Advertisement putting area optimization method and system based on multi-dimensional data analysis

Abstract

The application provides an advertisement putting area optimization method and system based on multidimensional data analysis, relates to the technical field of data analysis, and comprises the following steps: performing pre-research based on multiple advertisement putting resource nodes, and obtaining multiple multidimensional research data sets; performing clustering analysis, and demarcating M advertisement putting sub-areas according to a clustering analysis result; performing adaptive analysis, and obtaining an advertisement-area adaptive degree matrix; performing advertisement putting area configuration, performing node-level putting simulation after obtaining an initial advertisement putting scheme, and obtaining M putting simulation data sets; performing global optimization of advertisement putting, and generating a target advertisement putting scheme; and performing advertisement putting management. The application solves the technical problem that the prior art relies on simple audience division and rules, leading to inaccurate configuration of advertisement resources in an area, and further leading to low advertisement effect.

Description

Technical Field

[0001] This invention relates to the field of data analysis technology, specifically to a method and system for optimizing advertising placement areas based on multidimensional data analysis. Background Technology

[0002] The goal of advertising is not only to maximize exposure, but also to ensure precise targeting and high returns. Traditional advertising strategies often rely on simple audience segmentation and basic time slot selection. However, these strategies fail to fully consider the impact of multidimensional data and regional characteristics, leading to inefficient advertising and wasted resources. Specifically, current technologies typically rely on coarse audience segmentation for advertising, such as selecting target regions based on basic demographic data like age and gender. However, this approach ignores multidimensional factors such as geographical features, user behavior, and social interactions within the advertising region, resulting in insufficient precision in advertising targeting. This leads to ineffective reach of the target audience and imprecise allocation of advertising resources within the region, resulting in poor advertising performance. For example, an ad might be placed in a densely populated area but fail to effectively reach high-potential target audiences, resulting in underutilized advertising budgets and low ROI. Summary of the Invention

[0003] This application provides a method and system for optimizing advertising delivery areas based on multidimensional data analysis, aiming to solve the technical problem that existing technologies rely on simple audience segmentation and rules, resulting in insufficient accuracy in the allocation of advertising resources within a region, and consequently, low advertising effectiveness.

[0004] The first aspect disclosed in this application provides a method for optimizing advertising placement areas based on multidimensional data analysis. The method includes: before advertising placement, conducting preliminary research on multiple advertising resource nodes in a predetermined advertising placement area to obtain multiple multidimensional research datasets; performing cluster analysis on the multiple advertising resource nodes based on the multiple multidimensional research datasets, and defining M advertising placement sub-regions according to the cluster analysis results; obtaining N advertising placement requirements for N ads to be placed, performing adaptation analysis on the M advertising placement sub-regions to obtain an ad-region adaptation matrix; configuring advertising placement areas according to the ad-region adaptation matrix, obtaining an initial advertising placement plan, and then performing node-level placement simulation to obtain M placement simulation datasets; performing global optimization of the initial advertising placement plan based on the M placement simulation datasets to generate a target advertising placement plan with recommended advertising placement time slots and recommended advertising placement platforms; and managing advertising placement within the predetermined advertising placement area according to the target advertising placement plan.

[0005] The second aspect of this application discloses an advertising placement area optimization system based on multidimensional data analysis. This system is used in the aforementioned advertising placement area optimization method based on multidimensional data analysis. The system includes: a pre-survey module, used to conduct pre-surveys based on multiple advertising resource nodes in a predetermined advertising placement area before advertising placement, obtaining multiple multidimensional survey datasets; a cluster analysis module, used to perform cluster analysis on the multiple advertising resource nodes based on the multiple multidimensional survey datasets, and delineate M advertising placement sub-regions according to the cluster analysis results; and an adaptation analysis module, used to obtain N advertising placement requirements for N advertisements to be placed, and perform... The system performs adaptation analysis on M advertising sub-regions to obtain an ad-region adaptation matrix; a node-level advertising simulation module configures advertising regions based on the ad-region adaptation matrix, obtains an initial advertising plan, and then performs node-level advertising simulation to obtain M advertising simulation datasets; a global advertising optimization module optimizes the initial advertising plan based on the M advertising simulation datasets to generate a target advertising plan with recommended advertising time slots and recommended advertising platforms; and an advertising management module manages advertising within the predetermined advertising regions according to the target advertising plan.

[0006] One or more technical solutions provided in this application have at least the following beneficial effects: Before ad placement, preliminary research is conducted to collect multidimensional research datasets, providing a solid foundation for in-depth market and regional data analysis. This ensures that ad placement decisions are based on comprehensive and reliable multidimensional data, thereby improving the accuracy of ad targeting. Through cluster analysis of ad placement resource nodes, the target ad placement area can be divided into multiple sub-regions based on the effectiveness characteristics of different ad resources. Each sub-region represents an area with similar ad resources and audience characteristics. This process makes ad placement strategies more regionally targeted and reduces resource waste. Ad demand for the ads to be placed is obtained and matched with the defined ad placement sub-regions, generating an ad-region fit matrix. Quantitative analysis of the fit ensures that each ad is placed in the most suitable region. Improve advertising effectiveness; through node-level delivery simulation, predict the effects of different advertising delivery plans and obtain multiple delivery simulation datasets. This allows the optimization of advertising delivery plans to not only rely on theoretical analysis but also combine actual simulation data, enhancing the practicality and execution effectiveness of the plans. Based on the simulation data, perform global optimization on the initial advertising delivery plan, optimizing the selection of advertising delivery time periods and platforms. Through global optimization, the best delivery strategy can be selected according to the actual needs of advertising and regional characteristics to maximize advertising effectiveness. After the target advertising delivery plan is generated, dynamic management of advertising delivery is carried out in different advertising delivery areas, monitoring advertising effectiveness in real time and making adjustments to ensure the real-time nature and sustainability of advertising delivery. Through this management approach, advertising delivery can flexibly respond to market changes and improve the efficiency of advertising budget utilization.

