Internet advertisement quality evaluation method and system

Abstract

This invention discloses a method and system for evaluating the quality of internet advertising, belonging to the field of internet data processing and advertising quality assessment technology, and applied to government service platforms. The method includes: constructing a multi-level evaluation index system encompassing at least four dimensions: advertising compliance, supply chain fulfillment, government service scenario adaptability, and user conversion; acquiring real-time and offline index data of the advertisement to be evaluated under each dimension; performing data fusion and assigning evaluation weights to each dimension based on a dynamic weight model, which dynamically adjusts the weights according to the platform's real-time operating status, the advertisement's lifecycle, and external public opinion; and calculating a comprehensive quality score based on the fused data and weights. This invention also provides a corresponding evaluation system and storage medium. Through multi-dimensional fusion and dynamic weight adaptation, this invention comprehensively and accurately evaluates the quality of advertisements on government-related internet platforms, balancing commercial effectiveness and public service attributes, and supports early warning and suspension of advertising, effectively safeguarding government credibility.

Description

Technical Field

[0001] This invention relates to the field of internet data processing and advertising quality assessment technology, and in particular to an internet advertising quality evaluation method and system, which is especially suitable for multi-dimensional comprehensive quality evaluation of advertising content in e-commerce and advertising systems built within government mobile applications, government mini-programs, and government public accounts. Background Technology

[0002] As local governments accelerate their digital transformation, numerous government apps, mini-programs, and official WeChat accounts are not only providing convenient government services but also gradually integrating e-commerce platforms, regional supply chains, and advertising systems to promote local products, preferential policies for businesses, and public services. To ensure platform activity and operational effectiveness, the e-commerce and advertising modules of these platforms are typically outsourced to third-party professional agencies. On these government-led internet platforms, advertising content not only possesses commercial promotional attributes but also directly relates to government credibility and the image of public services, making its quality requirements significantly different from those of purely commercial internet media.

[0003] Currently, internet advertising quality evaluation largely follows established methods from the commercial advertising field, primarily focusing on commercial performance indicators such as click-through rate (CTR), conversion rate, and return on investment (ROI), combined with some user behavior data and ad creative scores to evaluate the effectiveness of ad placement. Some evaluation schemes also consider the matching degree between ad content and user interest tags, as well as the ad landing page experience. However, these methods have significant shortcomings when dealing with government-affiliated internet platforms: On the one hand, ads on government platforms need to meet higher standards of content compliance, public interest orientation, and coordination with the government service environment. Relying solely on commercial indicators such as CTR can easily lead to ads with "high clicks but low compliance" being incorrectly rated as high-quality ads. Once content that violates regulations or is inconsistent with the government's image appears, it will seriously affect public trust in the platform. On the other hand, government apps, mini-programs, and official accounts operate with integrated e-commerce and supply chains. The goods or services promoted by ads are often deeply tied to the backend supply chain's fulfillment capabilities, product quality, and after-sales complaints. Ad quality is not only reflected in front-end display and clicks but also extends to the entire chain of actual transaction fulfillment. Existing methods for evaluating advertising quality generally analyze advertising performance in isolation, lacking the technical means to systematically integrate dimensions such as advertising compliance, supply chain service quality, user complaints, and adaptability to government scenarios, making it difficult to truly reflect the overall quality of advertising on government platforms.

[0004] Furthermore, under the outsourced operation model, the operator needs to continuously monitor and optimize advertising content to demonstrate the health and overall effectiveness of advertising to government administrators. However, existing technologies lack an advertising quality evaluation system tailored to the characteristics of government internet platforms, failing to simultaneously consider multiple objectives such as commercial value, public attributes, supply chain coordination, and user experience. This results in lagging advertising quality control, blind optimization directions, and a high risk of negative public opinion, which can weaken the public-benefiting effects of digital government construction.

[0005] Therefore, there is an urgent need for an internet advertising quality evaluation method and system that can effectively integrate multi-source heterogeneous evaluation indicators to meet the actual needs of advertising operation and management in government-related internet platforms.

[0006] The information disclosed in this background section is intended only to enhance the understanding of the overall background of the invention and should not be construed as an admission or in any way implying that the information constitutes prior art known to those skilled in the art. Summary of the Invention

[0007] The purpose of this invention is to provide a method and system for evaluating the quality of internet advertising that can effectively integrate multi-source heterogeneous evaluation indicators, so as to meet the actual needs of advertising operation and management in government-related internet platforms.

