Optimization analysis method, system and medium based on web page advertising feedback data

Abstract

The present invention discloses an optimization analysis method, system and medium based on web page advertisement feedback data, relates to the technical field of advertisement delivery optimization, and comprises the following steps: collecting feedback data of web page advertisements, and cleaning the feedback data; evaluating delivery effects of advertisement versions of web page advertisements, which is named as version delivery effects; dividing audience groups of web page advertisements according to age, and calculating the best delivery ratio of web page advertisements based on feedback data of web page advertisements in different audience ages; optimizing advertisement delivery strategies based on version delivery effects and the best delivery ratio; the present invention is used to solve the problems that existing advertisement delivery optimization technologies still have a large number of invalid pushes and the amount of calculation is too large, resulting in poor effect of advertisement delivery and the stability of program operation being easily affected.

Description

Technical Field

[0001] The present invention relates to the technical field of advertisement delivery optimization, and in particular to an optimization analysis method, system and medium based on web page advertisement feedback data. Background Art

[0002] Advertising optimization technology refers to the technical methods and strategies that use the feedback data collected during the advertising process to improve the effectiveness and return rate of advertising through analysis and optimization.

[0003] Existing advertising delivery optimization technologies usually analyze and deliver advertisements based on users' purchase and browsing behaviors. However, usually, after users have purchased such products, they no longer need to purchase such products within a certain period of time. In many existing advertising delivery optimization technologies, after users have purchased similar products, such products are repeatedly recommended to users, resulting in a large number of invalid push notifications. In addition, advertisements can play a certain role in brand promotion when being pushed. Even if users do not purchase products, they can deepen users' impressions. Therefore, advertisements cannot be delivered solely for the purpose of profit when being pushed. The promotion effect of products among different users should also be considered. In addition, many existing advertising delivery optimization technologies have problems such as analyzing products based on individuals. The method of determining the placement of advertisements based on the object, while analyzing based on individuals will result in too much calculation for advertisement placement, and one or more analyses need to be performed for each person, which is not conducive to the stable operation of the program. For example, in the Chinese patent application publication number CN111178954A, an advertisement placement method, system and electronic device are disclosed. The scheme is to analyze the user behavior data of individual users as the analysis object, so as to select advertisements for placement. This method will generate huge computing data, which is not conducive to the stability of program operation. The existing advertisement placement optimization technology still has many invalid pushes and too much calculation, resulting in poor effect of advertisement placement and the stability of program operation is easily affected. Summary of the invention

[0004] The present invention aims to solve one of the technical problems in the prior art to at least a certain extent, by collecting feedback data of web page advertisements, cleaning the feedback data, evaluating the version delivery effect of the advertisement version of the web page advertisement, and then dividing the audience groups of the web page advertisements according to age, and calculating the optimal delivery ratio of the web page advertisements based on the feedback data of the web page advertisements among different audience ages, and finally optimizing the advertisement delivery strategy based on the version delivery effect and the optimal delivery ratio, so as to solve the problems that the existing advertisement delivery optimization technology still has a large number of invalid pushes and the amount of calculation is too large, resulting in poor effect of advertisement delivery and the stability of program operation is easily affected.

[0005] To achieve the above objectives, in a first aspect, the present application provides an optimization analysis method based on web page advertisement feedback data, comprising the following steps:

[0006] Collect feedback data of web page advertisements and clean the feedback data;

[0007] Evaluate the delivery effect of the advertising versions of web ads, named version delivery effect;

[0008] Divide the audience groups of web ads by age, and calculate the optimal placement ratio of web ads based on the feedback data of web ads among audiences of different ages;

[0009] Optimize advertising delivery strategies based on version delivery effects and optimal delivery ratios.

[0010] Furthermore, collecting feedback data of web page advertisements and cleaning the feedback data include the following sub-steps:

[0011] Collect feedback data about web ads;

[0012] Eliminate invalid data from feedback data.

[0013] Furthermore, collecting feedback data of web page advertisements includes the following sub-steps:

[0014] When a user clicks on a web ad and enters the ad page, a record of page views is recorded, and the number of times the web ad is pushed is counted, marked as push volume. The record click rate of the web ad is calculated by dividing the push volume by the record of page views.

[0015] When a user enters an advertisement page, the length of time the user browses the advertisement page is recorded and marked as recorded browsing time. The recorded browsing time is the length of time from when the user enters the advertisement page to when the user exits the advertisement page. The average value of the recorded browsing time is calculated and named as recorded average stay time.

[0016] Record the number of times users purchase products through the ad page, named as the number of conversions. Divide the number of recorded views by the number of conversions to calculate the record conversion rate of the web ad.

[0017] The recorded page views, recorded click-through rates, recorded average dwell time, and recorded conversion rates are collectively referred to as feedback data.