[0007] The above description is only an overview of the technical solution of this application. In order to better understand the technical means of this application and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of this application more obvious and understandable, the following are specific embodiments of this application. Attached Figure Description

[0008] Figure 1 This is a schematic diagram of the advertising placement area optimization method based on multidimensional data analysis provided in the embodiments of this application.

[0009] Figure 2 A schematic diagram of the advertising delivery area optimization system based on multidimensional data analysis provided in this application embodiment.

[0010] The attached diagrams are labeled as follows: Pre-research module 10, Cluster analysis module 20, Adaptation analysis module 30, Node-level delivery simulation module 40, Global optimization module for ad delivery 50, and Ad delivery management module 60. Detailed Implementation

[0011] To further illustrate the technical means and effects of the present invention in achieving its intended purpose, the following detailed description of the specific implementation methods, structures, features, and effects of the present invention, in conjunction with the accompanying drawings and preferred embodiments, is provided below.

[0012] Example 1, as Figure 1 As shown in the embodiments of this application, an advertising placement area optimization method based on multidimensional data analysis is provided, the method including: Before launching an advertisement, conduct preliminary research in the designated advertising area based on multiple advertising resource nodes to obtain multiple multi-dimensional research datasets.

[0013] Before actual advertising campaigns, research is conducted on the designated advertising areas. This involves a comprehensive analysis of multiple advertising resource nodes within the region, collecting relevant data to better understand the area's advertising potential. Advertising resource nodes include different types of resources, such as geographic location, audience information, and advertising platforms; each node represents a potential advertising placement point. The preliminary research data is not limited to a single dimension but includes multiple dimensions such as geographic location, user behavior, social interaction, and demographics. Data from each dimension is used for subsequent advertising campaign analysis.

[0014] Based on the multiple multidimensional survey datasets, cluster analysis is performed on the multiple advertising resource nodes, and M advertising sub-regions are defined according to the cluster analysis results.

[0015] Cluster analysis is used to group multiple advertising resource nodes. Cluster analysis uses methods such as K-means and DBSCAN to classify data points into several categories based on their similarity. Specifically, the effectiveness of each advertising resource node is first evaluated, which is achieved through machine learning models or weighted scoring. The evaluation effectiveness coefficient combines multiple factors, such as the potential of advertising and coverage. Based on the node effectiveness evaluation results, nodes with different effectiveness are divided into different intervals. According to the clustering results, the region is divided into M sub-regions. These sub-regions are divided according to the effectiveness and attributes of the advertising resources. Different regions are suitable for different types of advertising needs.

[0016] Obtain N ad placement requirements for N ads to be placed, perform adaptation analysis on the M ad placement sub-regions, and obtain the ad-region adaptation matrix.

[0017] Collect advertising requirements to be placed, including the target audience, geographic location requirements, budget, and time period. Based on the collected advertising requirements and the data of M sub-regions, conduct adaptation analysis. Adaptation analysis evaluates the matching degree between each advertisement and each advertising region, generating an advertisement-region adaptation matrix. Each item in this matrix represents the adaptation degree of a certain advertisement in a certain region. Advertisements with high adaptation degree are more suitable for the audience characteristics and needs of that region.

[0018] After configuring the advertising delivery area based on the advertiser-region adaptation matrix and obtaining the initial advertising delivery plan, node-level delivery simulation is performed to obtain M delivery simulation datasets.

[0019] Based on the ad-region fit matrix, the most suitable sub-region for each ad is first determined. Then, ad placement regions are configured, selecting the optimal placement locations and ad resources based on the ad-region fit, determining which ads to place in which regions, and initially configuring ad resource allocation and budget. After obtaining the initial ad placement plan, node-level placement simulations are performed. This means simulating each ad placement node to test how different combinations of ad resources affect ad performance. Factors considered during the simulation include ad budget, time period limitations, and ad platform selection. Through node-level placement simulations, multiple simulation datasets are obtained, including ad performance predictions and potential user reach for each sub-region.

[0020] Based on the M simulated advertising datasets, the initial advertising campaign is globally optimized to generate a target advertising campaign with recommended advertising time slots and platforms.

[0021] The purpose of global optimization is to optimize the initial advertising campaign, find the optimal advertising strategy, and generate a target advertising campaign. The goal is to optimize the advertising strategy based on the results of multiple simulation datasets, considering advertising effectiveness, resource utilization, and other constraints. Optimization includes the selection of advertising time slots and platforms. The global optimization algorithm selects the best time slots and platforms based on the target audience, performance on different platforms, advertising budget, and time constraints. Specifically, the optimal time slot is determined based on advertising effectiveness and audience activity times; for example, if an ad performs better in the evening, the optimization strategy will favor that time slot. The most suitable platform is selected based on audience data and advertising performance on different platforms; for example, some ads perform better on social media platforms, while others are better suited for search engines. Finally, after global optimization, a target advertising campaign is generated, including recommended advertising time slots and platforms. This campaign is the optimal solution obtained through comprehensive analysis and simulation optimization of multi-dimensional data.

[0022] Based on the target advertising placement plan, manage the advertising placement within the predetermined advertising placement area.

[0023] Based on the target advertising campaign plan, allocate corresponding advertising resources in different advertising regions, including budget allocation, platform selection, and ad placement schedules. During the advertising campaign, continuously monitor ad performance and adjust strategies based on real-time data feedback. Adjust the campaign strategy in real time based on data such as click-through rate, conversion rate, and impressions. For example, if an ad performs worse than expected in a specific region, adjust its placement time or change the platform to achieve the best advertising results.

