Media flow credibility detection method and system

Abstract

The invention provides a method and a system for detecting credibility of media traffic. The method comprises the following steps: acquiring and buying media traffic based on a DSP platform, and putting advertisement materials based on the acquired and bought media traffic; obtaining related data involved in media traffic purchase and advertisement material putting processes; determining three traffic characteristic credibility according to an exposure index and a 302 index, a click index and an arrival index, and an arrival index and a two-hop index, and determining traffic characteristic IP credibility based on an IP reported by traffic exposure according to a traffic request parameter, a landing page acquisition parameter and a server-side acquisition legitimacy; determining the flow characteristic UA credibility based on a browser standard protocol header reported by flow exposure, and determining the equipment characteristic credibility according to the flow characteristic IP credibility 4 and the flow characteristic UA credibility 5; and according to each acquired credibility, determining whether the purchased media traffic is credible based on a media traffic credibility detection model. The system comprises each function module for correspondingly realizing each step. According to the invention, the authenticity of the media traffic can be quickly verified.

Description

Technical Field

[0001] This invention belongs to the field of digital advertising, and more specifically, relates to a method and system for detecting the credibility of media traffic. Background Technology

[0002] Programmatic advertising exchanges, through digital platforms, automate ad buying and delivery via ADX (Advanced Demand Optimization) from an audience matching perspective, and provide real-time data analysis. However, the following problems arise in this process:

[0003] 1. Fake traffic leads to wasted advertising budgets, failure to accurately reach real users, and inflated customer acquisition costs;

[0004] Second, data fraud undermines the evaluation of advertising effectiveness, weakens trust in digital marketing, and has a long-term negative impact on brand reputation.

[0005] Third, low-quality content gains a competitive advantage by inflating its traffic, squeezing out compliance space and creating a "bad money drives out good money" effect;

[0006] Fourth, new technologies are being used to generate highly realistic fake traffic, which traditional risk control methods are inefficient at identifying. Summary of the Invention

[0007] In view of this, the present invention provides a method and system for detecting the credibility of media traffic.

[0008] According to a first aspect of the present invention, a method for detecting the credibility of media traffic is provided, the method comprising the following steps:

[0009] Media traffic is purchased using a DSP platform, and advertising creatives are delivered based on the purchased media traffic.

[0010] Acquire relevant data involved in media traffic purchase and advertising material delivery, including traffic request parameters, landing page acquisition parameters, server-side legal acquisition parameters, exposure metrics, 302 metrics, click metrics, arrival metrics, and second-hop metrics;

[0011] The credibility of traffic features is determined based on the following criteria: 1. Exposure metrics and 302 metrics; 2. Click metrics and arrival metrics; 3. Arrival metrics and second-hop metrics; 4. Traffic feature IP credibility is determined based on traffic request parameters, landing page acquisition parameters, server-acquired legal parameters, and the IP address reported for traffic exposure; 5. Traffic feature UA credibility is determined based on traffic request parameters, landing page acquisition parameters, server-acquired legal parameters, and the browser standard protocol header reported for traffic exposure; 6. Device feature credibility is determined based on traffic feature IP credibility 4 and traffic feature UA credibility 5.

[0012] Based on the obtained credibility levels, and using a pre-built media traffic credibility detection model, it is determined whether the purchased media traffic is credible.

[0013] Optionally, the step of determining the credibility of traffic features 1 based on the exposure index and the 302 index specifically includes:

[0014] The ratio of the exposure index to the 302 index is used as the confidence level of the traffic feature.

[0015] Optionally, the step of determining the credibility of traffic features based on click metrics and arrival metrics specifically includes:

[0016] The ratio of arrival metrics to click metrics is used as the confidence level of traffic features.

[0017] Optionally, the step of determining the reliability of traffic characteristics 3 based on arrival indicators and two-hop indicators specifically includes:

[0018] The ratio of the arrival index to the second-hop index is used as the confidence level of the traffic feature.

[0019] Optionally, step 4 of determining the credibility of traffic feature IPs based on traffic request parameters, landing page acquisition parameters, server-acquired valid parameters, and traffic exposure reported IPs further includes:

[0020] Obtain the consistency between each traffic request parameter and the IP address reported for traffic exposure (1);

[0021] 2. Obtain the consistency between the parameters obtained from each landing page and the IPs reported for traffic exposure;

[0022] 3. Obtain the consistency between the legal parameters obtained by each server and the IPs reported for traffic exposure;

[0023] The average of consistency scores 1, 2, and 3 is used as the traffic feature IP credibility score 4.

