Screening method of advertisement putting sites

Abstract

The invention relates to the technical field of advertisement putting site screening methods, in particular to an advertisement putting site screening method, which specifically comprises the following steps of 1, collecting multi-dimensional data related to advertisement putting sites; step 2, carrying out cleaning, conversion and normalization processing on the collected multi-dimensional data; step 3, constructing an advertisement putting site evaluation index system; 4, index weights are determined, and the weights of all indexes in the evaluation index system are determined through an analytic hierarchy process; according to the method, in the data level, multi-dimensional data collection and fine preprocessing are carried out, it is ensured that data are comprehensive, accurate and standard, evaluation errors caused by data quality problems are avoided, and a solid and reliable foundation is provided for subsequent analysis. A constructed evaluation index system and a scientific weight determination method comprehensively consider key elements of advertisement putting, subjective randomness and limitation of a single index are avoided, site values can be accurately evaluated, and screening results are more objective and credible.

Description

Technical Field

[0001] The invention relates to the technical field of angle iron line drawing equipment, in particular to a method for screening advertisement delivery sites. Background Art

[0002] In the digital information age, advertising is an important means for companies to promote products and services and enhance brand awareness. The effectiveness of advertising directly affects the company's market competitiveness and economic benefits.

[0003] Traditional methods for screening ad sites rely primarily on manual experience and simple statistical analysis. For example, based on past experience, advertisers subjectively judge the degree of compatibility between the audience of a particular site and the target audience. This approach has numerous limitations: First, manual experience is highly subjective, and different advertisers may come up with different screening results, resulting in a lack of objectivity and accuracy. Second, simple statistical indicators cannot fully reflect the advertising value of a site. For example, focusing solely on visit volume may overlook important factors such as the site's audience's spending power and interests, resulting in ads failing to accurately reach the target audience and wasting advertising resources. Summary of the Invention

[0004] In view of the technical problem that the traditional method for screening advertisement delivery sites mainly relies on manual experience and simple statistical analysis and has many limitations, the present invention provides a method for screening advertisement delivery sites.

[0005] The technical solution adopted by the present invention is: a method for screening advertising sites, which specifically includes the following steps:

[0006] Step 1: Collect multi-dimensional data related to the advertising delivery site;

[0007] Step 2: Clean, transform and normalize the collected multi-dimensional data;

[0008] Step 3: Build an evaluation index system for advertising sites;

[0009] Step 4: Determine the indicator weights. Use the analytic hierarchy process to determine the weights of each indicator in the evaluation indicator system.

[0010] Step 5: Calculate the comprehensive score of the advertising site;

[0011] Step 6: Filter the advertising sites.

[0012] In one embodiment, in step 1, the multi-dimensional data related to the advertisement delivery site is collected, including but not limited to the following categories:

[0013] Traffic data: the site's average daily visits (V), average monthly visits (MV), time distribution of visits, page views (PV), unique visitors (UV), etc.

[0014] Audience data: audience age distribution, gender ratio, geographical distribution, occupation distribution, and interest preferences;

[0015] Advertising history data: the type, quantity, delivery time period, advertiser information, ad click-through rate (CTR), and conversion rate (CR) of advertisements previously delivered by the site.

[0016] In one embodiment, in step 2, the specific steps of cleaning, converting, and normalizing the collected multi-dimensional data are as follows:

[0017] Data cleaning: remove duplicate data, erroneous data and missing values;

[0018] Data transformation: converting unstructured or semi-structured data into structured data;

[0019] Data normalization: converting data of different dimensions into the same numerical range.

[0020] In one embodiment, in step 3, the evaluation index system for advertisement delivery sites is constructed as follows:

[0021] The advertising site evaluation index system includes indicators in the following dimensions:

[0022] Depth of visit D: It is used to measure the depth of user browsing behavior within the site. The calculation formula is:

[0023]

[0024] Among them, n represents the number of days in the statistical period, PV i represents the page views on day i, UV represents the number of unique visitors;

[0025] Access frequency F: reflects the user's dependence on the site, and the calculation formula is:

[0026]

[0027] Among them, V i represents the number of visits on day i;

[0028] Target audience matching M: used to evaluate the matching degree between the site audience and the advertising target audience;

[0029] Historical average click-through rate (CTR) of advertisements avg , reflecting the site's ability to attract advertisements, and the calculation formula is:

[0030]

[0031] Among them, m represents the number of advertisements previously placed by the site, CTR i represents the click-through rate of the i-th advertisement;

[0032] Historical advertising average conversion rate CR avg : Calculate the average conversion rate of the site's previous ads. The calculation formula is:

[0033]

[0034] Among them, CR i represents the conversion rate of the i-th advertisement;

[0035] Competition for the same type of ads C: This evaluates the intensity of competition for the same type of ads on the site. The calculation formula is:

[0036]

[0037] Among them, N s Indicates the number of ads of the same type as the one to be placed on the site, N t Indicates the total number of ads on the site.