[0024] Furthermore, based on the aforementioned multiple multidimensional survey datasets, cluster analysis is performed on the multiple advertising resource nodes, including: The multiple multidimensional survey datasets are input into the node performance evaluation network to evaluate the node performance of the multiple advertising resource nodes and obtain multiple advertising performance coefficients. Based on a predetermined advertising performance interval, the multiple advertising performance coefficients are divided into intervals. The multiple advertising resource nodes are clustered according to the interval division results, and M node clusters are output as the clustering analysis results. Each node cluster includes several advertising resource nodes.

[0025] The Node Performance Evaluation Network is a multi-layer neural network model used to evaluate the performance of advertising resource nodes. Multiple multi-dimensional survey datasets are input into the Node Performance Evaluation Network. The evaluation process analyzes each advertising resource node to assess its effectiveness in advertising delivery. Evaluation metrics include ad reach, potential audience matching, and ad delivery results. The evaluation results are expressed as an advertising performance coefficient, which is a digital indicator that measures the effectiveness of an advertising node. A higher advertising performance coefficient indicates a better performance of the node in advertising delivery.

[0026] Based on predetermined advertising performance intervals, different advertising nodes are divided into different intervals according to their performance coefficients. For example, nodes with higher performance coefficients are classified as high-efficiency intervals, and nodes with lower performance coefficients are classified as low-efficiency intervals. Based on the interval division results using performance coefficients, the advertising resource nodes are clustered. Each cluster represents an advertising sub-region, containing several advertising resource nodes with similar performance. Cluster analysis helps identify which regions or nodes are suitable for advertising a certain type of ad and which nodes may be ineffective.

[0027] Furthermore, the node performance evaluation network includes an advertising risk analysis layer, an advertising revenue analysis layer, and a fully connected layer. Based on the advertising risk analysis layer, advertising risk analysis is performed on the multiple multi-dimensional survey datasets, outputting an advertising risk coefficient. Based on the advertising revenue analysis layer, advertising revenue analysis is performed on the multiple multi-dimensional survey datasets, outputting an advertising revenue coefficient. Based on the fully connected layer, the advertising risk coefficient and the advertising revenue coefficient are weighted and fused to output the multiple advertising performance coefficients.

[0028] The Node Performance Evaluation Network is a multi-layer neural network model used to evaluate the performance of advertising resource nodes. This network comprises multiple layers, each responsible for different aspects of analysis. The advertising risk analysis layer focuses on risk factors during the advertising process, including changes in the external environment, fluctuations in user behavior, and the stability of the advertising platform. By analyzing these risks, it can predict potential problems with advertising, such as low exposure and low conversion rates. The advertising revenue analysis layer assesses the potential revenue of advertising, including metrics such as ad exposure rate, click-through rate, and conversion rate. This layer analyzes the profitability of advertising resources, helping to identify which resource nodes generate higher revenue. The fully connected layer, the final layer of the neural network, integrates the results of risk and revenue analysis. It outputs the final advertising performance coefficient through a weighted fusion method, serving as the evaluation result for each resource node.

[0029] By analyzing multiple multidimensional survey datasets, the risks faced by each advertising resource node are assessed. Risk assessment indicators include: the stability of the advertising platform (e.g., the effectiveness of ads on some platforms may be affected by technical issues or changes in user trust); changes in the external environment (e.g., economic or social events causing fluctuations in ad performance); and the suitability of ad content (e.g., whether the ad content meets the needs of the target audience and whether it receives negative feedback). By analyzing these factors, a risk coefficient is output for each resource node. Nodes with higher risk coefficients indicate that advertising at that node faces greater uncertainty or a greater risk of failure.

[0030] Multiple multidimensional survey datasets were analyzed to evaluate the revenue potential of advertising at various resource nodes. Revenue assessment metrics included: ad impressions (whether an ad at a given node reaches a sufficient number of the target audience); click-through rate (CTR), the probability of an ad being clicked, reflecting the match between the ad and the audience; conversion rate, the actual sales or user behavior changes resulting from the ad, measuring the actual effectiveness of the advertising campaign; and return on investment (ROI), which measures the return on ad spending to determine the efficiency of the resources invested. Based on these revenue assessment metrics, a revenue coefficient was calculated for each node; nodes with higher revenue coefficients indicate that the advertising campaign will generate greater benefits.

[0031] In the fully connected layer, the risk and return coefficients of ad placement are weighted and combined according to certain weights. The purpose of this weighting is to comprehensively consider the risks and returns of ad placement to obtain a comprehensive performance coefficient, which represents the overall performance of resource nodes in ad placement. The weighting is based on specific business objectives and strategies. For example, if the current placement objective is more focused on efficient user conversion, then the return coefficient will be given a higher weight; conversely, if the ad platform is unstable and carries a higher risk, then the risk coefficient will be given a higher weight. A node with a high ad placement performance coefficient indicates that the node has both high revenue potential and remains within the risk control range in ad placement.

[0032] Furthermore, by obtaining the N ad placement requirements for N ads to be placed, and performing adaptation analysis on the M ad placement sub-regions, an ad-region adaptation matrix is ​​obtained, including: Extract the first multidimensional survey dataset of the first advertising sub-region, wherein the first multidimensional survey dataset includes first geographic location data, first user behavior data, first social interaction data, and first demographic data; extract the first advertising placement requirements of the first advertisement to be placed, wherein the first advertising placement requirements include first geographic location requirements and first target audience requirements; perform adaptation analysis based on the first multidimensional survey dataset and the first advertising placement requirements to obtain the first advertisement-region adaptation degree; and so on, combine and enumerate the M advertising sub-regions and N advertisements to be placed, output M×N advertisement-region adaptation degrees, and then fill the matrix to obtain the advertisement-region adaptation degree matrix.

