A traffic processing method and apparatus, an electronic device, and a storage medium

Abstract

Embodiments of the present application are suitable for the technical field of information processing, and provide a traffic processing method and device, electronic equipment and storage medium, wherein the traffic processing method comprises: iteratively updating a first parameter of each advertising scheme of at least two advertising schemes based on operation data corresponding to each advertising scheme, until the first parameter converges; the first parameter represents a probability of a first operation event of the corresponding advertising scheme; the operation data comprises operation data related to the first operation event and a second operation event; the first operation event represents an operation event related to a purchase intention; the second operation event represents an operation event of clicking a corresponding commodity page; and determining advertising traffic allocated to each advertising scheme based on the first parameter of each advertising scheme of the at least two advertising schemes when the first parameter converges.

Description

Technical Field

[0001] This invention relates to the field of information processing technology, and in particular to a traffic processing method, apparatus, electronic device, and storage medium. Background Technology

[0002] When e-commerce platforms run ads, they continuously adjust the proportion of traffic allocated to different ad campaigns based on target variables, promptly reassigning traffic from underperforming campaigns to more effective ones. Related technologies only support target variables with rapid feedback, such as click-through rate (CTR). However, CTR and other rapidly responding target variables do not necessarily increase product sales, and traffic allocation strategies based on CTR are not very effective. Summary of the Invention

[0003] To address the aforementioned problems, embodiments of the present invention provide a traffic processing method, apparatus, electronic device, and storage medium to at least resolve the issue of revenue generation from traffic allocation schemes based on click-through rates in related technologies.

[0004] The technical solution of this invention is implemented as follows:

[0005] In a first aspect, embodiments of the present invention provide a traffic processing method, the method comprising:

[0006] Based on the operation data corresponding to each of the at least two advertising schemes, the first parameter of each of the at least two advertising schemes is iteratively updated until the first parameter converges; the first parameter represents the probability of a first operation event occurring in the corresponding advertising scheme; the operation data includes operation data related to the first operation event and the second operation event; the first operation event represents an operation event related to purchase intention; the second operation event represents an operation event of clicking the corresponding product page;

[0007] Based on the first parameter at the convergence of each of the at least two advertising schemes, the advertising traffic allocated to each advertising scheme is determined.

[0008] In the above scheme, the step of iteratively updating the first parameter of each of the at least two advertising schemes based on the operation data corresponding to each of the at least two advertising schemes until the first parameter converges includes:

[0009] Based on the operational data and the first parameters obtained for each advertising scheme during the previous iteration update, a first weight is determined for each second operational event; the first weight represents the strength of the purchase intention.

[0010] Based on the first weight of each second operation event, update the first parameter of each of the at least two advertising schemes;

[0011] Determine whether the updated first parameter has converged.

[0012] In the above scheme, determining whether the updated first parameter has converged includes:

[0013] Calculate the difference between the first parameter obtained from two consecutive iterations;

[0014] If the difference is less than a set value, the first parameter is determined to have converged.

[0015] In the above scheme, determining the first weight of each second operation event based on the operation data and the first parameter obtained during the previous iteration update for each advertising scheme includes:

[0016] If the first operation event occurs after the second operation event, the first set value is determined as the first weight of the corresponding second operation event;

[0017] If the first operation event does not occur after the second operation event occurs, the first weight of the corresponding second operation event is calculated based on the first parameter, the first time interval, and the second parameter obtained in the previous iteration update; the first time interval represents the time interval from the occurrence time of the corresponding second operation event to the start time of the first iteration update; the second parameter represents the number of times the second operation event occurs per unit time.

[0018] In the above scheme, updating the first parameter of each of the at least two advertising schemes based on the first weight of each second operation event includes:

[0019] Based on the first weight, the second parameter, the first time interval corresponding to each second operation event, and the number of occurrences of the first and second operation events, the first parameter after iterative update of each advertising scheme is calculated; the second parameter represents the number of times the second operation event occurs per unit time.

[0020] Wherein, if no first operation event occurs after the second operation event occurs, the first time interval represents the time interval from the occurrence time of the corresponding second operation event to the start time of the first iteration update; if the first operation event occurs after the second operation event occurs, the first time interval represents the time interval from the occurrence time of the corresponding second operation event to the occurrence time of the first operation event.

[0021] In the above scheme, determining the advertising traffic allocated to each advertising scheme based on the first parameter at the convergence of each of the at least two advertising schemes includes:

[0022] Determine the beta distribution interval corresponding to the first parameter of each advertising campaign;

[0023] Each time, a random number is randomly selected from the beta distribution interval corresponding to each advertising scheme, until the set number of times is reached;

[0024] Determine the advertising scheme to which the largest random number in each selection belongs, and determine the number of wins for each advertising scheme;

[0025] Based on the number of wins for each advertising campaign, determine the ad traffic allocated to each advertising campaign.