[0008] To achieve the above objectives, the present invention provides the following solution: A method for evaluating the quality of internet advertising, applied to a government service platform, the method comprising: Construct a multi-level evaluation index system, which includes at least the dimensions of advertising compliance, supply chain fulfillment, government affairs scenario adaptability, and user conversion. Obtain real-time and offline indicator data for the advertisement to be evaluated under each of the aforementioned dimensions; The real-time indicator data and offline indicator data are fused together, and evaluation weights are assigned to each dimension based on a preset dynamic weight model. The dynamic weight model dynamically adjusts the evaluation weights based on at least one or more factors, including the real-time operating status of the government service platform, the life cycle stage of the advertisement to be evaluated, and external public opinion data. Based on the fused data and the assigned evaluation weights, the overall quality score of the advertisement to be evaluated is calculated.

[0009] Among them, the comprehensive quality score S The calculation formula is:

[0010] In the formula, w c , w s ,w g , w u The dynamic weights are for advertising compliance, supply chain fulfillment, government service scenario adaptability, and user conversion, respectively. w c + w s + w g + w u =1; f c , f s , f g , f u For the standardized scoring function of the corresponding dimension; D c , D s , D g , D u This is the fused indicator dataset for the corresponding dimensions.

[0011] Optionally, the advertising compliance dimension f c( D c The score is calculated using the following formula:

[0012] In the formula, v i For the advertisement in the first i The number of violations in the necessary compliance review items; e i For the first i Total number of exposures for each review item; n This represents the total number of necessary compliance review items. r j For the first j The severity level weight of user complaints; k j The number of times the advertisement was complained about in the j-th type of user complaints; m This represents the total number of user complaint categories.

[0013] Optionally, the supply chain fulfillment dimension f s ( D s The score is calculated using the following formula:

[0014] In the formula, T avg The historical average shipping time for the product corresponding to the advertisement; N succ The number of orders successfully delivered; N total Total number of orders; R defect Product defect rate; S stock This is a standardized value for inventory health. The preset positive weighting coefficients, and .

[0015] Optionally, the government affairs scenario adaptability dimension f g ( D g The score is calculated using the following formula:

[0016] In the formula, Sim ( C ad ,C gov The semantic similarity between the theme of the advertisement content and the core service theme of the government platform on which it is placed; PS Rate the public service attributes of the advertisement; This refers to the deviation between the timing of the advertisement placement and the current policy hotspot. These are preset weighting coefficients; e is the base of the natural logarithm.

[0017] Optionally, the user conversion dimension f u ( D u The score is calculated using the following formula:

[0018] In the formula, CTR For ad click-through rate; CVR Conversion rate; NPS is Net Promoter Score. CTR 0, CVR 0, NPS 0 represents the baseline value of the corresponding indicator; k , λ , v This is a factor to adjust the steepness of the curve; The preset weighting coefficients, and .

[0019] Optionally, the dynamic weight model is w i Adjustment function:

[0020] In the formula, w i ( t ) for in t Time dimension i The weights; For dimension i The baseline weights; This refers to the dynamic adjustment of weights, representing the real-time operating status. O ( t ), Advertising Statement Cycle Stage L ( t ) and external public opinion factors P ( t The function is defined as follows: when the external public opinion factor P(t) exceeds a preset threshold, the weight of the advertising compliance dimension is preferentially increased. w c .

[0021] Optionally, the method further includes: When the overall quality score S When the ad quality falls below the preset first-level warning threshold, an ad quality warning message is generated. when S When the threshold for the second-level circuit breaker is lowered, the advertising delivery circuit breaker mechanism will be automatically triggered, suspending the delivery of the corresponding advertisement.

[0022] An internet advertising quality evaluation system, comprising: A multi-level indicator system construction module is used to build and maintain an evaluation indicator system that includes at least the dimensions of advertising compliance, supply chain fulfillment, government affairs scenario adaptability, and user conversion. The multi-source data acquisition module is used to collect real-time and offline indicator data for each dimension from heterogeneous data sources. The dynamic weight calculation engine is used to calculate and adjust the evaluation weights of each dimension in real time based on the status of the government service platform, the life cycle of the advertisement, and external public opinion data. The comprehensive scoring fusion module is used to calculate the overall quality score of the advertisement based on the fused data and the evaluation weights.

[0023] Optionally, the multi-source data acquisition module specifically includes: The interface call unit is used to connect to the advertising delivery system and obtain exposure, click, and conversion data. The database synchronization unit is used to connect to the supply chain management system to obtain order fulfillment, inventory, and quality control data. The text and image parsing unit is used to perform compliance checks and semantic analysis on advertising content; The public opinion collection unit is used to collect public opinion information related to the platform and advertisers from external channels.