[0033] The first advertising sub-region is any one of the M advertising sub-regions, serving as the current analysis object. The first geographic location data reflects the geographic environment information of this sub-region, such as traffic flow, business centers, and transportation hubs. Geographic location data helps determine the population density and activity locations of potential audiences in this area. User behavior data includes information such as users' consumption habits, purchase history, and browsing preferences within the target area. This type of data helps analyze the needs and behavioral patterns of potential audiences. Social interaction data reflects the activity level, sharing behavior, and interaction frequency of users on social media within this area. Social interaction data helps identify the social behavior and influence of the audience, further maximizing advertising effectiveness. Demographic data includes information such as the age structure, gender distribution, income level, and educational background of the population. Advertising can use this data to refine the audience and ensure more targeted advertising.

[0034] The first geolocation requirement indicates the desire to display ads only within a specific geographic area or to an audience in certain locations. For example, if an advertiser is selling products from a local store, the ad requirements would specify that the ads should only be displayed in a certain city or region. The target audience requirements include age, gender, income level, interests, etc. These requirements ensure that ads are only shown to the user group most likely to make a purchase. For example, an ad requirement for a fashion product targeting young people would specifically require the audience to be between 18 and 35 years old.

[0035] Based on the extracted first multidimensional survey dataset and the first advertising demand, adaptation analysis is performed. The goal of adaptation analysis is to determine the matching degree between the advertisement and the region, that is, the degree of fit between the advertising demand and the resources, audience and other characteristics of the advertising region. The results of the adaptation analysis output a value called the advertisement-region matching degree. The matching degree value represents the degree of matching between the advertisement and the region. The higher the value, the better the advertising effect in the region.

[0036] For each ad and each sub-region, an adaptation analysis is performed, and the ad-region adaptation score is output. All the calculated ad-region adaptation score values ​​are filled into the matrix to obtain the complete M×N ad-region adaptation score matrix. Each row of the matrix represents a sub-region, each column represents an ad, and each cell contains the adaptation score value of the ad in that region.

[0037] Furthermore, based on the first multidimensional survey dataset and the first advertising demand, an adaptation analysis is performed to obtain the first advertising-region adaptation degree, including: Based on the first geographic location data and the first geographic positioning requirements, a geographic location adaptation analysis is performed to generate a first geographic location adaptation score; based on the first user behavior data, the first social interaction data, and the first demographic data, a regional user characteristic analysis is performed to generate a first regional user characteristic; after matching and obtaining the first target audience group based on the first target audience requirements, a target audience group characteristic analysis is performed to generate a first target audience group characteristic; based on the first regional user characteristic and the first target audience group characteristic, an advertising audience adaptation analysis is performed to generate a first advertising audience adaptation score; the first geographic location adaptation score and the first advertising audience adaptation score are weighted and calculated to obtain the first advertising-region adaptation score.

[0038] Geographic location fit analysis aims to determine whether the geographic location requirements of an advertisement match the geographic location data within the advertising area. By comparing the geographic location data with the geographic location requirements in the advertising needs, it assesses whether the advertisement can be effectively displayed in the target location. Through analysis, a first geographic location fit score is generated. This fit score reflects the degree of matching between the geographic area where the advertisement is placed and the advertising needs. The higher the fit score, the more the area matches the geographic requirements of the advertisement.

[0039] The first set of user behavior data includes users' consumption behavior, purchase history, and online behavior, such as website browsing, search keywords, and shopping preferences. This data represents users' interests and needs in a specific region. The first set of social interaction data provides information on users' activity levels on social platforms, discussion topics, and interaction frequency. This data can reveal the social behavior and hot topics of interest of users in this region, thereby identifying advertising opportunities. The first set of demographic data includes information on the age structure, gender ratio, income level, and education level of the region. This data reveals whether the user group in this region matches the target audience of the advertisement. By analyzing this data, a regional user profile is generated for the advertising sub-region, reflecting the overall characteristics of the user group in this region, including behavioral patterns, social activities, and demographic attributes.

[0040] Based on the primary target audience requirements, a target audience group that meets the requirements is matched. This means specifying the target group for the advertisement, such as age, gender, income level, interests, etc., identifying the most suitable audience group for the advertisement, further analyzing the characteristics of these audience groups, and establishing a detailed profile of the target audience group, including users' purchasing behavior, social interaction frequency, content interests, activity preferences, etc. Based on the above analysis, the primary target audience characteristics are generated, which describe the specific characteristics of the target audience of the advertisement.

[0041] By comparing the characteristics of users in the first region with those of the first target audience, an ad audience fit analysis is conducted. The goal is to assess whether the audience of the ad in a particular region matches the advertiser's target group. If the user group in the region matches the characteristics of the target audience, the fit is high, and the ad will perform well in that region. Ad audience fit indicates the degree of matching between the user group in the ad placement area and the ad's target audience. Regions with higher fit indicate that the ad can more effectively reach the target audience, thereby improving ad performance.

[0042] First-order geographic location fit and first-order ad audience fit reflect the degree of match between the ad and the geographical requirements of the region and the target audience. To obtain a comprehensive ad-region fit score, these two scores need to be weighted. The weighting values ​​are configured according to actual needs. For example, if audience matching is more important, ad audience fit score can be given a higher weight; if the geographical distribution of the ad is crucial to its effectiveness, geographic location fit score can be given a higher weight. Through weighted calculation, a comprehensive first-order ad-region fit score is finally obtained. This value reflects the overall fit of the ad in the region. The higher the fit score, the more suitable the region is for the ad placement, and the more likely the ad performance is to meet expectations.

[0043] Furthermore, the first multidimensional survey dataset of the first advertising delivery sub-region is extracted, including: Obtain the first node cluster of the first advertising delivery sub-region, which includes several designated advertising delivery resource nodes; perform a union integration of several designated multidimensional survey datasets of the several designated advertising delivery resource nodes to form the first multidimensional survey dataset.

[0044] Focusing on the first advertising sub-region, all relevant nodes in this sub-region are aggregated into a node cluster. Each node cluster represents the distribution and collection of advertising resources, integrating advertising resources from multiple relevant nodes within the same sub-region. These nodes are resources used for advertising within a specific region, representing specific locations, platforms, devices, or specific advertising spaces.