[0018] Furthermore, eliminating invalid data in the feedback data includes the following sub-steps:

[0019] Obtain the recorded browsing time, compare the recorded browsing time with the first valid time, and if the recorded browsing time is less than or equal to the first valid time, output an invalid browsing signal; if the recorded browsing time is greater than the first valid time, output a valid browsing signal;

[0020] If a valid browsing signal is output, the recorded browsing time is marked as the valid time, and the effective average stay time of the web page advertisement is recalculated based on the valid time;

[0021] Mark the number of effective durations as effective pageviews, and recalculate the effective click rate and effective conversion rate based on the effective pageviews;

[0022] Effective page views, effective click-through rate, effective conversion rate and effective average stay time are collectively referred to as effective feedback data.

[0023] Furthermore, evaluating the delivery effect of the advertisement version of the webpage advertisement, which is named as version delivery effect, includes the following sub-steps:

[0024] The effective feedback data is calculated separately for different advertisement versions;

[0025] For any advertisement version, obtain the corresponding valid feedback data and mark it as data to be evaluated;

[0026] The delivery effect calculation formula is used to calculate the data to be evaluated to obtain the delivery effect of the advertising version;

[0027] The delivery effect calculation formula is configured as follows: ; Among them, Vae is the version delivery effect, CT is the effective click-through rate, CR is the effective conversion rate, TA is the effective average stay time, and α is the time parameter.

[0028] Furthermore, the audience groups of web page advertisements are divided according to age, and based on the feedback data of web page advertisements among audiences of different ages, the optimal placement ratio of web page advertisements is calculated, which includes the following sub-steps:

[0029] A first age stage, a second age stage, a third age stage and a fourth age stage are set, and a first audience group, a second audience group, a third audience group and a fourth audience group are obtained based on the first age stage, the second age stage, the third age stage and the fourth age stage;

[0030] Obtaining valid feedback data corresponding to the first audience group, the second audience group, the third audience group, and the fourth audience group, respectively naming them as first feedback data, second feedback data, third feedback data, and fourth feedback data;

[0031] Calculate the product of the effective click rate and the effective conversion rate in the first feedback data, and name the calculation result as the first delivery effect; calculate the product of the effective click rate and the effective conversion rate in the second feedback data, and name the calculation result as the second delivery effect; calculate the product of the effective click rate and the effective conversion rate in the third feedback data, and name the calculation result as the third delivery effect; calculate the product of the effective click rate and the effective conversion rate in the fourth feedback data, and name the calculation result as the fourth delivery effect; mark the first delivery effect, the second delivery effect, the third delivery effect, and the fourth delivery effect as audience delivery effects;

[0032] Sort the audience delivery effects from largest to smallest, obtain the audience groups corresponding to the top two audience delivery effects, and mark them as target delivery groups;

[0033] Analyze the effective feedback data of the target group and calculate the optimal delivery ratio of web ads.

[0034] Furthermore, analyzing the effective feedback data of the target delivery group includes the following sub-steps:

[0035] The target delivery groups are numbered and represented by symbols T1 and T2;

[0036] Get the audience delivery effects corresponding to T1 and T2, represented by symbols S1 and S2;

[0037] Obtaining the preset delivery volume and expected conversion volume of the page advertisement, wherein the preset delivery volume is the number of times the page advertisement needs to be delivered preset in the platform; the expected conversion volume is the number of orders expected by the merchant to which the page advertisement belongs;

[0038] Set the increase parameter and construct the equation ; Where Ecr is the expected conversion volume, Pip is the increase parameter, PsV is the preset delivery volume, a is the delivery ratio of T1, and b is the delivery ratio of T2;

[0039] Solving the equation to get a and b is the optimal delivery ratio.

[0040] Furthermore, the optimization of the advertising delivery strategy based on the version delivery effect and the best delivery ratio includes the following sub-steps:

[0041] Get the maximum value among the version delivery effects and mark it as the best delivery version;

[0042] Deliver page ads to the target audience according to the best delivery version and deliver them according to the best delivery ratio.

[0043] In a second aspect, the present application provides an optimization analysis system based on web page advertising feedback data, including a feedback data collection module, a version delivery evaluation module, an audience delivery evaluation module and an advertising delivery optimization module; the feedback data collection module, the version delivery evaluation module and the audience delivery evaluation module are respectively connected to the advertising delivery optimization module data;

[0044] The feedback data collection module is used to collect feedback data of web page advertisements and clean the feedback data;

[0045] The version delivery evaluation module is used to evaluate the delivery effect of the advertisement version of the webpage advertisement, which is named as version delivery effect;

[0046] The audience delivery evaluation module is used to divide the audience groups of web page advertisements according to age, and calculate the optimal delivery ratio of web page advertisements based on the feedback data of web page advertisements among audiences of different ages;

[0047] The advertisement delivery optimization module is used to optimize the advertisement delivery strategy based on the version delivery effect and the best delivery ratio.