[0026] In the above scheme, determining the ad traffic allocated to each ad scheme based on the number of wins for each ad scheme includes:

[0027] The proportion of wins for each advertising campaign is determined as the proportion of advertising traffic allocated to each advertising campaign.

[0028] In a second aspect, embodiments of the present invention provide a flow processing apparatus, the apparatus comprising:

[0029] An update module is used to iteratively update the first parameter of each of the at least two advertising schemes based on the operation data corresponding to each of the at least two advertising schemes, until the first parameter converges; the first parameter represents the probability of a first operation event occurring in the corresponding advertising scheme; the operation data includes operation data related to the first operation event and the second operation event; the first operation event represents an operation event related to purchase intention; the second operation event represents an operation event of clicking the corresponding product page;

[0030] The determination module is used to determine the advertising traffic allocated to each advertising scheme based on a first parameter at the convergence of each of the at least two advertising schemes.

[0031] Thirdly, embodiments of the present invention provide an electronic device, including a processor and a memory, which are interconnected. The memory is used to store a computer program, which includes program instructions. The processor is configured to invoke the program instructions to execute the steps of the traffic processing method provided in the first aspect of the present invention.

[0032] Fourthly, embodiments of the present invention provide a computer-readable storage medium, comprising: the computer-readable storage medium storing a computer program. When the computer program is executed by a processor, it implements the steps of the traffic processing method provided in the first aspect of the present invention.

[0033] This invention, in its embodiments, iteratively updates the first parameter of each of at least two advertising schemes based on the operation data corresponding to each of the at least two advertising schemes, until the first parameter converges. Based on the converged first parameter of each of the at least two advertising schemes, the advertising traffic allocated to each advertising scheme is determined. The first parameter represents the probability of a first operation event occurring in the corresponding advertising scheme. The operation data includes operation data related to the first and second operation events. The first operation event represents an operation event related to purchase intention, and the second operation event represents an operation event involving clicking the corresponding product page. This invention uses the first parameter related to purchase intention as an iterative update indicator, iteratively updating the first parameter using the operation data corresponding to each advertising scheme until the first parameter converges, and allocating advertising traffic based on the converged first parameter. Compared to using a rapid feedback indicator, this invention uses the first parameter for adaptive experimentation, which maximizes the return on advertising investment. Ultimately, the advertising scheme with the largest advertising traffic allocation ratio will be the advertising scheme with the largest return, improving resource utilization. Attached Figure Description

[0034] Figure 1 This is a schematic diagram illustrating the implementation process of a traffic processing method provided in an embodiment of the present invention;

[0035] Figure 2 This is a flowchart of calculating the probability of traffic allocation provided in an embodiment of the present invention;

[0036] Figure 3 This is a flowchart illustrating a small loop provided in an embodiment of the present invention;

[0037] Figure 4 This is a schematic diagram illustrating the implementation process of another traffic processing method provided in an embodiment of the present invention;

[0038] Figure 5 This is a schematic diagram illustrating the implementation process of another traffic processing method provided in an embodiment of the present invention;

[0039] Figure 6 This is a schematic diagram illustrating the implementation process of another traffic processing method provided in an embodiment of the present invention;

[0040] Figure 7 This is a schematic diagram of a flow processing device provided in an embodiment of the present invention;

[0041] Figure 8 This is a schematic diagram of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0042] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0043] To boost product sales, businesses choose to advertise on online advertising platforms. To maximize the effectiveness of their advertising strategies, they conduct experiments. Currently, two common types of experiments are split-sample experiments and adaptive experiments. Split-sample experiments distribute samples equally (or at a fixed ratio) across each advertising campaign, representing the most classic approach. The advantage of this method is its simplicity and ability to compare a wider range of target variables. However, its disadvantages include the inability to maximize resource utilization and dynamically reduce investment in underperforming campaigns due to the fixed ratio, and the potential for lengthy or extensive data accumulation to obtain experimental results. Adaptive experiments, through reinforcement learning, continuously adjust the proportion of ad traffic allocated to each campaign based on existing results. This allows for more efficient resource utilization and timely redirection of ad traffic from underperforming campaigns to more effective ones.

[0044] Currently, most adaptive experiments in the industry only support target variables with rapid feedback, such as click-through rate (CTR). However, many important target variables have delayed feedback, such as conversion rate (CVR). Conversions include actions like placing an order or adding items to the cart. Unlike clicks, which occur immediately after an ad is displayed, conversions may happen over a longer period (days or even weeks). Without special handling, adaptive experiments cannot consider delayed outcome variables when adjusting traffic allocation, potentially leading to counterproductive results—for example, allocating more traffic to ad campaigns with low conversion rates while aiming to increase conversion rates.

[0045] To address the shortcomings of the aforementioned related technologies, embodiments of the present invention provide a traffic processing method that can at least improve the revenue of advertising campaigns. To illustrate the technical solution described in this invention, specific embodiments are provided below.