[0024] Compared with the prior art, the present invention has the following beneficial effects: This invention abandons the drawbacks of traditional commercial advertising that relies solely on single business indicators such as click-through rate and conversion rate. Instead, it constructs a multi-level evaluation indicator system that includes four dimensions: advertising compliance, supply chain coverage, adaptability to government scenarios, and user conversion. This system can comprehensively reflect the overall quality of advertising on government-related internet platforms, taking into account both commercial value and public service attributes.

[0025] By introducing a dynamic weighting model, the evaluation weights of each dimension are adjusted in real time based on the real-time operating status of the government service platform, the lifecycle stage of the advertisement, and external public opinion data. For example, when negative public opinion surges, the weight of the planning dimension is automatically increased, making the evaluation results more responsive to actual risks and avoiding misjudgments of "high clicks, low compliance".

[0026] By incorporating supply chain metrics such as order fulfillment success rate, delivery time, product defect rate, and inventory health into the evaluation system, the quality of advertising is not only reflected in front-end display and clicks, but also extends to the entire chain of actual transactions and after-sales service, truly reflecting the comprehensive fulfillment level of the goods or services promoted by the advertisement.

[0027] When the overall quality score falls below the first-level warning threshold, an early warning message is generated. When it falls below the second-level circuit breaker threshold, advertising is automatically suspended, effectively preventing low-quality advertisements from damaging the government's credibility and reducing public opinion risks.

[0028] Through various units such as interface calls, database synchronization, text and image parsing, and public opinion collection, the system achieves unified collection and integration of heterogeneous data sources such as advertising placement systems, supply chain management systems, and external public opinion channels, providing a solid data foundation for evaluation. Attached Figure Description

[0029] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0030] Figure 1 This is a schematic diagram of the process for evaluating the quality of internet advertising provided in an embodiment of the present invention.

[0031] Figure 2 This is a schematic diagram of the structure of an internet advertising quality evaluation system provided in an embodiment of the present invention. Detailed Implementation

[0032] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0033] The purpose of this invention is to provide a method and system for evaluating the quality of internet advertising that can effectively integrate multi-source heterogeneous evaluation indicators, so as to meet the actual needs of advertising operation and management in government-related internet platforms.

[0034] This invention provides a method and system for evaluating the quality of internet advertising, specifically designed for the unique needs of government service platforms (such as government apps, government mini-programs, and e-commerce and advertising systems embedded in government official accounts). The core of this invention lies in constructing a multi-dimensional, dynamically adjustable comprehensive evaluation system that deeply integrates and evaluates the commercial attributes of advertising with its public service attributes and supply chain fulfillment capabilities.

[0035] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0036] Example 1: This embodiment details the specific steps involved in evaluating advertising quality. For example... Figure 1 As shown, the method includes the following key steps: Step 1: Construct a multi-level evaluation index system.

[0037] First, the system constructs a multi-level evaluation index system for the government service platform. In this embodiment, the system includes at least four core dimensions: 1. Advertising compliance dimension: This measures whether advertising content complies with laws and regulations, platform policies, and public order and good morals.

[0038] 2. Supply chain fulfillment dimension: This measures the actual delivery capability of the goods or services promoted in the advertisement in terms of transaction, logistics, quality control, and other back-end aspects.

[0039] 3. Government Service Scenario Adaptability Dimension: This dimension measures the degree to which the advertising content aligns with the service objectives, policy hotspots, and public image of the government service platform.

[0040] 4. User Conversion Dimension: Used to measure the commercial effectiveness of ads in attracting users to click and complete expected behaviors (such as placing orders or registering).

[0041] Step 2: Obtain multi-source indicator data.

[0042] The system collects data on the advertisements to be evaluated across the aforementioned dimensions using a multi-source data acquisition module. This data is divided into two categories: Real-time metrics data: such as real-time click-through rate, impressions, and user complaints for advertisements.

[0043] Offline metrics data: such as historical order fulfillment success rate, average product delivery time, inventory turnover rate, etc.

[0044] Step 3: Data fusion and dynamic weight allocation.

[0045] The system performs cleaning, normalization, and time alignment on the collected real-time and offline data to create a standardized indicator dataset. D c , D s , D g , D u .