[0024] Optionally, the step of determining the credibility of the traffic feature UA based on traffic request parameters, landing page acquisition parameters, server-acquired valid parameters, and browser standard protocol headers for traffic exposure reporting further includes:

[0025] 4. Obtain the consistency between each traffic request parameter and the browser standard protocol header for traffic exposure reporting;

[0026] The consistency between the parameters obtained from each landing page and the browser standard protocol headers for traffic exposure reporting is 5.

[0027] 6. Obtain the consistency between the legal parameters obtained by each server and the browser standard protocol headers for traffic exposure reporting;

[0028] The average of consistency scores 4, 5, and 6 is used as the traffic characteristic IP credibility score 5.

[0029] Optionally, the step of determining the device characteristic confidence level 6 based on traffic characteristic IP confidence level 4 and traffic characteristic UA confidence level 5 specifically includes:

[0030] The product of traffic feature IP credibility 4 and traffic feature UA credibility 5 is taken as device feature credibility 6.

[0031] Alternatively, the media traffic credibility detection model may be implemented using a machine learning model or a neural network model.

[0032] Optionally, the steps for obtaining relevant data involved in media traffic purchase and advertising material delivery are implemented based on monitoring links bound to advertising materials.

[0033] According to a second aspect of the present invention, a media traffic credibility detection system is provided, the system comprising the following functional modules:

[0034] The programmatic advertising buying module is used to purchase media traffic based on the DSP platform and to deliver advertising creatives based on the purchased media traffic.

[0035] The raw data acquisition module is used to acquire relevant data involved in the process of media traffic purchase and advertising material delivery. The relevant data includes traffic request parameters, landing page acquisition parameters, server-side legal acquisition parameters, exposure metrics, 302 metrics, click metrics, arrival metrics, and second-hop metrics.

[0036] The feature data acquisition module is used to determine the credibility of traffic features 1 based on exposure metrics and 302 metrics; determine the credibility of traffic features 2 based on click metrics and arrival metrics; determine the credibility of traffic features 3 based on arrival metrics and second-hop metrics; determine the credibility of traffic feature IP based on traffic request parameters, landing page acquisition parameters, legal parameters acquired by the server, and IP of traffic exposure reporting; determine the credibility of traffic feature UA based on traffic request parameters, landing page acquisition parameters, legal parameters acquired by the server, and browser standard protocol header of traffic exposure reporting; and determine the credibility of device features 6 based on traffic feature IP credibility 4 and traffic feature UA credibility 5.

[0037] The media traffic credibility detection module is used to determine whether the purchased media traffic is credible based on the obtained credibility scores and a pre-built media traffic credibility detection model.

[0038] The beneficial effects of this invention are as follows:

[0039] This invention provides a method and system for detecting the credibility of media traffic. This invention significantly improves the filtering and identification of invalid and fraudulent traffic, thereby reducing customer economic losses, enhancing brand trust, maintaining market order, and strengthening technological countermeasures. Compared to existing methods that rely on customer feedback on traffic quality received a day or more later to detect the credibility of media traffic, this invention can quickly verify the authenticity of media traffic.

[0040] Other features and advantages of the present invention will be described in detail in the following detailed description section. Attached Figure Description

[0041] The present invention can be better understood by referring to the following description taken in conjunction with the accompanying drawings, in which the same or similar reference numerals are used throughout the drawings to denote the same or similar parts.

[0042] Figure 1 A flowchart illustrating the implementation of a media traffic credibility detection method according to an embodiment of the present invention is shown.

[0043] Figure 2 A principle block diagram of a media traffic credibility detection method according to an embodiment of the present invention is shown;

[0044] Figure 3 A structural block diagram of a media traffic credibility detection system according to an embodiment of the present invention is shown. Detailed Implementation

[0045] To enable those skilled in the art to more fully understand the technical solutions of the present invention, exemplary embodiments of the present invention will be described more comprehensively and in detail below with reference to the accompanying drawings. Obviously, the one or more embodiments of the present invention described below are merely one or more specific ways to implement the technical solutions of the present invention, and are not exhaustive. It should be understood that other ways belonging to a general inventive concept can be used to implement the technical solutions of the present invention, and should not be limited to the embodiments described exemplary. Based on one or more embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0046] Example: Figure 1 A flowchart illustrating the implementation of the media traffic credibility detection method according to an embodiment of the present invention is shown. (Refer to...) Figure 1 The media traffic credibility detection method of this invention includes the following steps:

[0047] Step S100: Purchase media traffic based on the DSP platform and deliver advertising materials based on the purchased media traffic;

[0048] Step S200: Obtain relevant data involved in the process of media traffic purchase and advertising material delivery. The relevant data includes traffic request parameters, landing page acquisition parameters, server-side legal acquisition parameters, exposure metrics, 302 metrics, click metrics, arrival metrics, and second-hop metrics.