[0045] By performing a union integration of the multidimensional survey datasets of these designated advertising resource nodes, the relevant data of all nodes are merged into a unified first multidimensional survey dataset. This integration ensures that all relevant information is fully considered while integrating various advertising resource nodes, thereby providing comprehensive data support for advertising placement decisions.

[0046] Furthermore, the initial advertising delivery plan is obtained by configuring the advertising delivery area based on the aforementioned ad-region adaptation matrix, including: Based on the ad-region adaptability matrix, extract Q first-adaptive ads for the first ad placement sub-region, where each first-adaptive ad is an ad to be placed whose ad-region adaptability meets a first adaptability threshold. Based on the Q ad-region adaptability scores of the Q first-adaptive ads, allocate resources to generate ad placement resource ratios. Sort the Q ad-region adaptability scores in descending order to generate ad placement priorities. Based on the ad placement priorities and ad placement resource ratios, generate a first ad placement plan for the first ad placement sub-region and add it to the initial ad placement plan.

[0047] The ad-region fit matrix reflects the fit of each ad in each region. Each ad is filtered based on whether its fit in the first ad placement sub-region meets the preset first fit threshold. If the fit value of an ad is greater than or equal to this threshold, then the ad is considered to be suitable for placement in that region, i.e., the first fit ad.

[0048] A high ad fit in a region means the ad will perform better in that region, thus warranting more resources. For example, highly fitted ads will receive more impressions, more budget, or longer exposure periods. The result of resource allocation is the ad delivery resource ratio, which indicates how resources are distributed among the Q fitted ads.

[0049] Ads are sorted in descending order of their relevance, with the ads having the highest relevance placed at the top. This prioritizes ads that have the highest relevance in a specific region, as these are considered more likely to produce good advertising results. By sorting ads in descending order, the priority of each ad is determined, with ads having higher relevance being prioritized and ads having lower relevance being delayed or given less resources.

[0050] Based on ad placement priorities and allocated resource ratios, a specific placement plan is developed for each ad. This means determining the specific details of each ad's placement time, platform, and budget allocation. Specifically, based on ad placement priorities, more resources or more suitable placement time slots are allocated to ads with higher priority. Then, based on the allocated resource ratios, it is ensured that each ad receives reasonable placement resources; for example, higher-priority ads receive more display time or are placed on the most suitable advertising platform. Upon completion, a first ad placement plan is obtained, which includes the specific placement arrangements for all ads, ensuring that each ad receives optimal resource support in the appropriate region according to its priority and suitability.

[0051] Furthermore, based on the advertising placement priority and advertising resource ratio, a first advertising placement plan for the first advertising placement sub-region is generated, including: The system locally calls the first resource restriction information of the first advertising sub-region, parses the advertising time period restriction information and advertising budget restriction information to construct the first constraint condition; extracts the Q basic exposure durations of the Q first adapted ads; based on the first constraint condition, combined with the advertising priority and advertising resource ratio, performs parallel resource scheduling optimization of several designated advertising resource nodes in the first advertising sub-region around the Q basic exposure durations to generate the first advertising placement scheme.

[0052] The first resource constraint information includes resource limitations such as available ad slots and ad budget limits within the region. These constraints directly affect ad delivery methods. For example, some ad slots may have limited availability, or budgets may be limited, requiring optimization of ad delivery. Ad slot restrictions indicate that ads can only be run during certain specific time periods; for example, some ads can only be displayed during peak hours or on specific dates. Ad budget constraints indicate that each ad has a limited budget, and the frequency and duration of ad delivery must be determined based on budget allocation.

[0053] The basic exposure duration is determined based on advertising needs and the availability of platform resources. Specifically, it is determined based on factors such as advertising budget, time of day, and recommendations from the advertising platform. For example, within a short period of time, some ads will have shorter exposure durations, while some high-priority ads will have longer exposure periods.

[0054] Based on the aforementioned first constraint, ad placement priority, and ad resource allocation, resource scheduling optimization is performed. The task of resource scheduling is to optimize ad placement time and platform selection to ensure maximum ad effectiveness while meeting resource constraints such as budget and time slots. Parallel scheduling here means scheduling across multiple ad placement resource nodes simultaneously to maximize resource utilization. Scheduling is performed around the basic exposure duration of each ad to determine the specific time slot and duration for ad placement at each node. Adjustments to exposure duration must consider ad priority, resource allocation, and budget constraints. Through this parallel resource scheduling and optimization process, a first ad placement plan is ultimately generated, which specifies the details of ad placement.

[0055] Furthermore, based on the M simulated advertising datasets, the initial advertising campaign is globally optimized to generate a target advertising campaign with recommended advertising time slots and platforms, including: Based on the multiple multidimensional survey datasets, regional feature analysis of advertising placement is performed to establish strategy coordination rules for the M advertising placement sub-regions. Based on the M advertising simulation datasets, M regional resource configurations and M advertising placement effects are extracted. According to the strategy coordination rules, the M regional resource configurations and M advertising placement effects are traversed to detect and correct advertising resource conflicts, generating the target advertising placement plan. The regional resource configuration includes advertising platform configuration and advertising budget configuration. Based on the advertising platform configuration and advertising budget configuration, global advertising placement optimization is performed to generate recommended advertising placement time slots and recommended advertising placement platforms.