[0048] In a third aspect, the present application provides a storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the steps in the above method are performed.

[0049] Beneficial effects of the present invention: The present invention obtains effective feedback data by collecting feedback data of web page advertisements and then eliminating invalid data in the feedback data. The advantage is that during the browsing process, users may enter the advertisement page due to accidental touch or improper operation for various reasons. In this case, users usually exit the advertisement page quickly, but the click will be recorded in the feedback data of the background. Once the pageview statistics are wrong, various subsequent feedback data will be wrong. Therefore, it is necessary to analyze the effectiveness of the feedback data, which improves the effectiveness and rationality of the data in the optimization analysis of advertisement delivery.

[0050] The present invention evaluates the version delivery effect of the web page advertisement based on effective feedback data. The advantage is that there are different versions of advertisements nowadays, such as video version and pop-up version, and users have different acceptance of different versions of advertisements. If the advertisement is delivered to users with the version with the lowest user acceptance, it will make users disgusted. Therefore, it is necessary to find the advertisement version with the highest user acceptance for delivery, which further improves the effectiveness and rationality of advertisement delivery optimization.

[0051] The present invention divides the audience groups of web page advertisements according to age, calculates the optimal delivery ratio of web page advertisements based on the feedback data of web page advertisements in different audience ages, and finally optimizes the advertisement delivery strategy based on the delivery effect of each version and the optimal delivery ratio. The advantage is that different products have different audience groups. Compared with analyzing the user's personal behavior data, directly analyzing the audience groups can greatly reduce the amount of calculation of the advertisement delivery strategy operation, and can reduce the probability of repeated push to users who have recently purchased similar products, thereby improving the accuracy and effectiveness of advertisement delivery. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1 is a functional block diagram of the system of the present invention;

[0053] Figure 2 A schematic diagram of data cleaning of the present invention;

[0054] Figure 3 is a flow chart of the steps of the method of the present invention;

[0055] Figure 4 It is a schematic structural diagram of the electronic device of the present invention. DETAILED DESCRIPTION

[0056] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0057] Example 1, please refer to Figure 1 As shown, the present application provides an optimization analysis system based on web page advertising feedback data, including a feedback data collection module, a version delivery evaluation module, an audience delivery evaluation module and an advertising delivery optimization module; the feedback data collection module, the version delivery evaluation module and the audience delivery evaluation module are respectively connected to the advertising delivery optimization module data;

[0058] The feedback data collection module is used to collect feedback data of web page advertisements and clean the feedback data at the same time; the feedback data collection module includes a feedback data collection unit and a feedback data cleaning unit;

[0059] The feedback data collection unit is used to collect feedback data of web page advertisements;

[0060] The feedback data collection unit is configured with a feedback data collection strategy, which includes:

[0061] When a user clicks on a web ad and enters the ad page, a record of page views is recorded, and the number of times the web ad is pushed is counted, marked as push volume. The record click rate of the web ad is calculated by dividing the push volume by the record of page views.

[0062] When a user enters an ad page, the length of time the user browses the ad page is recorded and marked as recorded browsing time. The recorded browsing time is the length of time from when the user enters the ad page to when the user exits the ad page. The average value of the recorded browsing time is calculated and named as recorded average stay time.

[0063] Record the number of times users purchase products through the ad page, named as the number of conversions. Divide the number of recorded views by the number of conversions to calculate the record conversion rate of the web ad.

[0064] Recording page views, recording click-through rates, recording average dwell time, and recording conversion rates are collectively referred to as feedback data;

[0065] In actual applications, the user referred to in the feedback data is not a single user, but all users on the platform. The page advertisement refers to the advertisement of a specific product, such as "XXX laptop 2023 model"; the record page views are recorded as 11872, the record push volume is 86544, and the record click rate is calculated to be 0.1372, that is, 13.72%. The calculation result is rounded to four decimal places. The timing starts after the user enters the advertisement page, and the timing stops after the user exits the advertisement page to obtain the recorded browsing time. Since the user records the browsing time once every time he browses, the amount of data recorded for the browsing time is too much, so it is not specifically displayed in this embodiment. The average recorded stay time is 10.8s through statistical calculation, and the number of conversions is recorded as 2327. The record conversion rate is calculated to be 0.1960, that is, 19.60%, and the calculation result is rounded to four decimal places;

[0066] See also Figure 2 As shown, the feedback data cleaning unit is used to remove invalid data in the feedback data;

[0067] The feedback data cleaning unit is configured with a feedback data cleaning strategy, which includes:

[0068] Obtain the recorded browsing time, compare the recorded browsing time with the first valid time, and if the recorded browsing time is less than or equal to the first valid time, output an invalid browsing signal; if the recorded browsing time is greater than the first valid time, output a valid browsing signal;

[0069] If a valid browsing signal is output, the recorded browsing time is marked as the valid time, and the effective average stay time of the web page advertisement is recalculated based on the valid time;

[0070] Mark the number of effective durations as effective pageviews, and recalculate the effective click rate and effective conversion rate based on the effective pageviews;

[0071] The effective page views, effective click rate, effective conversion rate and effective average stay time are collectively referred to as effective feedback data;

[0072] In actual applications, after a large number of social experiments, it is finally concluded that when a user enters a page by accidental touch or erroneous operation, the user will exit the page within two to three seconds. Taking into account the network speed and reaction and other factors, the first effective time is set to 5s to distinguish between users who accidentally touch and enter the advertising page and users who click to enter the advertising page by themselves; the recorded browsing time is obtained to be 10s, and through comparison, it is found that the recorded browsing time is greater than the first effective time, and a valid browsing signal is output, and the recorded browsing time recorded this time is marked as the effective time. After analyzing all the recorded browsing times, 9096 effective times are obtained, that is, the effective browsing volume is 9096, and the average value is recalculated to obtain an effective average stay time of 15.8s. Then the effective 9096 / 86544 is calculated, and the effective click rate is 0.1051, that is, 10.51%, and the calculation result is rounded to four decimal places. Calculate 2327 / 9096 and the effective conversion rate is 0.2558, that is, 25.58%, and the calculation result is rounded to four decimal places.

[0073] The version delivery evaluation module is used to evaluate the delivery effect of the advertising version of the web page advertisement, and is named as version delivery effect;

[0074] The version delivery evaluation module is configured with a version delivery evaluation strategy, which includes:

[0075] Valid feedback data is calculated separately for different ad versions;

[0076] For any advertisement version, obtain the corresponding valid feedback data and mark it as data to be evaluated;

[0077] The delivery effect calculation formula is used to calculate the data to be evaluated to obtain the delivery effect of the advertising version;

[0078] The calculation formula for delivery effect is configured as follows: ; Where Vae is the version delivery effect, CT is the effective click-through rate, CR is the effective conversion rate, TA is the effective average stay time, and α is the duration parameter;

[0079] In actual applications, advertising versions include video version, pop-up version and grid version. The grid version has the same format as other products in the shopping interface. This embodiment takes the video version of the web page advertisement as an example to calculate the version delivery effect of the video version, which is named video delivery effect. The calculation process of the version delivery effects of other versions is the same as the calculation process of the video delivery effect, so no specific demonstration is performed in this embodiment. The effective click rate is 10.51%, the effective conversion rate is 25.58%, and the effective average stay time is 15.8s. The setting of α is to make the data type and data size of the effective average stay time similar to the effective click rate and the effective conversion rate, so as to balance the influence of the three on the version delivery effect. The setting of the constant 10000 is to make the calculation result easier to observe and has no practical significance. In this embodiment, α is set to 0.01, and the video delivery effect is calculated to be 42.48, and the calculation result is rounded to two decimal places. Similarly, the pop-up delivery effect is calculated to be 28.98, and the grid delivery effect is 39.72.

[0080] The audience delivery evaluation module is used to divide the audience groups of web page advertisements according to age, and calculate the optimal delivery ratio of web page advertisements based on the feedback data of web page advertisements in different audience ages; the audience delivery evaluation module includes a delivery group selection unit and a delivery ratio analysis unit;

[0081] The delivery group selection unit is configured with delivery group selection strategies, which include:

[0082] A first age stage, a second age stage, a third age stage and a fourth age stage are set, and a first audience group, a second audience group, a third audience group and a fourth audience group are obtained based on the first age stage, the second age stage, the third age stage and the fourth age stage;

[0083] In actual applications, teenagers under the age of 12 usually do not have consumption power, so the audience group under the age of 12 is not considered. Teenagers aged 12 to 18 are usually junior high school students and can have certain consumption power with the consent of their parents. Therefore, [12,18) is regarded as the first age stage, that is, the first audience group; the consumption power of adults aged 18 to 30 is similar. Adults aged 18 to 22 are usually studying in school. Although they do not have large consumption power, they are similar to those aged 22 to 30 in general commodities. Therefore, [18,30) is divided into the second age stage, that is, the second audience group; most adults aged 30 to 55 have entered the palace of marriage, and their consumption at this time is mostly linked to the family, so [30,55) is divided into the third age stage, that is, the third audience group; the remaining adults over 55 are divided into the fourth age stage, that is, the fourth audience group;