[0046] Subsequently, the system invokes a preset dynamic weight model. This model does not use fixed weights, but rather calculates and assigns dynamic weights to the four dimensions based on factors such as the real-time operational status of the government service platform (e.g., server load, user activity), the lifecycle stage of the advertisement to be evaluated (e.g., testing phase, scaling-up phase, decline phase), and external public opinion data (e.g., negative mentions on social media, news complaints). w c , w s , w g , w u .

[0047] For example, when the platform load is high during the "Double Eleven" promotion, the weight of the user conversion dimension can be appropriately reduced; when a collective complaint is detected about a certain type of product, the weight of the advertising compliance dimension and the supply chain fulfillment dimension should be immediately increased.

[0048] Step 4: Calculate the overall quality score.

[0049] Finally, based on the fused indicator data and dynamic weights, the system calculates the overall quality score of the advertisement. S In this invention, S The calculation formula is:

[0050] in, f c , f s , f g , f u These are the standardized scoring functions for the corresponding dimensions, and their output values ​​are all normalized to the [0,1] interval.

[0051] Specific implementation details regarding advertising compliance scoring: In this embodiment, advertising compliance scoring f c ( D c It considers not only the number of violations but also the severity of user complaints. The calculation formula is as follows:

[0052] In the formula, v i For the advertisement in the first i The number of violations in the necessary compliance review items; e i For the first i Total number of exposures for each review item; n This represents the total number of necessary compliance review items. r j For the first j The severity level weight of user complaints; k j For the advertisement in the first j The number of times a complaint was filed against a user in a category of user complaints; m This represents the total number of user complaint categories.

[0053] For example, the number of times an advertisement violates a necessary compliance review item (such as "use of absolute terms") v i For 2 times, total exposure e i If the number of violations is 10,000, the violation rate is 0.0002. Simultaneously, the advertisement received a complaint for "false advertising" (severity level weight). r j= 0.8) 3 times ( k j =3), and complaints related to "poor after-sales service" (severity level weight). r j =0.3) 5 times. The system standardizes the two types of penalty items and deducts them from the full score of 1 to obtain the final compliance score.

[0054] Specific implementation details of supply chain fulfillment scoring: Supply chain fulfillment rating f s ( D s This factor integrates logistics, order success rate, product quality, and inventory health. The calculation formula is as follows:

[0055] In the formula, T avg The historical average shipping time for the product corresponding to the advertisement; N succ The number of orders successfully delivered; N total Total number of orders; R defect Product defect rate; S stock This is a standardized value for inventory health. The preset positive weighting coefficients, and .

[0056] For example, the average delivery time for a certain product T avg =2 days, order success rate Defect rate R defect =0.01, Inventory Health S stock =0.95. If set , , , Then, the specific performance score can be calculated.

[0057] Specific implementation details of the scoring for the adaptability dimension of government service scenarios: Government Affairs Scenarios Adaptability Score f g ( D g The formula for calculating ) is:

[0058] In the formula, Sim( C ad , C gov The semantic similarity between the advertisement's content theme and the core service theme of the government platform to which it is placed is denoted; PS is the public service attribute score of the advertisement. This refers to the deviation between the timing of the advertisement placement and the current policy hotspot. w g The preset weighting coefficient (taken in this embodiment)w g =1, so as to normalize the score); e is the base of the natural logarithm.

[0059] The system uses natural language processing techniques (such as the BERT model) to calculate the advertising theme. C ad With the core service theme of the government affairs platform C gov The semantic similarity (Sim) of advertisements (such as "social security processing" and "business establishment") is evaluated. Simultaneously, the system assesses the public service attributes of the advertisements. PS (e.g., whether to promote pro-agricultural policies and local specialties), and calculate the time deviation between the advertising placement time and the recent policy hotspot window period △ T For example, around "315 Consumer Rights Day," public service advertisements about "honest business practices" are placed, with their Sim and PS The score will be very high, △ T It is very small, so the overall score is close to 1.

[0060] Specific implementation details regarding user conversion dimensional scoring: User conversion score f u ( D u A sigmoid function (such as a logistic function) is used to address the diminishing marginal returns of the indicator. The calculation formula is as follows:

[0061] In the formula, CTR For ad click-through rate; CVR Conversion rate; NPS Net Promoter Score (NPS) for users; CTR 0, CVR 0, NPS 0 represents the baseline value of the corresponding indicator; k ,λ, v This is a factor to adjust the steepness of the curve; w 1, w 2, w 3 is the preset weighting coefficient, and w 1+ w 2+ w 3 = 1.

[0062] This design results in slow score growth when the metric is below the baseline, increased sensitivity when it is above the baseline, and eventually saturation.