[0038] In one embodiment, in step 4, determining the indicator weights by using the analytic hierarchy process to determine the weights of each indicator in the evaluation indicator system is performed in the following specific steps:

[0039] Construct a hierarchical model: decompose the problem of advertising site screening into the target layer, the criterion layer, and the plan layer;

[0040] Construct a judgment matrix: Compare the importance of each indicator in the criterion layer relative to the target layer and construct a judgment matrix.

[0041] In one embodiment, in step 5, the specific method for calculating the comprehensive score of the advertisement delivery site is as follows:

[0042] Based on the constructed evaluation index system and the determined index weights, the comprehensive score S of each advertising delivery site is calculated. The calculation formula is S;

[0043]

[0044] Among them, p represents the number of evaluation indicators, W i represents the weight of the i-th indicator, X i Represents the normalized value of the i-th indicator.

[0045] The beneficial effects of this invention are as follows: Compared with existing technologies, at the data level, this invention utilizes multi-dimensional data collection and meticulous preprocessing to ensure comprehensive, accurate, and standardized data, avoiding evaluation errors caused by data quality issues and providing a solid and reliable foundation for subsequent analysis. The constructed evaluation index system and scientific weighting method comprehensively consider the key factors of advertising delivery, avoiding subjectivity and the limitations of a single indicator, accurately assessing site value, and making screening results more objective and credible. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 It is a schematic flow diagram of the present invention; DETAILED DESCRIPTION

[0047] In the description of the present invention, it should be noted that the terms "front", "up", "down", "left", "right", "vertical", "horizontal", etc., indicating orientations or positional relationships, are based on the orientations or positional relationships shown in the accompanying drawings. They are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation. Therefore, they cannot be understood as limiting the present invention.

[0048] refer to Figure 1 In order to solve the problems existing in the background technology, the present application proposes the following technical solution: a method for screening advertising sites, specifically comprising the following steps:

[0049] Step 1: Collect multi-dimensional data related to the advertising delivery site;

[0050] In step 1, collect multi-dimensional data related to the advertising delivery site, including but not limited to the following categories:

[0051] Traffic data: the site's average daily visits V, average monthly visits MV, and time distribution of visits (such as visits V in different time periods). t , t represents time), page views PV, number of independent visitors UV, etc.

[0052] Audience data: audience age distribution (percentage of people in each age group i , i represents age group), gender ratio (male proportion G m,女性占比 G f ), regional distribution (ratio of visits in each region j , j represents region), occupational distribution (the proportion of visits by each occupational group O k , k represents occupation), interest preference (the preference degree of each interest category obtained by mining user behavior data I l , l represents interest category) etc.

[0053] Historical advertising data: The site's historical advertising data includes the type, quantity, timeframe, advertiser information, click-through rate (CTR), and conversion rate (CR). This data can be collected through various channels, such as the site's own statistical data interface, third-party data monitoring platforms, and user surveys. This collected data is stored in a database, providing a foundation for subsequent analysis and screening. This comprehensive data collection covers a wide range of multi-dimensional data from advertising sites, including basic information, traffic, audience, and advertising history. This comprehensive data provides a solid foundation for subsequent analysis, avoiding evaluation biases caused by missing data and ensuring that screening results are based on authentic, rich data support, ensuring that the screening results are aligned with actual needs.

[0054] Step 2: Clean, transform and normalize the collected multi-dimensional data;

[0055] In step 2, the specific steps for cleaning, converting, and normalizing the collected multi-dimensional data are as follows:

[0056] Data cleaning: Remove duplicate data, erroneous data, and missing values; for the treatment of missing values, different methods are used according to the characteristics of the data. If the number of missing values ​​is small, the mean, median, or mode can be used to fill in the missing values; if the number of missing values ​​is large and the data has a significant impact on subsequent analysis, consider re-collecting the data or using machine learning algorithms for prediction and filling. For example, for the missing values ​​of the site's average daily visits V, if the number of missing values ​​is small, the average daily visits of other similar sites can be calculated to fill in the missing values. If the number of missing values ​​is large, other data related to the visits (such as page views PV, unique visitors UV, etc.) can be used to predict the missing visits through a linear regression model.