[0056] This study analyzes the regional characteristics of advertising campaigns using multiple multidimensional survey datasets, including geographic information, user behavior, social interactions, and demographic data. The goal of this analysis is to identify the unique features of each advertising sub-region to better formulate advertising strategies. Different regions have different audience characteristics and advertising potential; for example, some densely populated areas have higher ad exposure, while others have a more targeted audience but lower exposure. Based on the regional feature analysis, strategy coordination rules are established for each advertising sub-region. These rules aim to adjust advertising based on regional characteristics, such as: time slot selection rules, which establish rules for selecting ad slots based on activity patterns and the active times of the ad audience in different regions; budget allocation rules, which formulate budget allocation strategies based on advertising performance and budget constraints in different regions, allocating more budget to high-performing regions; and resource priority rules, which establish priority rules based on the suitability of ads and regions to ensure that resources are allocated preferentially to the most suitable ads and regions.

[0057] Each simulation dataset represents the simulated results of ad placement in a specific sub-region, including predicted ad performance. From each simulation dataset, the regional resource configuration for each sub-region is extracted, including the ad platform configuration and budget configuration within that region. Different ad platforms, such as social media, search engines, and video platforms, have different audiences and performance characteristics; extracting these platform configurations helps understand ad performance and resource allocation on different platforms. The budget configuration for each sub-region represents the ad budget for that region, which affects ad display frequency, exposure duration, and other factors, directly impacting ad performance. Additionally, the ad performance for each sub-region is extracted, including key metrics such as click-through rate, conversion rate, audience reach, and ad engagement. These metrics are used to evaluate ad performance across different sub-regions.

[0058] The process iterates through the resource allocation and ad performance of each region, comparing the configurations and performance across different regions to ensure they comply with strategy coordination rules. For example, if a region's budget is too high but its performance is poor, adjustments are needed; if an ad performs poorly on a specific platform, the budget needs to be reallocated or another platform selected. Ad resources may conflict, such as budget overruns, an ad consuming too many resources during the same time slot, or uneven distribution of ad resources across different platforms. These resource conflicts are detected. Upon detection, corrections are made according to strategy coordination rules, including reallocating ad budgets, adjusting ad delivery times, or selecting new platforms. The goal is to ensure that the resource allocation and ad performance of each ad delivery sub-region align with the overall optimization objectives.

[0059] Advertising platform configuration refers to the selection of different advertising platforms for each advertising sub-region. These platforms include social media, search engines, video platforms, etc. Each platform has different advertising effects, and the appropriate advertising platform needs to be selected based on the audience and advertising objectives. Advertising budget configuration is an important factor in determining the frequency and duration of ad display. Each advertising sub-region has a corresponding budget configuration, and the budget needs to be allocated among different regions to ensure that the ads can effectively reach the target audience.

[0060] Global ad optimization involves optimizing resource allocation across all ad placement sub-regions, platforms, and budgets to maximize ad performance. Different times of day affect ad effectiveness; for example, audience activity differs between day and night. Based on ad performance data and regional characteristics, the selection of ad placement times is optimized. Recommended ad placement times are optimized based on ad performance (e.g., click-through rate, conversion rate) and strategy coordination rules (e.g., budget allocation). Platform selection is optimized based on platform performance and budget constraints. For example, some ads perform better on social media platforms, while others are better suited for search engines. Through global optimization, the best ad placement platforms are recommended. Finally, based on the global optimization results, recommended ad placement times and platforms are generated to ensure maximum ad reach and effectiveness across different sub-regions, while ensuring the most efficient use of resources.

[0061] Example 2, based on the same inventive concept as the advertising placement area optimization method based on multidimensional data analysis in the foregoing examples, such as... Figure 2 As shown in the embodiment of this application, an advertising placement area optimization system based on multidimensional data analysis is provided. The system includes: The pre-survey module 10 is used to conduct pre-surveys based on multiple advertising resource nodes in a predetermined advertising area before advertising is launched, and to obtain multiple multi-dimensional survey datasets. The clustering analysis module 20 is used to perform clustering analysis on the multiple advertising resource nodes based on the multiple multi-dimensional survey datasets, and to delineate M advertising sub-regions based on the clustering analysis results. The adaptation analysis module 30 is used to obtain N advertising requirements for N ads to be launched, and to perform adaptation analysis on the M advertising sub-regions to obtain an ad-region adaptation matrix. The node-level launch simulation module 40 is used to configure the advertising area according to the ad-region adaptation matrix, and after obtaining an initial advertising launch plan, to perform node-level launch simulation to obtain M launch simulation datasets. The global optimization module 50 is used to perform global optimization of the initial advertising launch plan based on the M launch simulation datasets, and to generate a target advertising launch plan with recommended advertising launch time periods and recommended advertising launch platforms. The advertising launch management module 60 is used to manage advertising launches within the predetermined advertising area according to the target advertising launch plan.

[0062] Furthermore, the clustering analysis module 20 is used to perform the following operation steps: The multiple multidimensional survey datasets are input into the node performance evaluation network to evaluate the node performance of the multiple advertising resource nodes and obtain multiple advertising performance coefficients. Based on a predetermined advertising performance interval, the multiple advertising performance coefficients are divided into intervals. The multiple advertising resource nodes are clustered according to the interval division results, and M node clusters are output as the clustering analysis results. Each node cluster includes several advertising resource nodes.

[0063] Furthermore, the node performance evaluation network includes an advertising risk analysis layer, an advertising revenue analysis layer, and a fully connected layer. Based on the advertising risk analysis layer, advertising risk analysis is performed on the multiple multi-dimensional survey datasets, outputting an advertising risk coefficient. Based on the advertising revenue analysis layer, advertising revenue analysis is performed on the multiple multi-dimensional survey datasets, outputting an advertising revenue coefficient. Based on the fully connected layer, the advertising risk coefficient and the advertising revenue coefficient are weighted and fused to output the multiple advertising performance coefficients.