[0046] Figure 1 This is a schematic diagram illustrating the implementation flow of a traffic processing method provided in an embodiment of the present invention. The execution subject of the traffic processing method is an electronic device, including desktop computers, laptops, and servers. (Reference) Figure 1 Traffic processing methods include:

[0047] S101, based on the operation data corresponding to each of the at least two advertising schemes, iteratively update the first parameter of each of the at least two advertising schemes until the first parameter converges; the first parameter represents the probability of the corresponding advertising scheme experiencing a first operation event; the operation data includes operation data related to the first operation event and the second operation event; the first operation event represents an operation event related to purchase intention; the second operation event represents an operation event of clicking the corresponding product page.

[0048] In the embodiments of the invention, the first parameter is illustrated by conversion rate as an example, where conversion refers to a user completing an order.

[0049] The adaptive experiment system is a closed-loop system. In the initial stage of the experiment, each ad campaign has the same display probability. Then, the system collects user behavior feedback, estimates the conversion rate for each ad campaign, and adjusts the traffic allocation probability accordingly. The system then displays each ad campaign to the user according to the new traffic allocation probability, and this process repeats. During this process, ad campaigns with higher estimated conversion rates will gradually receive higher traffic allocation probabilities until the ad campaign with the best conversion rate receives all (or nearly all) of the traffic allocation probability, at which point the experiment ends.

[0050] refer to Figure 2 , Figure 2 This is a flowchart illustrating how to calculate traffic allocation probability according to an embodiment of the present invention. In the experimental group, corresponding advertising schemes, click exposure logs recorded user operation data. Based on this data, the conversion rate was estimated for each advertising scheme. Then, the allocation probability of the traffic corresponding to the advertising scheme was calculated based on the conversion rate. The advertising schemes were then displayed to the user according to the new traffic allocation probability. This process was repeated until the advertising scheme with the optimal conversion rate received all (or nearly all) of the traffic allocation probability, at which point the experiment ended.

[0051] This invention employs an iterative update process. First, a first parameter is iteratively updated until it converges. Then, the ad traffic allocated to each ad campaign is adjusted using the converged first parameter. This process is called a large loop, which can be performed at predetermined time intervals, such as 30 minutes. Each iteration of updating the first parameter is called a small loop, and the output of the small loop serves as the input for the next small loop, until the output (first parameter) converges.

[0052] Here, the first parameter is a delayed metric, which includes conversion rate, favorite rate, add-to-cart rate, and purchase completion rate (conversion rate only considers placing an order; placing an order and not canceling or returning the item is considered a purchase completion).

[0053] Compared to metrics that provide immediate feedback, there are more metrics that provide delayed feedback because most user actions don't occur immediately after an ad is displayed, such as conversion rates, add-to-cart metrics, and total sales. Metrics that provide immediate feedback are primarily limited to click-through rates and other browsing-related metrics. Furthermore, delayed feedback metrics are more closely linked to advertisers' and merchants' revenue; a higher conversion rate means higher sales can be achieved with the same advertising investment. However, a higher click-through rate does not necessarily lead to higher sales.

[0054] Here, the operational data includes operational data related to the first and second operational events. The first operational event represents an operational event related to purchase intention, and the second operational event represents the operational event of clicking the corresponding product page. After a user clicks on a product page and decides to place an order, there will be a delay before the purchase is completed. The shorter the delay, the stronger the user's purchase intention is considered to be.

[0055] Based on the operational data corresponding to each advertising campaign, the first parameter of the advertising campaign is iteratively updated until the first parameter converges. Here, the purchase intention value of each user can be determined based on the interval between the time the user clicks on the product page and the time the user places an order. The first parameter is then calculated based on the overall purchase intention value of each advertising campaign.

[0056] To accurately estimate latency metrics, it's necessary to know their latency distribution. Taking conversion rates as an example, after a user clicks on an ad campaign and decides to place an order, there will be a delay before the purchase is completed. This delay follows a certain probability distribution, known as the latency distribution.

[0057] Different advertising campaigns may correspond to different latency distributions, therefore, it is necessary to estimate the latency metric for each campaign. Assume the latency distribution follows an exponential distribution, corresponding to the parameter λ. The Expectation-Maximization (EM) algorithm is used to estimate the first parameter. Figure 3 This is a flowchart illustrating a small loop provided in an embodiment of the present invention. Figure 3 This diagram illustrates a flowchart of updating the first parameter during a single large loop for an advertising campaign. The first step reads the initial values, which consist of the initial conversion rate (θ0) and the initial delay distribution parameter (λ0). An exponential distribution is used to fit the delay distribution, so λ0 is the rate parameter of the exponential distribution, representing the number of times an event (e.g., order completion) occurs per unit of time. If the experiment has just begun, this initial value is manually set. During the experiment, the initial value is the result of the previous large loop calculation. A smaller loop then follows, consisting of two steps: calculating weights and updating the parameters.