[0057] Data conversion: converting unstructured or semi-structured data into structured data for subsequent analysis and processing. For example, converting user interest preference data from text descriptions into numerical preference values. l .

[0058] Data normalization: converting data of different dimensions to the same numerical range to eliminate the influence of dimensions on data analysis. Common normalization methods include minimum-maximum normalization and Z-score normalization. Here we use the minimum-maximum normalization method. For data x, the normalized result x ′ The calculation formula is:

[0059]

[0060] Among them, x min and x maxare the minimum and maximum values ​​in the data column, respectively. For example, for a site's average daily visits, V, the above formula is used to normalize it to the range [0, 1], making visits of different magnitudes comparable and facilitating subsequent comprehensive analysis. Data quality is improved through cleaning, conversion, and normalization. Invalid data is removed, and data formats and dimensions are standardized, making the data more standardized and comparable. This effectively prevents data issues from impacting analysis accuracy, laying a solid foundation for subsequent, accurate evaluation of the value of advertising sites.

[0061] Step 3: Build an evaluation index system for advertising sites;

[0062] In step 3, the evaluation index system for advertising delivery sites is constructed as follows:

[0063] The advertising site evaluation index system includes indicators in the following dimensions:

[0064] Depth of visit D: It is used to measure the depth of user browsing behavior within the site. The calculation formula is:

[0065]

[0066] Among them, n represents the number of days in the statistical period, PV i represents the number of page views on day i, and UV represents the number of unique visitors. The greater the visit depth D, the longer the user stays on the site, the higher their interest in the site content, and the better the site's traffic quality.

[0067] Access frequency F: reflects the user's dependence on the site, and the calculation formula is:

[0068]

[0069] Among them, V i The higher the access frequency F, the more frequently users visit the site, and the stronger the user stickiness of the site.

[0070] Target audience matching degree M: used to evaluate the matching degree between the site audience and the advertising target audience. Assume that the age distribution of the advertising target audience is A t,i , the gender ratio is G t,m and G t,f , the geographical distribution is R t,j , occupation distribution is O t,k , interest preference is I t,l , then the calculation formula for the target audience matching degree M is:

[0071] M=α∑ i |A t,i -A i |+β|G t,m -Gm |+γ|G t,f -G f |+δ∑ j |R t,j -R j |+∈∑ k |O t,k -O k |+ζ∑ l |I t,l -I l |;

[0072] Where α, β, γ, δ, ∈, and ζ are weight coefficients, set based on the advertiser's emphasis on different audience characteristics, with α + β + γ + δ + ∈ + ζ = 1. The smaller the value of M, the more closely the site's audience matches the ad's target audience, and the more suitable the site is for the ad.

[0073] Historical average click-through rate (CTR) of advertisements avg :The average click-through rate of ads previously placed on the site is calculated to reflect the site's ability to attract ads. The calculation formula is:

[0074]

[0075] Among them, m represents the number of advertisements previously placed by the site, CTR i Indicates the click-through rate of the i-th advertisement. CTR avg The higher it is, the more likely it is that the ads on the site will attract users to click on them, and the more likely the advertising effect will be.

[0076] Historical advertising average conversion rate CR avg : Calculate the average conversion rate of ads previously placed on the site to measure the conversion effect of ads on the site. The calculation formula is:

[0077]

[0078] Among them, CR i Indicates the conversion rate of the i-th advertisement. avg The higher it is, the more effectively advertising on this site can achieve advertising goals (such as product sales, user registration, etc.).

[0079] Competition for the same type of ads C: This evaluates the intensity of competition for the same type of ads on the site. The calculation formula is:

[0080]

[0081] Among them, N s Indicates the number of ads of the same type as the one to be placed on the site, N tRepresents the total number of ads on the site. A larger C value indicates more intense competition among similar ads on the site, making it more difficult for new ads to attract user attention.

[0082] The above technical solution builds metrics based on traffic quality, audience matching, advertising effectiveness, and the competitive landscape. This comprehensively considers key factors in advertising, avoiding the limitations of a single metric. This allows for a more accurate assessment of site value, helping advertisers find sites that align with their target audience, deliver high advertising effectiveness, and maintain appropriate competition.

[0083] Step 4: Determine the indicator weights. Use the analytic hierarchy process to determine the weights of each indicator in the evaluation indicator system.