[0084] Obtaining valid feedback data corresponding to the first audience group, the second audience group, the third audience group, and the fourth audience group, respectively naming them as first feedback data, second feedback data, third feedback data, and fourth feedback data;

[0085] Calculate the product of the effective click rate and the effective conversion rate in the first feedback data, and name the calculation result as the first delivery effect; calculate the product of the effective click rate and the effective conversion rate in the second feedback data, and name the calculation result as the second delivery effect; calculate the product of the effective click rate and the effective conversion rate in the third feedback data, and name the calculation result as the third delivery effect; calculate the product of the effective click rate and the effective conversion rate in the fourth feedback data, and name the calculation result as the fourth delivery effect; mark the first delivery effect, the second delivery effect, the third delivery effect, and the fourth delivery effect as audience delivery effects;

[0086] Sort the audience delivery effects from largest to smallest, obtain the audience groups corresponding to the top two audience delivery effects, and mark them as target delivery groups;

[0087] Analyze the effective feedback data of the target advertising group and calculate the optimal advertising ratio for web pages;

[0088] In actual application, the effective click rate obtained from the first feedback data is 20.88%, the effective conversion rate is 3.73%, and the first delivery effect is calculated to be 0.007788. The calculation result is retained for 6 decimal places. To facilitate data observation, in this embodiment, the audience delivery effects are multiplied by 10000, and the first delivery effect is 77.88; the effective click rate obtained in the second feedback data is 10.72%, and the effective conversion rate is 15.68%, and the second delivery effect is calculated to be 168.09; the effective click rate obtained in the third feedback data is 18.26%, and the effective conversion rate is 10.85%, and the third delivery effect is calculated to be 198.12; the effective click rate obtained in the fourth feedback data is 12.74%, and the effective conversion rate is 5.65%, and the fourth delivery effect is calculated to be 71.98; the audience delivery effects of the top two are obtained by sorting and comparison, which are 198.12 and 168.09, that is, the second audience group and the third audience group are marked as target delivery groups; since the product has applicability, that is, the product has a certain applicable age, according to a large number of experiments and data, the applicable population span of most products will not exceed two audience groups, so the audience group groups corresponding to the top two audience delivery effects are selected as the target delivery groups for analysis;

[0089] The delivery ratio analysis unit is configured with delivery ratio analysis strategies, which include:

[0090] The target delivery groups are numbered and represented by symbols T1 and T2;

[0091] Get the audience delivery effects corresponding to T1 and T2, represented by symbols S1 and S2, where S1 is greater than S2;

[0092] Get the preset delivery volume and expected conversion volume of the page advertisement. The preset delivery volume is the number of times the page advertisement needs to be delivered preset in the platform; the expected conversion volume is the number of orders expected by the merchant to which the page advertisement belongs;

[0093] Set the increase parameter and construct the equation ; Where Ecr is the expected conversion volume, Pip is the increase parameter, PsV is the preset delivery volume, a is the delivery ratio of T1, and b is the delivery ratio of T2;

[0094] Solve the equation to get a and b, which is the optimal placement ratio;

[0095] In actual applications, the third audience group is marked as T1, and the second audience group is marked as T2. The audience delivery effect in the equation is the original value, not the increased value, and S1 is 0.019812 and S2 is 0.016809. When signing an advertising contract with the platform, the company to which the advertisement belongs usually negotiates the revenue that the company to which the advertisement belongs expects to bring from promoting the advertisement on the platform, that is, the expected conversion volume. The platform has a large number of advertisements, so the total delivery volume of each advertisement is set, that is, the preset delivery volume. At the same time, while completing the task of the expected conversion volume, it is necessary to promote the web ads to a certain extent, and all web ads are delivered to one audience group. Although the profit can be greater than that of targeting two audience groups, it cannot play a promotional role and cannot expand the market. Therefore, when considering the placement of advertisements, we should not only care about the profit, but also the promotion of advertisements. The purpose of setting the increase parameter is to balance the relationship between profit and promotion. The larger the increase parameter, the more profit, but the less promotion among different audience groups; the smaller the increase parameter, the less profit, but the greater the promotion among different audience groups. The value range is 1 to 1.5. The specific value can be formulated according to the actual situation. In this embodiment, the increase parameter is set to 1.2; the preset delivery volume Psv is 200000, and the expected conversion volume is 2000. Substituting it into the equation, we get , solving the system of equations yields a=0.71, b=0.29.

[0096] The advertising delivery optimization module is used to optimize the advertising delivery strategy based on the version delivery effect and the best delivery ratio;

[0097] The advertising delivery optimization module is configured with advertising delivery optimization strategies, which include:

[0098] Get the maximum value among the version delivery effects and mark it as the best delivery version;

[0099] Deliver page ads to the target audience according to the best delivery version and deliver them according to the best delivery ratio;

[0100] In actual applications, the ratio of the second audience group is 0.29, and the ratio of the third audience group is 0.71. Therefore, page ads are delivered to the second audience group and the third audience group at a ratio of 29% and 71% respectively.

