a display screen.
[0069] One or more modules are stored in memory 42, and when executed by one or more processors 41, they perform actions such as... Figure 1 The method shown.
[0070] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The program can be stored in a computer-readable storage medium. When executed, the program can include the processes of the embodiments of the motor control methods described above. The storage medium can be a magnetic disk, optical disk, read-only memory (ROM), random access memory (RAM), flash memory, hard disk drive (HDD), or solid-state drive (SSD), etc.; the storage medium can also include combinations of the above types of memory.
[0071] Although embodiments of the present disclosure have been described in conjunction with the accompanying drawings, those skilled in the art can make various modifications and variations without departing from the spirit and scope of the present disclosure, and such modifications and variations all fall within the scope defined by the appended claims.
Claims
1. An information processing method for advertising competition, characterized in that, include: After obtaining user data from the current traffic, the user data is filtered to determine valid user data. The filtering of user data to determine valid user data includes: removing user data that does not meet the preset qualifications; and removing user data that does not meet the conversion rules at the preset node in the advertising business. Based on a pre-established conversion rate prediction model, the conversion rate corresponding to the effective user data after ad delivery is predicted. The ad delivery business based on RTA is divided into multiple nodes, including click node, registration node, first login node, business application node, application business implementation node, evaluation node, usage node, and payment node. The conversion rate prediction model is pre-trained. During training, the actual results of each conversion node are used as the target variable, and the relevant information of the user ID sent by the information flow is used as the dependent variable. The multi-objective machine learning model learns the user characteristics with high conversion efficiency throughout the entire process. When the RTA bidding request is made, this multi-objective model is called to make predictions and obtain the conversion probability of the user throughout the entire process. The user profile of the valid user data is predicted using a user qualification profile prediction model; Based on the predicted conversion rate and the user profile, a bid estimation model is used to determine the bidding price corresponding to the current user data. This process includes: determining a bidding strategy based on the predicted conversion rate and the user profile; determining the value of the effective user data using the bid estimation model; and determining the bid price corresponding to the effective user data based on the bidding strategy and the user value. Bidding is based on the conversion rate and user profile; different bidding bases have different bidding strategies, including strong participation strategies and strong non-participation strategies. Different bidding strategies have different bid coefficient values, which are used to calculate the bid value. The method prior to determining the bidding price also includes: determining whether to participate in the bidding for user data in the current traffic, including: using a pre-established prediction model to predict the risk level of multiple businesses for each valid user data; and based on the prediction results, determining whether to participate in the bidding for advertising in the current traffic.
2. The information processing method for advertising competition according to claim 1, characterized in that, The user data is filtered to determine the valid user data, including: User data that does not meet the preset qualifications will be removed; In addition, for preset nodes in the advertising delivery business, user data that does not meet the conversion rules at the preset node will be removed.
3. The information processing method for advertising competition according to claim 1, characterized in that, Based on the above predictions, the bidding process for determining whether to run ads on the current traffic includes: Based on the above predictions and the business details for each user provided by a third party, it is determined whether to participate in the bidding for advertising based on the current traffic.
4. The information processing method for advertising competition according to claim 1, characterized in that, Based on the predicted conversion rate and the user profile, the bidding price corresponding to the current user data is determined using the bid prediction model, including: The bidding strategy is determined based on the predicted conversion rate and the user profile. The value of the effective user data is determined using a bid prediction model; The bidding bid corresponding to the valid user data is determined based on the bidding strategy and the user value.
5. An information processing device for advertising competition, characterized in that, include: The data filtering unit is configured to filter the user data in the current traffic after obtaining the user data to determine the valid user data. The filtering of user data to determine the valid user data includes: removing user data that does not meet the preset qualifications; and removing user data that does not meet the conversion rules at the preset node in the advertising delivery business. The conversion rate prediction unit is configured to predict the conversion rate corresponding to the effective user data after ad delivery based on a pre-established conversion rate prediction model. The ad delivery business based on RTA is divided into multiple nodes, including click nodes, registration nodes, first login nodes, business application nodes, business application implementation nodes, evaluation nodes, usage nodes, and payment nodes. The conversion rate prediction model is pre-trained, using the actual results of each conversion node as the target variable and the relevant information of the user ID sent in the information stream as the dependent variable. A multi-objective machine learning model learns the characteristics of users with high conversion efficiency across the entire conversion chain. This multi-objective model is invoked to predict the conversion probability of the user across the entire chain when an RTA bidding request is made. The profile prediction unit is configured to predict the user profile of the valid user data using a user qualification profile prediction model. The bidding unit is configured to determine the bidding price corresponding to the current user data based on the predicted conversion rate and the user profile, using a bidding prediction model. This determination includes: determining a bidding strategy based on the predicted conversion rate and the user profile; determining the value of the effective user data using the bidding prediction model; and determining the bidding bid corresponding to the effective user data based on the bidding strategy and the user value. Bidding is based on the conversion rate and user profile. Different bidding bases have different bidding strategies, including strong participation and strong non-participation strategies. Different bidding strategies have different bidding coefficient values, which are used to calculate the bid value. Before determining the bidding price, it is necessary to determine whether to participate in the bidding based on user data in the current traffic. This includes: using a pre-established prediction model to predict the risk level of multiple businesses for each valid user data; and based on the results of the above prediction, determining whether to participate in the bidding for advertising on the current traffic.
6. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions for causing the computer to perform the information processing method for advertising competition as described in any one of claims 1-4.
7. An electronic device, characterized in that, include: At least one processor; And a memory communicatively connected to the at least one processor; wherein the memory stores a computer program executable by the at least one processor, the computer program being executed by the at least one processor to cause the at least one processor to perform the information processing method for advertising competition as described in any one of claims 1-4.