[0084] In step 4, determine the indicator weights using the analytic hierarchy process to determine the weights of each indicator in the evaluation indicator system. The specific steps are as follows:

[0085] Constructing a hierarchical model: Decomposing the problem of ad placement site screening into the target layer (selecting the optimal ad placement site), the criteria layer (traffic quality indicators, audience matching indicators, ad effectiveness indicators, and competitive environment indicators), and the solution layer (each ad placement site);

[0086] Constructing a judgment matrix: Invite advertising industry experts, data analysts and other professionals to compare the importance of each indicator in the criterion layer relative to the target layer, and construct a judgment matrix; the element a of the judgment matrix ij It indicates the importance of indicator i relative to indicator j, and its value range is 1-9, where 1 means that the two indicators are equally important, 3 means that indicator i is slightly more important than indicator j, 5 means that indicator i is obviously more important than indicator j, 7 means that indicator i is strongly more important than indicator j, and 9 means that indicator i is extremely more important than indicator j. 2, 4, 6, and 8 are the median values ​​of the above adjacent judgments.

[0087] For example, if the traffic quality indicator is considered slightly more important than the audience matching indicator, then the element a at the corresponding position in the judgment matrix 12 =3, 3. Calculate the weight vector: Perform a consistency test on the judgment matrix. If it passes the consistency test (it is generally believed that when the consistency ratio CR<0.1, the judgment matrix has satisfactory consistency), the weight vector is calculated using the square root method or the sum-product method. Taking the square root method as an example, the calculation steps are as follows: - Calculate the judgment matrix A = (a ij ) n×n The product M of each row of elements i :

[0088]

[0089] Calculate M i The nth root of

[0090]

[0091] -Pair vector Normalize and get the weight vector W=(W1,W2,…,W n ) T :

[0092]

[0093] Among them, W i is the weight of each indicator.

[0094] By applying the Analytic Hierarchy Process (AHP) and inviting professionals to determine weights, we transform qualitative judgments into quantitative weights, avoiding subjectivity and arbitrariness. This makes the weights of each indicator more scientific and reasonable, accurately reflects the importance of the indicator in the comprehensive evaluation, and enhances the objectivity and credibility of the screening results.

[0095] Step 5: Calculate the comprehensive score of the advertising site;

[0096] In step 5, the specific method for calculating the comprehensive score of the advertising site is as follows:

[0097] Based on the constructed evaluation index system and the determined index weights, the comprehensive score S of each advertising delivery site is calculated. The calculation formula is S;

[0098]

[0099] Among them, p represents the number of evaluation indicators, W i represents the weight of the i-th indicator, X i Represents the normalized value of the ith indicator. For example, for a certain advertising site, the traffic quality indicator weight is known to be W1, and its access depth D and access frequency F after normalization are X 11 、X 12 , then the comprehensive value of the flow quality index is W1(X 11 +X 12 ), calculate the comprehensive values ​​of other indicators in the same way, and finally add up the comprehensive values ​​of all indicators to get the comprehensive score S of the site. The higher the comprehensive score S, the higher the overall value of the advertising site is and the more suitable it is for advertising.

[0100] The above technical solution integrates multi-dimensional evaluation results by calculating scores based on an indicator system and weights. This simplifies complex site evaluations into quantitative scores, making it easier for advertisers to intuitively compare the pros and cons of various sites, quickly identify high-value sites, and improve screening efficiency and accuracy.

[0101] Step 6: Filter advertising sites

[0102] Based on the calculated comprehensive score S of the advertising site, the sites are ranked from high to low. Advertisers can set a screening threshold T based on their own advertising budget, delivery goals, and other requirements, and select sites with a comprehensive score S ≥ T as candidate advertising sites. For example, if an advertiser wants to maintain a certain level of advertising effectiveness and control advertising costs, they can set the screening threshold T to a moderate value and select sites with a comprehensive score that meets or exceeds this threshold from the ranked sites, ensuring that the selected sites have high advertising value.

[0103] Step 7: Dynamic Optimization and Adjustment

[0104] Because the advertising market is dynamic, site traffic, audience, and advertising effectiveness data will also change over time. Therefore, a dynamic monitoring and optimization mechanism is necessary to regularly (e.g., weekly or monthly) update and re-evaluate the data of selected advertising sites. When a site's overall score falls below threshold T, or when a new, more valuable site emerges (with a higher overall score than an existing site), timely adjustments to the advertising sites are made, replacing the less valuable site with a more suitable one to ensure optimal advertising results.

[0105] In summary:

[0106] At the data level, multi-dimensional data collection and meticulous pre-processing ensure that the data is comprehensive, accurate, and standardized, avoiding evaluation errors caused by data quality issues and providing a solid and reliable foundation for subsequent analysis.