[0101] Example 2, please refer to Figure 3 As shown, the present application provides an optimization analysis method based on web page advertisement feedback data, comprising the following steps:

[0102] Step S1, collecting feedback data of web page advertisements and cleaning the feedback data; Step S1 includes the following sub-steps:

[0103] Step S101, collecting feedback data of web page advertisements;

[0104] Step S101 includes the following sub-steps:

[0105] Step S1011, when a user clicks on a webpage advertisement to enter the advertisement page, a record of page views is recorded, and the number of times the webpage advertisement is pushed is counted, marked as push volume, and the record click rate of the webpage advertisement is calculated by dividing the push volume by the record of page views;

[0106] Step S1012, when the user enters the advertisement page, the duration of the user browsing the advertisement page is recorded and marked as the recorded browsing duration. The recorded browsing duration is the duration from the user entering the advertisement page to the user exiting the advertisement page. The average value of the recorded browsing duration is calculated and named as the recorded average stay duration.

[0107] Step S1013, record the number of times users purchase products through the advertisement page, named as the number of conversions, and divide the number of recorded views by the number of conversions to calculate the record conversion rate of the webpage advertisement;

[0108] Step S1014, recording the page views, recording the click rate, recording the average stay time and recording the conversion rate are collectively referred to as feedback data;

[0109] Step S102, eliminating invalid data in the feedback data;

[0110] Step S102 includes the following sub-steps:

[0111] Step S1021, obtaining the recorded browsing time, comparing the recorded browsing time with the first effective time, if the recorded browsing time is less than or equal to the first effective time, outputting an invalid browsing signal; if the recorded browsing time is greater than the first effective time, outputting a valid browsing signal;

[0112] Step S1022: if a valid browsing signal is output, the recorded browsing duration is marked as a valid duration, and the effective average stay duration of the web page advertisement is recalculated based on the valid duration;

[0113] Step S1023, marking the number of effective durations as effective page views, and recalculating effective click rate and effective conversion rate based on the effective page views;

[0114] Step S1024, the effective page views, effective click rate, effective conversion rate and effective average stay time are collectively referred to as effective feedback data;

[0115] Step S2, evaluating the delivery effect of the advertisement version of the webpage advertisement, which is named as version delivery effect; Step S2 includes the following sub-steps:

[0116] Step S201, valid feedback data is calculated separately for different advertisement versions;

[0117] Step S202, for any advertisement version, obtain corresponding valid feedback data and mark it as data to be evaluated;

[0118] Step S203, calculating the data to be evaluated using a delivery effect calculation formula to obtain a delivery effect of the advertisement version;

[0119] The calculation formula for delivery effect is configured as follows: ; Where Vae is the version delivery effect, CT is the effective click-through rate, CR is the effective conversion rate, TA is the effective average stay time, and α is the duration parameter;

[0120] Step S3, dividing the audience groups of the web page advertisement according to age, and calculating the optimal placement ratio of the web page advertisement based on the feedback data of the web page advertisement among audiences of different ages; Step S3 includes the following sub-steps:

[0121] Step S301, setting a first age stage, a second age stage, a third age stage and a fourth age stage, and obtaining a first audience group, a second audience group, a third audience group and a fourth audience group based on the first age stage, the second age stage, the third age stage and the fourth age stage;

[0122] Step S302, obtaining valid feedback data corresponding to the first audience group, the second audience group, the third audience group, and the fourth audience group, respectively named as first feedback data, second feedback data, third feedback data, and fourth feedback data;

[0123] Step S303, calculating the product between the effective click rate and the effective conversion rate in the first feedback data, and naming the calculation result as the first delivery effect; calculating the product between the effective click rate and the effective conversion rate in the second feedback data, and naming the calculation result as the second delivery effect; calculating the product between the effective click rate and the effective conversion rate in the third feedback data, and naming the calculation result as the third delivery effect; calculating the product between the effective click rate and the effective conversion rate in the fourth feedback data, and naming the calculation result as the fourth delivery effect; marking the first delivery effect, the second delivery effect, the third delivery effect, and the fourth delivery effect as audience delivery effects;

[0124] Step S304, sorting the audience delivery effects in descending order, obtaining the audience groups corresponding to the top two audience delivery effects, and marking them as target delivery groups;

[0125] Step S305, analyzing the effective feedback data of the target delivery group and calculating the optimal delivery ratio of the web page advertisement;

[0126] Step S305 includes the following sub-steps:

[0127] Step S3051, numbering the target delivery groups, represented by symbols T1 and T2;