[0058] refer to Figure 4In one embodiment, the step of iteratively updating the first parameter of each of the at least two advertising schemes based on the operation data corresponding to each of the at least two advertising schemes until the first parameter converges includes:

[0059] S401, based on the operation data and the first parameters obtained by each advertising scheme in the previous iteration update, determine the first weight of each second operation event; the first weight represents the strength of the purchase intention.

[0060] In the first step of the small loop in this embodiment of the invention, the weights are calculated.

[0061] Here, in the small loop process, if the experiment has just started, the first parameter used in the first small loop is manually set, and the first parameter used in each subsequent small loop is the first parameter obtained from the previous small loop iteration update.

[0062] In one embodiment, determining the first weight of each second operation event based on the operation data and the first parameter obtained during the previous iteration update for each advertising scheme includes:

[0063] If the first operation event occurs after the second operation event, the first set value is determined as the first weight of the corresponding second operation event;

[0064] If the first operation event does not occur after the second operation event occurs, the first weight of the corresponding second operation event is calculated based on the first parameter, the first time interval, and the second parameter obtained in the previous iteration update; the first time interval represents the time interval from the occurrence time of the corresponding second operation event to the start time of the first iteration update; the second parameter represents the number of times the second operation event occurs per unit time.

[0065] Here, a first weight is calculated for each user click, representing the strength of their purchase intention. For example, if a user clicks on a product page and then converts, the first weight is set to 1. If a user clicks on a product page but does not convert, the first weight is determined based on the time elapsed since the click. The maximum first weight is 1; if no conversion occurs, the first weight is less than 1.

[0066] In practical applications, the first weight of each second operation event can be calculated using the following formula:

[0067]

[0068] Let t be the number of updates in the main loop (t = 1, 2, 3...), and let TIME(t) be the clock time at the start of the t-th update in the main loop (e.g., 3:30 PM on November 5th). Let l be the number of updates in the smaller loops within the main loop (l = 1, 2, 3...). This represents the first weight of click i during the l-th sub-loop update within the t-th major loop.

[0069] and In the first small loop, the result of the previous large loop (t-1th large loop) should be used. If it is the first large loop calculation, the initial values ​​θ0 and λ0 should be used. In subsequent small loops, and This represents the result of the previous small loop. Y it This indicates whether a conversion has occurred up to TIME(t) after clicking i. If a conversion occurs after clicking i (Y... it When =1), e it This represents the time interval from click time to conversion time; when click i does not convert (Y) it When =0), e it This represents the interval from the click time to TIME(t).

[0070] The first parameter obtained during the previous iteration update is The first time interval is e it The second parameter is

[0071] Based on the above formula, the first weight of each second operation event is calculated.

[0072] S402, based on the first weight of each of the at least two advertising schemes, update the first parameter of each advertising scheme.

[0073] In the second step of the small loop in this embodiment of the invention, the parameters are updated.

[0074] In one embodiment, updating the first parameter of each of the at least two advertising schemes based on the first weight of each second operation event includes:

[0075] Based on the first weight, the second parameter, the first time interval corresponding to each second operation event, and the number of occurrences of the first and second operation events, the first parameter after iterative update of each advertising scheme is calculated; the second parameter represents the number of times the second operation event occurs per unit time.

[0076] Wherein, if no first operation event occurs after the second operation event occurs, the first time interval represents the time interval from the occurrence time of the corresponding second operation event to the start time of the first iteration update; if the first operation event occurs after the second operation event occurs, the first time interval represents the time interval from the occurrence time of the corresponding second operation event to the occurrence time of the first operation event.

[0077] In practical applications, the updated first parameter can be calculated using the following formula:

[0078] First, the optimal θ and λ are obtained by solving the following optimization problem:

[0079]

[0080] The above optimization problem can be solved algebraically:

[0081]

[0082]

[0083] in, This represents the total number of transformations observed from the start of the experiment up to TIME(t). This represents the number of new clicks observed between the (s-1)th and sth major cycles.

[0084] Based on the optimization problem described above, the updated first and second parameters can be obtained.

[0085] S403, determine whether the updated first parameter has converged.

[0086] The first parameter is considered to have converged when the difference between the first parameter calculated in two consecutive iterations is not significant.

[0087] refer to Figure 5 In one embodiment, determining whether the updated first parameter has converged includes:

[0088] S501, calculate the difference between the first parameters obtained from two consecutive iterations.

[0089] S502, if the difference is less than a set value, determine that the first parameter has converged.

[0090] If the difference between the first parameter obtained from two consecutive iterations is less than a set value, the first parameter is considered to have converged.

[0091] In practical applications, since the second parameter is the lambda value of the first parameter, the small loop will only stop when the second parameter converges as well. The convergence of the first and second parameters can be determined using the following formula:

[0092] and The time is when it converges.