[0064] Furthermore, the adaptation analysis module 30 is used to perform the following operation steps: Extract the first multidimensional survey dataset of the first advertising sub-region, wherein the first multidimensional survey dataset includes first geographic location data, first user behavior data, first social interaction data, and first demographic data; extract the first advertising placement requirements of the first advertisement to be placed, wherein the first advertising placement requirements include first geographic location requirements and first target audience requirements; perform adaptation analysis based on the first multidimensional survey dataset and the first advertising placement requirements to obtain the first advertisement-region adaptation degree; and so on, combine and enumerate the M advertising sub-regions and N advertisements to be placed, output M×N advertisement-region adaptation degrees, and then fill the matrix to obtain the advertisement-region adaptation degree matrix.

[0065] Furthermore, the adaptation analysis module 30 is used to perform the following operation steps: Based on the first geographic location data and the first geographic positioning requirements, a geographic location adaptation analysis is performed to generate a first geographic location adaptation score; based on the first user behavior data, the first social interaction data, and the first demographic data, a regional user characteristic analysis is performed to generate a first regional user characteristic; after matching and obtaining the first target audience group based on the first target audience requirements, a target audience group characteristic analysis is performed to generate a first target audience group characteristic; based on the first regional user characteristic and the first target audience group characteristic, an advertising audience adaptation analysis is performed to generate a first advertising audience adaptation score; the first geographic location adaptation score and the first advertising audience adaptation score are weighted and calculated to obtain the first advertising-region adaptation score.

[0066] Furthermore, the adaptation analysis module 30 is used to perform the following operation steps: Obtain the first node cluster of the first advertising delivery sub-region, which includes several designated advertising delivery resource nodes; perform a union integration of several designated multidimensional survey datasets of the several designated advertising delivery resource nodes to form the first multidimensional survey dataset.

[0067] Furthermore, the node-level deployment simulation module 40 is used to perform the following operation steps: Based on the ad-region adaptability matrix, extract Q first-adaptive ads for the first ad placement sub-region, where each first-adaptive ad is an ad to be placed whose ad-region adaptability meets a first adaptability threshold. Based on the Q ad-region adaptability scores of the Q first-adaptive ads, allocate resources to generate ad placement resource ratios. Sort the Q ad-region adaptability scores in descending order to generate ad placement priorities. Based on the ad placement priorities and ad placement resource ratios, generate a first ad placement plan for the first ad placement sub-region and add it to the initial ad placement plan.

[0068] Furthermore, the node-level deployment simulation module 40 is used to perform the following operation steps: The system locally calls the first resource restriction information of the first advertising sub-region, parses the advertising time period restriction information and advertising budget restriction information to construct the first constraint condition; extracts the Q basic exposure durations of the Q first adapted ads; based on the first constraint condition, combined with the advertising priority and advertising resource ratio, performs parallel resource scheduling optimization of several designated advertising resource nodes in the first advertising sub-region around the Q basic exposure durations to generate the first advertising placement scheme.

[0069] Furthermore, the global optimization module 50 for ad placement is used to perform the following steps: Based on the multiple multidimensional survey datasets, regional feature analysis of advertising placement is performed to establish strategy coordination rules for the M advertising placement sub-regions. Based on the M advertising simulation datasets, M regional resource configurations and M advertising placement effects are extracted. According to the strategy coordination rules, the M regional resource configurations and M advertising placement effects are traversed to detect and correct advertising resource conflicts, generating the target advertising placement plan. The regional resource configuration includes advertising platform configuration and advertising budget configuration. Based on the advertising platform configuration and advertising budget configuration, global advertising placement optimization is performed to generate recommended advertising placement time slots and recommended advertising placement platforms.

[0070] Through the foregoing detailed description of the advertising placement area optimization method based on multidimensional data analysis, those skilled in the art can clearly understand the advertising placement area optimization system based on multidimensional data analysis in this embodiment. Since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and relevant parts can be referred to the method section.

[0071] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any modifications, equivalent changes, and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the scope of the present invention.

Claims

1. A method for optimizing advertising delivery areas based on multidimensional data analysis, characterized in that, The method includes: Before launching an advertisement, conduct preliminary research in the designated advertising area based on multiple advertising resource nodes to obtain multiple multi-dimensional survey datasets. Based on the multiple multidimensional survey datasets, cluster analysis is performed on the multiple advertising resource nodes, and M advertising sub-regions are defined according to the cluster analysis results. Obtain N ad placement requirements for N ads to be placed, perform adaptation analysis on the M ad placement sub-regions, and obtain the ad-region adaptation matrix; After configuring the advertising delivery area based on the advertiser-region adaptation matrix and obtaining the initial advertising delivery plan, perform node-level delivery simulation to obtain M delivery simulation datasets. Based on the M simulated advertising datasets, the initial advertising campaign is optimized globally to generate a target advertising campaign with recommended advertising time slots and recommended advertising platforms. Based on the target advertising placement plan, manage the advertising placement within the predetermined advertising placement area.

2. The advertising placement area optimization method based on multidimensional data analysis as described in claim 1, characterized in that, Based on the aforementioned multidimensional survey datasets, cluster analysis is performed on the multiple advertising resource nodes, including: The multiple multidimensional survey datasets are input into the node performance evaluation network to evaluate the node performance of the multiple advertising delivery resource nodes and obtain multiple advertising delivery performance coefficients. Based on the predetermined advertising performance range, the multiple advertising performance coefficients are divided into ranges. The multiple advertising resource nodes are clustered according to the range division results, and M node clusters are output as the clustering analysis results. Each node cluster includes several advertising resource nodes.

3. The advertising placement area optimization method based on multidimensional data analysis as described in claim 2, characterized in that, The node performance evaluation network includes an advertising risk analysis layer, an advertising revenue analysis layer, and a fully connected layer. Based on the aforementioned advertising placement risk analysis layer, advertising placement risk analysis is performed on the multiple multidimensional survey datasets, and an advertising placement risk coefficient is output. Based on the advertising revenue analysis layer, advertising revenue analysis is performed on the multiple multidimensional survey datasets, and advertising revenue coefficients are output. The advertising placement risk coefficient and advertising placement revenue coefficient are weighted and fused based on the fully connected layer to output the multiple advertising placement efficiency coefficients.