[0107] The constructed evaluation index system and scientific weight determination method comprehensively consider the key factors of advertising delivery, avoid subjective arbitrariness and the limitations of a single indicator, can accurately evaluate the value of the site, and make the screening results more objective and credible.

[0108] In terms of screening decisions, by calculating the comprehensive score and sorting and screening accordingly, complex site evaluations are converted into intuitive quantitative results. Thresholds are set based on advertiser needs, and sites that meet budgets and delivery goals are quickly identified, significantly improving screening efficiency and avoiding waste of resources caused by blind delivery.

[0109] In the face of a dynamic advertising market, a dynamic optimization and adjustment mechanism provides regular updates and assessments, promptly responding to site data fluctuations and shifts in the competitive market. This ensures that ads consistently land on high-value sites, maintains positive results, and enhances the flexibility and adaptability of advertising strategies. This solution effectively addresses many of the issues inherent in traditional screening methods, accurately matching ads with target audiences, improving click-through and conversion rates, and reducing advertising costs.

[0110] While the embodiments of the present invention have been shown and described, it will be apparent to those skilled in the art that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A method for screening advertisement delivery sites, characterized in that: The specific steps include: Step 1: Collect multi-dimensional data related to the advertising delivery site; Step 2: Clean, transform and normalize the collected multi-dimensional data; Step 3: Build an evaluation index system for advertising sites; Step 4: Determine the indicator weights. Use the analytic hierarchy process to determine the weights of each indicator in the evaluation indicator system. Step 5: Calculate the comprehensive score of the advertising site; Step 6: Filter the advertising sites.

2. The method for screening advertisement delivery sites according to claim 1, characterized in that: In step 1, collect multi-dimensional data related to the advertising delivery site, including but not limited to the following categories: Traffic data: the site's average daily visits (V), average monthly visits (MV), time distribution of visits, page views (PV), unique visitors (UV), etc. Audience data: audience age distribution, gender ratio, geographical distribution, occupation distribution, and interest preferences; Advertising history data: the type, quantity, delivery time period, advertiser information, ad click-through rate (CTR), and conversion rate (CR) of advertisements previously delivered by the site.

3. The method for screening advertisement delivery sites according to claim 2, characterized in that: In step 2, the specific steps for cleaning, converting, and normalizing the collected multi-dimensional data are as follows: Data cleaning: remove duplicate data, erroneous data and missing values; Data transformation: converting unstructured or semi-structured data into structured data; Data normalization: converting data of different dimensions into the same numerical range.

4. The method for screening advertisement delivery sites according to claim 3, characterized in that: In step 3, the evaluation index system for advertising delivery sites is constructed as follows: The advertising site evaluation index system includes indicators in the following dimensions: Depth of visit D: It is used to measure the depth of user browsing behavior within the site. The calculation formula is: Among them, n represents the number of days in the statistical period, PV i represents the page views on day i, UV represents the number of unique visitors; Access frequency F: reflects the user's dependence on the site, and the calculation formula is: Among them, V i represents the number of visits on day i; Target audience matching M: used to evaluate the matching degree between the site audience and the advertising target audience; Historical average click-through rate (CTR) of advertisements avg , reflecting the site's ability to attract advertisements, and the calculation formula is: Among them, m represents the number of advertisements previously placed by the site, CTR i represents the click-through rate of the i-th advertisement; Historical advertising average conversion rate CR avg : Calculate the average conversion rate of the site's previous ads. The calculation formula is: Among them, CR i represents the conversion rate of the i-th advertisement; Competition for the same type of ads C: This evaluates the intensity of competition for the same type of ads on the site. The calculation formula is: Among them, N s Indicates the number of ads of the same type as the one to be placed on the site, N t Indicates the total number of ads on the site.

5. The method for screening advertisement delivery sites according to claim 4, characterized in that: In step 4, determine the indicator weights using the analytic hierarchy process to determine the weights of each indicator in the evaluation indicator system. The specific steps are as follows: Construct a hierarchical model: decompose the problem of advertising site screening into the target layer, the criterion layer, and the plan layer; Construct a judgment matrix: Compare the importance of each indicator in the criterion layer relative to the target layer and construct a judgment matrix.

6. The method for screening advertisement delivery sites according to claim 5, characterized in that: In step 5, the specific method for calculating the comprehensive score of the advertising site is as follows: Based on the constructed evaluation index system and the determined index weights, the comprehensive score S of each advertising delivery site is calculated. The calculation formula is S; Among them, p represents the number of evaluation indicators, W i represents the weight of the i-th indicator, X i Represents the normalized value of the i-th indicator.

Images