[0093] If the first parameter converges, the loop exits, thus yielding the conversion rate for one advertising campaign. By performing the same calculation on all advertising campaigns, we can obtain the conversion rate for each campaign. Then, ad traffic is allocated based on the conversion rate.

[0094] If it is determined that the first parameter has not converged, repeat the small loop to calculate the weights and update the parameters until the first parameter converges.

[0095] S102, based on the first parameter at the convergence of each of the at least two advertising schemes, determine the advertising traffic allocated to each advertising scheme.

[0096] refer to Figure 6 In one embodiment, determining the ad traffic allocated to each ad campaign based on a first parameter at the convergence of each of the at least two ad campaigns includes:

[0097] S601, determine the beta distribution interval corresponding to the first parameter of each advertising scheme.

[0098] After obtaining the conversion rate for each advertising campaign, allocate advertising traffic to each campaign based on the conversion rate.

[0099] In this embodiment of the invention, it is assumed that the true conversion rate of each advertising campaign follows a Beta distribution (α). kt ,β kt Before the experiment begins (t=0), α is set for all advertising schemes k. k0 =1,β k0 =1.

[0100] In each subsequent major cycle update, the following settings are configured:

[0101]

[0102]

[0103] in, The first parameter represents the final estimate of advertising scheme k in the t-th major cycle.

[0104] During the t-th major cycle update, the Monte Carlo method can be used to calculate the probability that each advertising scheme has the highest first parameter.

[0105] Specifically, the first step is to determine the beta distribution interval corresponding to the first parameter of each advertising scheme. The beta distribution is a density function that serves as the conjugate prior distribution of the Bernoulli and binomial distributions. In probability theory, the beta distribution refers to a set of continuous probability distributions defined in the interval (0, 1).

[0106] S602: Each time, a random number is randomly selected from the beta distribution interval corresponding to each advertising scheme until the set number of times is reached.

[0107] For each advertising campaign, randomly select a random number from the beta distribution interval and record the advertising campaign to which the largest random number belongs. Increment the "win" count of that advertising campaign by 1. Repeat this step a set number of times, such as 1000 times.

[0108] S603, determine the advertising scheme to which the largest random number in each selected random number belongs, and determine the number of wins for each advertising scheme.

[0109] The statistics show the number of times each advertising campaign "wins" within the set number of attempts.

[0110] S604 determines the ad traffic allocated to each ad campaign based on the number of wins for each ad campaign.

[0111] In one embodiment, determining the ad traffic allocated to each ad campaign based on the number of wins for each ad campaign includes:

[0112] The proportion of wins for each advertising campaign is determined as the proportion of advertising traffic allocated to each advertising campaign.

[0113] For example, suppose there are a total of 4 advertising campaigns, and the winning ratio of the 4 advertising campaigns is 1:2:3:4. Then the advertising traffic will be allocated to the corresponding advertising campaigns in the same ratio of 1:2:3:4.

[0114] After determining the traffic allocation ratio, the next large loop begins until the best advertising plan receives all (or nearly all) of the traffic allocation probability, at which point the experiment ends. For example, a threshold can be set; if the traffic allocation ratio of a certain advertising plan exceeds this threshold, the experiment ends, and that advertising plan is determined as the one officially launched into the market.

[0115] Since the first parameter used in this embodiment of the invention is a delayed feedback metric, experimenters can choose any one of conversion rate, collection rate, add-to-cart rate, and purchase completion rate as the first parameter for their experiments. Delayed feedback metrics are more closely linked to the revenue of advertisers and merchants. Compared to rapidly responding metrics, using delayed feedback metrics for adaptive experiments ensures that the advertising scheme with the highest traffic allocation ratio is the one with the highest return, maximizing the return on advertising investment and improving resource utilization.

[0116] In practical applications, relevant experimental statistics can be displayed to the experimenter in the form of a control panel. For example, it can display the first parameter and winning probability (the proportion of winning times) for each advertising strategy in real time. It can also estimate the confidence interval of the first parameter. The estimated first parameter is a random variable, and its value may be higher or lower than the true first parameter due to noise in the data. Simply put, the confidence interval provides a range within which the true first parameter might exist. Specifically, the confidence interval of the first parameter can be determined by obtaining the 10th and 90th percentile values ​​of the beta distribution.

[0117] This invention, in its embodiments, iteratively updates the first parameter of each of at least two advertising schemes based on the operation data corresponding to each of the at least two advertising schemes, until the first parameter converges. Based on the converged first parameter of each of the at least two advertising schemes, the advertising traffic allocated to each advertising scheme is determined. The first parameter represents the probability of a first operation event occurring in the corresponding advertising scheme. The operation data includes operation data related to the first and second operation events. The first operation event represents an operation event related to purchase intention, and the second operation event represents an operation event involving clicking the corresponding product page. This invention uses the first parameter related to purchase intention as an iterative update indicator, iteratively updating the first parameter using the operation data corresponding to each advertising scheme until the first parameter converges, and allocating advertising traffic based on the converged first parameter. Compared to using a rapid feedback indicator, this invention uses the first parameter for adaptive experimentation, ultimately ensuring that the advertising scheme with the largest advertising traffic allocation ratio is the one with the highest return, maximizing the return on advertising investment and improving resource utilization.