[0128] Step S3052, obtaining the audience delivery effects corresponding to T1 and T2, represented by symbols S1 and S2;

[0129] Step S3053, obtaining the preset delivery volume and expected conversion volume of the page advertisement, wherein the preset delivery volume is the number of times the page advertisement needs to be delivered preset in the platform; and the expected conversion volume is the number of orders expected by the merchant to which the page advertisement belongs;

[0130] Step S3054, set the increase parameter and construct the equation ; Where Ecr is the expected conversion volume, Pip is the increase parameter, PsV is the preset delivery volume, a is the delivery ratio of T1, and b is the delivery ratio of T2;

[0131] Step S3055, solving the equation to obtain a and b, which is the optimal delivery ratio;

[0132] Step S4, optimizing the advertising delivery strategy based on the version delivery effect and the best delivery ratio; Step S4 includes the following sub-steps:

[0133] Step S401, obtaining the maximum value among the version delivery effects and marking it as the best delivery version;

[0134] Step S402: delivering the page advertisement to the target delivery group according to the best delivery version, and delivering the advertisement according to the best delivery ratio.

[0135] Example 3, please refer to Figure 4 As shown, Figure 4 The structural diagram of an electronic device is illustrated, and the electronic device may include: a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other through the communication bus. The memory stores computer-readable instructions, and the processor may call the instructions in the memory. When the computer-readable instructions are executed by the processor, the steps in the optimization analysis method based on web page advertisement feedback data are executed to achieve the following functions: collect feedback data of web page advertisements, and perform data cleaning on the feedback data at the same time; evaluate the delivery effect of the advertisement version of the web page advertisement, which is named version delivery effect; divide the audience groups of the web page advertisements according to age, and calculate the optimal delivery ratio of the web page advertisements based on the feedback data of the web page advertisements in different audience ages; optimize the advertisement delivery strategy based on the version delivery effect and the optimal delivery ratio.

[0136] In addition, the logic instructions in the above-mentioned memory can be implemented in the form of software functional units and can be stored in a computer-readable storage medium when sold or used as an independent product. Based on this understanding, the technical solution of the present application can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium, including several instructions to enable a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc. Various media that can store program codes.

[0137] Embodiment 4, the present application also provides a computer-readable storage medium, the present application provides a storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps in the optimization analysis method based on web page advertising feedback data are executed to achieve the following functions: collect feedback data of web page advertisements, and clean the feedback data at the same time; evaluate the delivery effect of the advertising version of the web page advertisement, which is named as version delivery effect; divide the audience groups of the web page advertisements according to age, and calculate the optimal delivery ratio of the web page advertisements based on the feedback data of the web page advertisements in different audience ages; optimize the advertising delivery strategy based on the version delivery effect and the optimal delivery ratio.

[0138] Through the description of the above implementation methods, the embodiments of the present invention can be provided as methods, systems or computer program products. Based on such an understanding, the above technical solutions can be essentially or partly contributed to the prior art in the form of software products, which can be stored in a computer-readable storage medium, such as ROM / RAM, a magnetic disk, an optical disk, etc., and include several instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0139] In the embodiments provided in the present application, it should be understood that the disclosed system or method can be implemented in other ways. The embodiments described above are merely schematic. For example, the division of modules or units is only a logical function division. There may be other division methods in actual implementation. For example, multiple modules or units can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some communication interfaces, and the indirect coupling or communication connection of systems, modules and units can be electrical, mechanical or other forms.