4. The advertising placement area optimization method based on multidimensional data analysis as described in claim 1, characterized in that, Obtain N ad placement requirements from N ads to be placed, perform adaptation analysis on the M ad placement sub-regions, and obtain the ad-region adaptation matrix, including: Extract the first multidimensional survey dataset of the first advertising sub-region, wherein the first multidimensional survey dataset includes first geographic location data, first user behavior data, first social interaction data, and first demographic data; Extract the first advertising placement requirements for the first advertisement to be placed, wherein the first advertising placement requirements include first geographic location requirements and first target audience requirements; Based on the first multidimensional survey dataset and the first advertising demand, an adaptation analysis is performed to obtain the first advertising-regional adaptation degree. Similarly, by combining and enumerating the M advertising sub-regions and N advertisements to be placed, and outputting M×N advertisement-region fit scores, the matrix is ​​filled to obtain the advertisement-region fit score matrix.

5. The advertising placement area optimization method based on multidimensional data analysis as described in claim 4, characterized in that, Based on the first multidimensional survey dataset and the first advertising demand, an adaptation analysis is performed to obtain the first advertising-region adaptation degree, including: Based on the first geographic location data and the first geographic location requirements, a geographic location adaptation analysis is performed to generate a first geographic location adaptation degree. Based on the first user behavior data, the first social interaction data, and the first demographic data, regional user characteristics are analyzed to generate the first regional user characteristics. After obtaining the first target audience group based on the first target audience requirements, the target audience group characteristics are analyzed to generate the first target audience group characteristics; Based on the user characteristics of the first region and the characteristics of the first target audience, an advertising audience fit analysis is performed to generate the first advertising audience fit score. The first ad-regional fit is obtained by weighting the first geographic location fit and the first ad audience fit.

6. The advertising placement area optimization method based on multidimensional data analysis as described in claim 2, characterized in that, Extract the first multidimensional survey dataset from the first advertising sub-region, including: Obtain the first node cluster of the first advertising delivery sub-region, which includes several specified advertising delivery resource nodes; The specified multidimensional survey datasets of the specified advertising resource nodes are combined into a union to form the first multidimensional survey dataset.

7. The advertising placement area optimization method based on multidimensional data analysis as described in claim 1, characterized in that, Based on the aforementioned ad-region adaptability matrix, ad delivery regions are configured to obtain an initial ad delivery plan, including: Based on the ad-region fit matrix, extract Q first fit ads for the first ad delivery sub-region, wherein the first fit ads are ads to be delivered whose ad-region fit satisfies the first fit threshold. Based on the Q ad-region adaptability of the Q first-adaptive ads, resources are allocated to generate the ad delivery resource ratio; Arrange the Q ads-region adaptability scores in descending order to generate ad delivery priority; Based on the advertising placement priority and advertising resource ratio, a first advertising placement plan for the first advertising placement sub-region is generated and added to the initial advertising placement plan.

8. The advertising placement area optimization method based on multidimensional data analysis as described in claim 7, characterized in that, Based on the advertising placement priority and advertising resource ratio, a first advertising placement plan is generated for the first advertising placement sub-region, including: Locally, the first resource restriction information of the first advertising sub-region is retrieved, and the advertising time restriction information and advertising budget restriction information are parsed to construct the first constraint condition; Extract the Q base exposure durations of the Q first-adaptation ads; Based on the first constraint, and in conjunction with the advertising placement priority and advertising resource ratio, the first advertising placement scheme is generated by optimizing the parallel resource scheduling of several designated advertising resource nodes within the first advertising placement sub-region around the Q basic exposure durations through the combination of advertising placement priority and advertising resource ratio.

9. The advertising placement area optimization method based on multidimensional data analysis as described in claim 1, characterized in that, Based on the M simulated advertising datasets, the initial advertising campaign is globally optimized to generate a target advertising campaign with recommended advertising time slots and platforms, including: Based on the multiple multidimensional survey datasets, the characteristics of advertising placement areas are analyzed, and strategy coordination rules for the M advertising placement sub-regions are established. Based on the M simulated advertising datasets, extract M regional resource configurations and M advertising performance metrics; Based on the strategy coordination rules, the resource configurations of the M regions and the advertising performance of the M ads are traversed to detect and correct advertising resource conflicts, and the target advertising plan is generated. The regional resource allocation includes advertising platform configuration and advertising budget configuration. Based on the configuration of the advertising platform and the advertising budget, global advertising optimization is performed to generate recommended advertising time slots and recommended advertising platforms.

10. An advertising placement area optimization system based on multidimensional data analysis, characterized in that: The system is used to implement the advertising placement area optimization method based on multidimensional data analysis as described in any one of claims 1-9, the system comprising: The pre-ad research module is used to conduct pre-ad research based on multiple ad resource nodes in the predetermined ad placement area before ad placement, and to obtain multiple multi-dimensional survey datasets. The clustering analysis module is used to perform clustering analysis on the multiple advertising resource nodes based on the multiple multidimensional survey datasets, and to delineate M advertising sub-regions based on the clustering analysis results; The adaptation analysis module is used to obtain the N ad placement requirements of N ads to be placed, perform adaptation analysis on the M ad placement sub-regions, and obtain the ad-region adaptation matrix. The node-level delivery simulation module is used to configure the advertising delivery area according to the ad-region adaptation matrix, obtain the initial advertising delivery plan, perform node-level delivery simulation, and obtain M delivery simulation datasets. The global optimization module for ad placement is used to perform global optimization of the initial ad placement plan based on the M ad placement simulation datasets, and generate a target ad placement plan with recommended ad placement time periods and recommended ad placement platforms; The advertising delivery management module is used to manage advertising delivery within the predetermined advertising delivery area according to the target advertising delivery plan.

Images