[0118] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0119] It should be understood that, when used in this specification and the appended claims, the terms "comprising" and "including" indicate the presence of the described features, integrals, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or collections thereof.

[0120] It should be noted that the technical solutions described in the embodiments of the present invention can be combined arbitrarily without conflict.

[0121] In addition, in the embodiments of the present invention, "first," "second," etc. are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence.

[0122] refer to Figure 7 , Figure 7 This is a schematic diagram of a flow processing device provided in an embodiment of the present invention, as shown below. Figure 7 As shown, the device includes an update module and a determination module.

[0123] An update module is used to iteratively update the first parameter of each of the at least two advertising schemes based on the operation data corresponding to each of the at least two advertising schemes, until the first parameter converges; the first parameter represents the probability of a first operation event occurring in the corresponding advertising scheme; the operation data includes operation data related to the first operation event and the second operation event; the first operation event represents an operation event related to purchase intention; the second operation event represents an operation event of clicking the corresponding product page;

[0124] The determination module is used to determine the advertising traffic allocated to each advertising scheme based on a first parameter at the convergence of each of the at least two advertising schemes.

[0125] In one embodiment, the update module iteratively updates a first parameter of each of the at least two advertising schemes based on the operation data corresponding to each of the at least two advertising schemes, until the first parameter converges, including:

[0126] Based on the operational data and the first parameters obtained for each advertising scheme during the previous iteration update, a first weight is determined for each second operational event; the first weight represents the strength of the purchase intention.

[0127] Based on the first weight of each second operation event, update the first parameter of each of the at least two advertising schemes;

[0128] Determine whether the updated first parameter has converged.

[0129] In one embodiment, the update module determines whether the updated first parameter has converged, including:

[0130] Calculate the difference between the first parameter obtained from two consecutive iterations;

[0131] If the difference is less than a set value, the first parameter is determined to have converged.

[0132] In one embodiment, when the update module determines the first weight of each second operation event based on the operation data and the first parameter obtained by each advertising scheme in the previous iteration update, it is used to:

[0133] If the first operation event occurs after the second operation event, the first set value is determined as the first weight of the corresponding second operation event;

[0134] If the first operation event does not occur after the second operation event occurs, the first weight of the corresponding second operation event is calculated based on the first parameter, the first time interval, and the second parameter obtained in the previous iteration update; the first time interval represents the time interval from the occurrence time of the corresponding second operation event to the start time of the first iteration update; the second parameter represents the number of times the second operation event occurs per unit time.

[0135] In one embodiment, when the update module updates the first parameter of each of the at least two advertising schemes based on the first weight of each second operation event, it is used to:

[0136] Based on the first weight, the second parameter, the first time interval corresponding to each second operation event, and the number of occurrences of the first and second operation events, the first parameter after iterative update of each advertising scheme is calculated; the second parameter represents the number of times the second operation event occurs per unit time.

[0137] Wherein, if no first operation event occurs after the second operation event occurs, the first time interval represents the time interval from the occurrence time of the corresponding second operation event to the start time of the first iteration update; if the first operation event occurs after the second operation event occurs, the first time interval represents the time interval from the occurrence time of the corresponding second operation event to the occurrence time of the first operation event.

[0138] In one embodiment, the determining module determines the ad traffic allocated to each ad campaign based on a first parameter at the convergence of each of the at least two ad campaigns, including:

[0139] Determine the beta distribution interval corresponding to the first parameter of each advertising campaign;

[0140] Each time, a random number is randomly selected from the beta distribution interval corresponding to each advertising scheme, until the set number of times is reached;

[0141] Determine the advertising scheme to which the largest random number in each selection belongs, and determine the number of wins for each advertising scheme;

[0142] Based on the number of wins for each advertising campaign, determine the ad traffic allocated to each advertising campaign.

[0143] In one embodiment, the determining module determines the ad traffic allocated to each ad campaign based on the number of wins for each ad campaign, including:

[0144] The proportion of wins for each advertising campaign is determined as the proportion of advertising traffic allocated to each advertising campaign.

[0145] In practical applications, the update module and the determination module can be implemented by processors in electronic devices, such as central processing units (CPUs), digital signal processors (DSPs), microcontroller units (MCUs), or field-programmable gate arrays (FPGAs).

[0146] It should be noted that the traffic processing device provided in the above embodiments is only illustrated by the division of the above modules. In practical applications, the above processing can be assigned to different modules as needed, that is, the internal structure of the device can be divided into different modules to complete all or part of the processing described above. In addition, the traffic processing device and the traffic processing method embodiments provided in the above embodiments belong to the same concept, and the specific implementation process can be found in the method embodiments, which will not be repeated here.