[0063] Regarding the specific implementation of dynamic weights: The core function of the dynamic weight model is:

[0064] In the formula, w i (t) represents the value in t Time dimension i The weights; For dimension i The benchmark weights; △ w i This refers to the dynamic adjustment of weights, representing the real-time operating status. O ( t ), Advertising lifecycle stages L ( t ) and external public opinion factors P ( t The function of ).

[0065] A specific adjustment logic is as follows: The system monitors external public opinion in real time through the public opinion collection unit. When a surge in negative public opinion related to the platform or advertisers is detected, leading to an increase in external public opinion factors... P ( t When the threshold is exceeded (e.g., 0.8), the system will prioritize improving the advertising compliance dimension. w c The weights of the dimensions are adjusted (e.g., increased by 0.2 while proportionally decreasing the weights of other dimensions) to improve the overall score. S They are more sensitive to compliance issues and may trigger subsequent warnings or circuit breaker mechanisms.

[0066] Regarding the specific implementation of early warning and circuit breaker mechanisms: This method also includes automated control steps. The system presets two thresholds: a level one warning threshold (such as...). S =0.6) and secondary circuit breaker threshold (e.g. S =0.4).

[0067] when S When the value is less than 0.6, the system generates an ad quality warning and notifies the operations staff via email or system message, suggesting that the ad be optimized or investigated.

[0068] when S When the value is less than 0.4, the system will automatically trigger the advertising delivery circuit breaker mechanism, immediately suspend the delivery of the advertisement on all channels, and report the result to the platform administrator until the problem is resolved and the advertisement is manually reviewed and restored.

[0069] Example 2: Internet advertising quality evaluation system.

[0070] This embodiment describes in detail the system architecture for implementing the above method. For example... Figure 2 As shown, the system includes the following core modules: Multi-level indicator system construction module: This module is responsible for constructing and maintaining the evaluation indicator system described in Example 1, which includes dimensions such as advertising compliance, supply chain fulfillment, government scenario adaptability, and user conversion, based on the needs of government service scenarios. Administrators can use this module to customize indicators, adjust indicator weight baselines, etc.

[0071] Multi-source data acquisition module: Responsible for collecting data from various heterogeneous data sources on the government service platform. Internally, it includes: Interface call unit: Connects to the advertising system via API to obtain real-time data such as ad exposure, clicks, and conversions.

[0072] Database synchronization unit: Synchronizes offline data such as order fulfillment, inventory levels, and product quality control from the supply chain management system using ETL tools.

[0073] Text and Image Analysis Unit: Utilizing OCR and NLP technologies, this unit automatically performs compliance checks and thematic semantic analysis on the text, images, and video content of advertisements.

[0074] Public opinion collection unit: Collects public opinion information related to the platform and advertisers from external channels such as Weibo, Xiaohongshu, and Black Cat Complaints through web crawlers or third-party APIs.

[0075] Dynamic Weight Calculation Engine: This is the core decision-making component of the system. It receives data in real time from the government service platform's operational status, the advertising lifecycle management module, and the public opinion collection unit, and calculates and outputs dynamic weights for each dimension.

[0076] The comprehensive scoring fusion module receives standardized indicator data from the multi-source data acquisition module and dynamic weights from the weight dynamic calculation engine, calls the formula described in claim 1 to calculate the comprehensive quality score of the advertisement, and outputs the scoring results to the outside world. At the same time, it triggers warnings or circuit breaker instructions based on the score.

[0077] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the systems disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the descriptions are relatively simple; relevant parts can be referred to the method section.

[0078] This document uses specific examples to illustrate the principles and implementation methods of the present invention. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of the present invention. Furthermore, those skilled in the art will recognize that, based on the ideas of the present invention, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. An Internet advertisement quality evaluation method characterized by comprising: Applied to a government service platform, the method includes: Construct a multi-level evaluation index system, which includes at least the dimensions of advertising compliance, supply chain fulfillment, government affairs scenario adaptability, and user conversion. Obtain real-time and offline indicator data for the advertisement to be evaluated under each of the aforementioned dimensions; The real-time indicator data and offline indicator data are fused together, and evaluation weights are assigned to each dimension based on a preset dynamic weight model. The dynamic weight model dynamically adjusts the evaluation weights based on at least one or more factors, including the real-time operating status of the government service platform, the life cycle stage of the advertisement to be evaluated, and external public opinion data. Based on the fused data and the assigned evaluation weights, the overall quality score of the advertisement to be evaluated is calculated; The comprehensive quality score is calculated according to the following formula: S The calculation formula is as follows: In the formula, w c , w s , w g , w u The dynamic weights are for advertising compliance, supply chain fulfillment, government service scenario adaptability, and user conversion, respectively. w c + w s + w g + w u =1; f c , f s , f g , f u For the corresponding dimension, a standardized scoring function is provided. D c , D s , D g , D u This is the fused index dataset for the corresponding dimensions.