[0049] Step S300: Determine the credibility of traffic features 1 based on exposure metrics and 302 metrics; determine the credibility of traffic features 2 based on click metrics and arrival metrics; determine the credibility of traffic features 3 based on arrival metrics and second-hop metrics; determine the credibility of traffic feature IP based on traffic request parameters, landing page acquisition parameters, server-acquired legal parameters, and the IP of traffic exposure reporting; determine the credibility of traffic feature UA based on traffic request parameters, landing page acquisition parameters, server-acquired legal parameters, and the browser standard protocol header of traffic exposure reporting; determine the credibility of device features 6 based on traffic feature IP credibility 4 and traffic feature UA credibility 5.

[0050] Step S400: Based on the obtained credibility scores and the pre-built media traffic credibility detection model, determine whether the purchased media traffic is credible.

[0051] Specifically, in this embodiment of the invention, 302 redirection is a server-side redirection technique (Temporarily Moved), 302 metrics refer to data obtained using 302 redirection technology; click metrics specifically refer to click-through rate; landing metrics specifically refer to first-trigger rate, also known as landing rate; and second-hop metrics specifically refer to second-hop rate (2nd-click rate).

[0052] Furthermore, in step 300 of this embodiment of the invention, the traffic feature confidence level 1 is the ratio of the exposure index to the 302 index.

[0053] Furthermore, in step 300 of this embodiment of the invention, the traffic feature confidence level 2 is the ratio of the arrival index to the click index.

[0054] Furthermore, in step 300 of this embodiment of the invention, the traffic feature confidence level 3 is the ratio of the arrival index to the second-hop index.

[0055] Furthermore, in step 300 of this embodiment of the invention, the method for determining the traffic feature IP credibility 4 is specifically as follows:

[0056] Obtain the consistency between each traffic request parameter and the IP address reported for traffic exposure (1);

[0057] 2. Obtain the consistency between the parameters obtained from each landing page and the IPs reported for traffic exposure;

[0058] 3. Obtain the consistency between the legal parameters obtained by each server and the IPs reported for traffic exposure;

[0059] The average of consistency scores 1, 2, and 3 is used as the traffic feature IP credibility score 4.

[0060] Furthermore, in step 300 of this embodiment of the invention, the method for determining the credibility 5 of the traffic feature UA is as follows:

[0061] 4. Obtain the consistency between each traffic request parameter and the browser standard protocol header for traffic exposure reporting;

[0062] The consistency between the parameters obtained from each landing page and the browser standard protocol headers for traffic exposure reporting is 5.

[0063] 6. Obtain the consistency between the legal parameters obtained by each server and the browser standard protocol headers for traffic exposure reporting;

[0064] The average of consistency scores 4, 5, and 6 is used as the traffic characteristic IP credibility score 5.

[0065] Furthermore, in step 300 of this embodiment of the invention, the device feature confidence level 6 is the product of the traffic feature IP confidence level 4 and the traffic feature UA confidence level 5.

[0066] Specifically, in this embodiment of the invention, the ratio of the exposure index to the 302 index is defined as the traffic feature confidence level 1. Generally speaking, when the traffic feature confidence level 1 is in the range of 0.95 to 1.05, it can indicate to a certain extent that the corresponding media traffic is in a normal state. The ratio of the arrival index to the click index is defined as the traffic feature confidence level 2. Generally speaking, when the traffic feature confidence level 2 is in the range of 0.95 to 1.05, it can indicate to a certain extent that the corresponding media traffic is in a normal state. The ratio of the arrival index to the second-hop index is defined as the traffic feature confidence level 3. Generally speaking, when the traffic feature confidence level 3 is greater than 0.3, it can indicate to a certain extent that the corresponding media traffic is in a normal state.

[0067] Specifically, in this embodiment of the invention, the average of the consistency 1 of each traffic request parameter relative to the IP reported for traffic exposure, the consistency 2 of each landing page acquisition parameter relative to the IP reported for traffic exposure, and the consistency 3 of each server-acquired legitimate parameter relative to the IP reported for traffic exposure is defined as the traffic feature IP credibility 4. For example, if 90 out of 100 traffic request parameters match the IP reported for traffic exposure, the corresponding consistency 1 is 0.9; if 80 out of 100 landing page acquisition parameters match the IP reported for traffic exposure, the corresponding consistency 2 is 0.8; if 70 out of 100 server-acquired legitimate parameters match the IP reported for traffic exposure, the corresponding consistency 3 is 0.7; and the corresponding traffic feature IP credibility 4 is 0.8.