[0147] Based on the hardware implementation of the above program modules, and in order to implement the method of the embodiments of this application, the embodiments of this application also provide an electronic device. Figure 8 This is a schematic diagram of the hardware structure of the electronic device according to an embodiment of this application, as shown below. Figure 8 As shown, the electronic device includes:

[0148] A communication interface enables information exchange with other devices, such as network devices.

[0149] The processor, connected to the communication interface, enables information interaction with other devices and, when running a computer program, executes the methods provided by one or more technical solutions on the electronic device side. The computer program is stored in memory.

[0150] Of course, in practical applications, the various components in an electronic device are coupled together through a bus system. It can be understood that the bus system is used to achieve communication and connection between these components. In addition to the data bus, the bus system also includes a power bus, a control bus, and a status signal bus. However, for the sake of clarity, in... Figure 8 The general will label all buses as bus systems.

[0151] The memory in this application embodiment is used to store various types of data to support the operation of the electronic device. Examples of such data include any computer program used to operate on the electronic device.

[0152] It is understood that memory can be volatile or non-volatile, or both. Non-volatile memory can be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), ferromagnetic random access memory (FRAM), flash memory, magnetic surface memory, optical disc, or compact disc read-only memory (CD-ROM); magnetic surface memory can be disk storage or magnetic tape storage. Volatile memory can be random access memory (RAM), which is used as an external cache. By way of example, but not limitation, many forms of RAM are available, such as Static Random Access Memory (SRAM), Synchronous Static Random Access Memory (SSRAM), Dynamic Random Access Memory (DRAM), Synchronous Dynamic Random Access Memory (SDRAM), Double Data Rate Synchronous Dynamic Random Access Memory (DDRSDRAM), Enhanced Synchronous Dynamic Random Access Memory (ESDRAM), SyncLink Dynamic Random Access Memory (SLDRAM), and Direct Rambus Random Access Memory (DRRAM). The memories described in the embodiments of this application are intended to include, but are not limited to, these and any other suitable types of memory.

[0153] The methods disclosed in the embodiments of this application can be applied to a processor or implemented by a processor. The processor may be an integrated circuit chip with signal processing capabilities. During implementation, each step of the above method can be completed by integrated logic circuits in the processor's hardware or by instructions in software form. The processor may be a general-purpose processor, a DSP, or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The processor can implement or execute the methods, steps, and logic block diagrams disclosed in the embodiments of this application. A general-purpose processor may be a microprocessor or any conventional processor, etc. The steps of the methods disclosed in the embodiments of this application can be directly manifested as execution by a hardware decoding processor, or execution by a combination of hardware and software modules in the decoding processor. The software modules may be located in a storage medium, which is located in memory. The processor reads the program from the memory and, in conjunction with its hardware, completes the steps of the aforementioned method.

[0154] Optionally, when the processor executes the program, it implements the corresponding processes implemented by the electronic device in the various methods of the embodiments of this application. For the sake of brevity, these will not be described in detail here.

[0155] In an exemplary embodiment, this application also provides a storage medium, namely a computer storage medium, specifically a computer-readable storage medium, such as a first memory storing a computer program, which can be executed by a processor of an electronic device to complete the steps described in the aforementioned method. The computer-readable storage medium may be a memory such as FRAM, ROM, PROM, EPROM, EEPROM, Flash Memory, magnetic surface memory, optical disc, or CD-ROM.

[0156] In the several embodiments provided in this application, it should be understood that the disclosed apparatus, electronic devices, and methods can be implemented in other ways. The apparatus embodiments described above are merely illustrative. For example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods, such as: multiple units or components may be combined, or integrated into another system, or some features may be ignored or not executed. In addition, the coupling, direct coupling, or communication connection between the various components shown or discussed may be through some interfaces, and the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0157] The units described above as separate components may or may not be physically separate. The components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of the units may be selected to achieve the purpose of this embodiment according to actual needs.

[0158] In addition, each functional unit in the various embodiments of this application can be integrated into one processing unit, or each unit can be a separate unit, or two or more units can be integrated into one unit; the integrated unit can be implemented in hardware or in the form of hardware plus software functional units.

[0159] Those skilled in the art will understand that all or part of the steps of the above method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it performs the steps of the above method embodiments. The aforementioned storage medium includes various media that can store program code, such as mobile storage devices, ROM, RAM, magnetic disks, or optical disks.

[0160] Alternatively, if the integrated units described above are implemented as software functional modules and sold or used as independent products, they can also be stored in a computer-readable storage medium. Based on this understanding, the technical solutions of the embodiments of this application, or the parts that contribute to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as mobile storage devices, ROM, RAM, magnetic disks, or optical disks.

[0161] It should be noted that the technical solutions described in the embodiments of this application can be combined arbitrarily without conflict.

[0162] In addition, in this application example, "first," "second," etc. are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence.