2. The Internet advertisement quality evaluation method according to claim 1, characterized by, the ad compliance dimension f c( D c ) is calculated by the following formula: In the formula, v i For the advertisement in the first i The number of violations in the necessary compliance review items; e i For the first i Total number of exposures for each review item; n This represents the total number of necessary compliance review items. r j For the first j The severity level weight of user complaints; k j The number of times the advertisement was complained about in the j-th type of user complaints; m This represents the total number of user complaint categories.

3. The Internet advertisement quality evaluation method according to claim 1, characterized by, The supply chain fulfillment dimension f s ( D s The score is calculated using the following formula: In the formula, T avg The historical average shipping time for the product corresponding to the advertisement; N succ The number of orders successfully delivered; N total Total number of orders; R defect Product defect rate; S stock This is a standardized value for inventory health. The preset positive weighting coefficients, and .

4. The method for evaluating the quality of internet advertising according to claim 1, characterized in that, the government affair scenario adaptability dimension f g ( D g ) is calculated by the following formula: In the formula, Sim ( C ad ,C gov The semantic similarity between the theme of the advertisement content and the core service theme of the government platform on which it is placed; PS Rate the public service attributes of the advertisement; This refers to the deviation between the timing of the advertisement placement and the current policy hotspot. These are preset weighting coefficients; e is the base of the natural logarithm.

5. The method for evaluating the quality of internet advertising according to claim 1, characterized in that, the user conversion dimension f u ( D u ) is calculated by the following formula: In the formula, CTR is the advertising click rate; CVR is the conversion rate; NPS is the user net recommendation value; CTR 0, CVR 0, NPS 0 represents the baseline value of the corresponding indicator; k , λ , v This is a factor for adjusting the steepness of the curve; The preset weighting coefficients, and .

6. The Internet advertisement quality evaluation method according to claim 1, characterized by, The dynamic weight model is w i adjustment function: In the formula, w i ( t ) for in t Time dimension i The weights; For dimension i The baseline weights; This refers to the dynamic adjustment of weights, representing the real-time operating status. O ( t ), Advertising Statement Cycle Stage L ( t ) and external public opinion factors P ( t The function is defined as follows: when the external public opinion factor P(t) exceeds a preset threshold, the weight of the advertising compliance dimension is preferentially increased. w c .

7. The Internet advertisement quality evaluation method according to claim 1, characterized by, The method further includes: when the comprehensive quality score S an advertisement quality early warning information is generated when the comprehensive quality score is lower than a preset first early warning threshold. When S When the preset secondary fusing threshold is lower than the current number of times of the corresponding advertisement being displayed, the advertisement display fusing mechanism is triggered automatically, and the display of the corresponding advertisement is suspended.

8. An Internet advertisement quality evaluation system characterized by comprising: The method described by any one of claims 1 to 7 comprises: A multi-level indicator system construction module is used to build and maintain an evaluation indicator system that includes at least the dimensions of advertising compliance, supply chain fulfillment, government affairs scenario adaptability, and user conversion. The multi-source data acquisition module is used to collect real-time and offline indicator data for each dimension from heterogeneous data sources. The dynamic weight calculation engine is used to calculate and adjust the evaluation weights of each dimension in real time based on the status of the government service platform, the life cycle of the advertisement, and external public opinion data. The comprehensive scoring fusion module is used to calculate the overall quality score of the advertisement based on the fused data and the evaluation weights.

9. The Internet advertisement quality evaluation system according to Claim 8, characterized by The multi-source data acquisition module specifically includes: The interface call unit is used to connect to the advertising delivery system and obtain exposure, click, and conversion data. The database synchronization unit is used to connect to the supply chain management system to obtain order fulfillment, inventory, and quality control data. The text and image parsing unit is used to perform compliance checks and semantic analysis on advertising content; The public opinion collection unit is used to collect public opinion information related to the platform and advertisers from external channels.

10. A computer-readable storage medium having stored thereon a computer program, characterized in that, When the program is executed by the processor, it implements the steps of the method as described in any one of claims 1 to 7.

Images