[0140] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit it. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. An optimization analysis method based on web page advertising feedback data, characterized in that: The steps include: Collect feedback data of web page advertisements and clean the feedback data; Evaluate the delivery effect of the advertising versions of web ads, named version delivery effect; Divide the audience groups of web ads by age, and calculate the optimal placement ratio of web ads based on the feedback data of web ads among audiences of different ages; Optimize advertising delivery strategies based on version delivery effects and optimal delivery ratios; Collecting feedback data of web page advertisements and cleaning the feedback data include the following sub-steps: Collect feedback data about web ads; Eliminate invalid data from feedback data; Collecting feedback data for web ads includes the following sub-steps: When a user clicks on a web ad and enters the ad page, a record pageview is recorded, and the number of times the web ad is pushed is counted, marked as the push volume. The record click-through rate of the web ad is calculated by dividing the push volume by the record pageview volume. When a user enters an advertisement page, the length of time the user browses the advertisement page is recorded and marked as recorded browsing time. The recorded browsing time is the length of time from when the user enters the advertisement page to when the user exits the advertisement page. The average value of the recorded browsing time is calculated and named as recorded average stay time. Record the number of times users purchase products through the ad page, named as the number of conversions. Divide the number of recorded views by the number of conversions to calculate the record conversion rate of the web ad. The recorded page views, recorded click-through rates, recorded average dwell time and recorded conversion rates are collectively referred to as feedback data; Eliminating invalid data from feedback data includes the following sub-steps: Obtain the recorded browsing time, compare the recorded browsing time with the first valid time, and if the recorded browsing time is less than or equal to the first valid time, output an invalid browsing signal; if the recorded browsing time is greater than the first valid time, output a valid browsing signal; If a valid browsing signal is output, the recorded browsing time is marked as the valid time, and the effective average stay time of the web page advertisement is recalculated based on the valid time; Mark the number of effective durations as effective pageviews, and recalculate the effective click rate and effective conversion rate based on the effective pageviews; The effective page views, effective click rate, effective conversion rate and effective average stay time are collectively referred to as effective feedback data; Evaluating the delivery effect of the advertising version of a web page advertisement, named version delivery effect, includes the following sub-steps: The effective feedback data is calculated separately for different advertisement versions; For any advertisement version, obtain the corresponding valid feedback data and mark it as data to be evaluated; The delivery effect calculation formula is used to calculate the data to be evaluated to obtain the delivery effect of the advertising version; The delivery effect calculation formula is configured as follows: ; Where Vae is the version delivery effect, CT is the effective click-through rate, CR is the effective conversion rate, TA is the effective average stay time, and α is the duration parameter; Divide the audience groups of web page advertisements by age, and calculate the optimal placement ratio of web page advertisements based on the feedback data of web page advertisements in different audience ages, including the following sub-steps: A first age stage, a second age stage, a third age stage and a fourth age stage are set, and a first audience group, a second audience group, a third audience group and a fourth audience group are obtained based on the first age stage, the second age stage, the third age stage and the fourth age stage; Obtaining valid feedback data corresponding to the first audience group, the second audience group, the third audience group, and the fourth audience group, respectively naming them as first feedback data, second feedback data, third feedback data, and fourth feedback data; Calculate the product of the effective click rate and the effective conversion rate in the first feedback data, and name the calculation result as the first delivery effect; calculate the product of the effective click rate and the effective conversion rate in the second feedback data, and name the calculation result as the second delivery effect; calculate the product of the effective click rate and the effective conversion rate in the third feedback data, and name the calculation result as the third delivery effect; calculate the product of the effective click rate and the effective conversion rate in the fourth feedback data, and name the calculation result as the fourth delivery effect; mark the first delivery effect, the second delivery effect, the third delivery effect, and the fourth delivery effect as audience delivery effects; Sort the audience delivery effects from largest to smallest, obtain the audience groups corresponding to the top two audience delivery effects, and mark them as target delivery groups; Analyze the effective feedback data of the target advertising group and calculate the optimal advertising ratio for web pages; The analysis of effective feedback data from the target audience includes the following sub-steps: The target delivery groups are numbered and represented by symbols T1 and T2; Get the audience delivery effects corresponding to T1 and T2, represented by symbols S1 and S2; Obtaining the preset delivery volume and expected conversion volume of the page advertisement, wherein the preset delivery volume is the number of times the page advertisement needs to be delivered preset in the platform; the expected conversion volume is the number of orders expected by the merchant to which the page advertisement belongs; Set the increase parameter and construct the equation ; Where Ecr is the expected conversion volume, Pip is the increase parameter, Psv is the preset delivery volume, a is the delivery ratio of T1, and b is the delivery ratio of T2; Solving the equation to get a and b is the optimal delivery ratio.

2. The optimization analysis method based on web page advertisement feedback data according to claim 1, characterized in that: The optimization of advertising delivery strategy based on the version delivery effect and the best delivery ratio includes the following sub-steps: Get the maximum value among the version delivery effects and mark it as the best delivery version; Deliver page ads to the target audience according to the best delivery version, and deliver them according to the best delivery ratio.

3. An optimization analysis system based on web page advertisement feedback data, used to implement the optimization analysis method based on web page advertisement feedback data according to claim 1 or 2, characterized in that: It includes a feedback data collection module, a version delivery evaluation module, an audience delivery evaluation module and an advertisement delivery optimization module; the feedback data collection module, the version delivery evaluation module and the audience delivery evaluation module are respectively connected with the advertisement delivery optimization module data; The feedback data collection module is used to collect feedback data of web page advertisements and clean the feedback data; The version delivery evaluation module is used to evaluate the delivery effect of the advertisement version of the webpage advertisement, which is named as version delivery effect; The audience delivery evaluation module is used to divide the audience groups of web page advertisements according to age, and calculate the optimal delivery ratio of web page advertisements based on the feedback data of web page advertisements among audiences of different ages; The advertisement delivery optimization module is used to optimize the advertisement delivery strategy based on the version delivery effect and the best delivery ratio.

4. A storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps in the method according to claim 1 or 2 are executed.

Images