[0163] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A traffic processing method, characterized by, The method comprises: determining a first weight of each second operation event based on operation data corresponding to each of the at least two advertisement schemes and a first parameter obtained by each of the at least two advertisement schemes in the last iteration update; the first weight representing a strength of a purchase intention; the first parameter representing a probability of a first operation event of a corresponding advertisement scheme; the operation data comprising operation data related to the first operation event and the second operation event; the first operation event representing an operation event related to the purchase intention; and the second operation event representing an operation event of clicking a corresponding product page; calculating the first parameter of each advertisement scheme after the iteration update based on the following formula: wherein, is a first parameter, is a second parameter, characterizes whether each second operation event has conversion, is a first weight of each second operation event, is a first time interval corresponding to each second operation event; the second parameter characterizes the number of second operation events occurring per unit time; wherein, in the case that no first operation event occurs after the occurrence of the second operation event, the first time interval characterizes the time interval from the occurrence time of the corresponding second operation event to the start time of the first iterative update; in the case that a first operation event occurs after the occurrence of the second operation event, the first time interval characterizes the time interval from the occurrence time of the corresponding second operation event to the occurrence time of the first operation event; determining advertisement traffic allocated to each of the at least two advertisement schemes based on the first parameter of each of the at least two advertisement schemes in the convergence.

2. The method of claim 1, wherein, The determination of whether the updated first parameter converges comprises: calculating a difference value of the first parameters obtained by two adjacent iteration updates; in a case where the difference value is less than a set value, determining that the first parameter converges.

3. The method of claim 1, wherein, The determination of the first weight of each second operation event based on the operation data corresponding to each of the at least two advertisement schemes and the first parameter obtained by each of the at least two advertisement schemes in the last iteration update comprises: in a case where the first operation event occurs after the second operation event, determining a first set value as the first weight of the corresponding second operation event; in a case where the first operation event does not occur after the second operation event, calculating the first weight of the corresponding second operation event based on the first parameter obtained by the last iteration update, a first time interval and a second parameter; the first time interval representing a time interval from a time of occurrence of the corresponding second operation event to a start time of the first iteration update; and the second parameter representing a number of times of occurrence of the second operation event per unit time.

4. The method of claim 1, wherein, The determination of the advertisement traffic allocated to each of the at least two advertisement schemes based on the first parameter of each of the at least two advertisement schemes in the convergence comprises: determining a beta distribution interval corresponding to the first parameter of each of the at least two advertisement schemes; randomly selecting a random number from the beta distribution interval corresponding to each of the at least two advertisement schemes each time until a set number of times is reached; determining an advertisement scheme to which a random number with a maximum value in the random numbers selected each time belongs, and determining a winning number of times of each of the at least two advertisement schemes; determining the advertisement traffic allocated to each of the at least two advertisement schemes based on the winning number of times of each of the at least two advertisement schemes.

5. The method of claim 4, wherein, The determination of the advertisement traffic allocated to each of the at least two advertisement schemes based on the winning number of times of each of the at least two advertisement schemes comprises: determining a proportion of the winning number of times of each of the at least two advertisement schemes as a proportion of the advertisement traffic allocated to each of the at least two advertisement schemes.

6. A traffic processing device, characterized by The method comprises: an updating module configured to determine a first weight of each second operation event based on operation data corresponding to each of the at least two advertisement schemes and a first parameter obtained by each of the at least two advertisement schemes in the last iteration update; calculating the first parameter of each advertisement scheme after the iteration update based on the following formula: wherein, is a first parameter, is a second parameter, characterizes whether each second operation event has conversion, is a first weight of each second operation event, is a first time interval corresponding to each second operation event; the first weight characterizes the strength of the purchase intention; the first parameter characterizes the probability of the corresponding advertisement scheme having the first operation event; the operation data includes operation data related to the first operation event and the second operation event; the first operation event characterizes an operation event related to the purchase intention; the second operation event characterizes an operation event of clicking the corresponding product page; the second parameter characterizes the number of times of the second operation event occurring per unit time; wherein, in the case that the first operation event does not occur after the second operation event occurs, the first time interval characterizes the time interval from the occurrence time of the corresponding second operation event to the start time of the first iteration update; in the case that the first operation event occurs after the second operation event occurs, the first time interval characterizes the time interval from the occurrence time of the corresponding second operation event to the occurrence time of the first operation event; A determining module is configured to determine whether the updated first parameter converges, and determine the advertising traffic allocated to each of the at least two advertising schemes based on the first parameter at the time of convergence of each of the at least two advertising schemes.

7. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, The processor implements the traffic processing method as claimed in any one of claims 1 to 5 when executing the computer program.

8. A computer-readable storage medium, characterized in that, The computer readable storage medium stores a computer program, and the computer program comprises program instructions which, when executed by a processor, cause the processor to perform the traffic processing method as claimed in any one of claims 1 to 5.

Images