[0068] Specifically, in this embodiment of the invention, the average of the consistency 4 of each traffic request parameter with respect to the browser standard protocol header reported for traffic exposure, the consistency 5 of each landing page acquisition parameter with respect to the browser standard protocol header reported for traffic exposure, and the consistency 6 of each server-acquired legitimate parameter with respect to the browser standard protocol header reported for traffic exposure is defined as the traffic feature IP credibility 5. For example, if 90 out of 100 traffic request parameters are consistent with the browser standard protocol header reported for traffic exposure, the corresponding consistency 4 is 0.9; if 80 out of 100 landing page acquisition parameters are consistent with the browser standard protocol header reported for traffic exposure, the corresponding consistency 5 is 0.8; if 70 out of 100 server-acquired legitimate parameters are consistent with the browser standard protocol header reported for traffic exposure, the corresponding consistency 6 is 0.7; and the corresponding traffic feature IP credibility 5 is 0.8.

[0069] Specifically, Figure 2 A principle block diagram of the media traffic credibility detection method according to an embodiment of the present invention is shown. The following is in conjunction with... Figure 2 The specific implementation process of the media traffic credibility detection method according to the embodiments of the present invention will be described in more detail as follows:

[0070] The specific implementation process of the media traffic credibility detection method in this embodiment of the invention mainly includes the following five steps:

[0071] 1) Configure necessary materials for monitoring behaviors such as landing page, impressions, 302 impressions, clicks, arrivals, and second-hops based on the DSP platform, and send the materials to the ADX, which then sends traffic;

[0072] 2) The data acquisition system automatically collects data and generates reports;

[0073] 3) Data collection is performed through the data collection system, including data cleaning, machine learning, model fitting, model evaluation, model generation, and model deployment.

[0074] 4) Generate feature table data;

[0075] 5) The ADX integrated model API allows the system to automatically determine traffic, reducing manual intervention.

[0076] In this embodiment of the invention, the media traffic credibility detection model is obtained by training based on credibility data. The specific training process is as follows:

[0077] 1) Preprocess the credibility data;

[0078] 2) Obtain the training dataset:

[0079] T=(x1,x2,x3,x4....,xN),(y1,y2,y3,y4....,yN),...,(z1,z2,z3,z4....,zN);

[0080] For a given training dataset and features, according to Bayes' theorem: P(C|X)=P(X|C)·P(C) / P(X), where P(X) is the evidence constant and can be omitted in classification decision, simplifying to P(C|X)∝P(X|C)·P(C);

[0081] Feature independence handling: In the feature vector X = (x1, x2, ..., x...) n Under the condition that P(X|C)=∏ i P(x i |C), which assumes that the features are independent given the class. Joint probability model:

[0082] P(i) = (px1 * px2 * ... * pN)

[0083] 3) Fitting;

[0084] 4) Model evaluation and validation;

[0085] 5) Theoretically, the closer the joint probability is to 1, the higher the credibility.

[0086] The media traffic credibility detection method of this invention has the following beneficial effects:

[0087] In existing technologies, programmatic verification of traffic credibility typically relies on app launch metrics for evaluation. However, this data often lacks real-time feedback, requiring feedback every other day. This process is complex, time-consuming, and inefficient. The gray area during this window, where the quality of media traffic is unknown, provides opportunities for fake traffic, data manipulation, low-quality content, and simulated traffic. The media traffic credibility detection method of this invention, after enabling automatic traffic monitoring, automatically configures test materials, binds monitoring, collects metrics, automatically cleans data, performs machine learning, updates the model, generates reports, and labels multiple credibility features, discarding traffic with low credibility. From the credibility features, it can be determined whether the traffic is mixed, swapped, rigged, or fake, etc. Testing efficiency is improved from daily feedback to minute-level feedback, a nearly 100-fold increase in efficiency. To further improve traffic credibility, this invention includes a dynamically adjustable second-hop button on the landing page to prevent black market operators from using group control to simulate landing page calls.

[0088] Accordingly, based on the media traffic credibility detection method of the present invention, the present invention also proposes a media traffic credibility detection system.

[0089] Figure 3 A structural block diagram of a media traffic reliability detection system according to an embodiment of the present invention is shown. (Refer to...) Figure 3 The media traffic credibility detection system of this invention includes:

[0090] The programmatic advertising buying module is used to purchase media traffic based on the DSP platform and to deliver advertising creatives based on the purchased media traffic.

[0091] The raw data acquisition module is used to acquire relevant data involved in the process of media traffic purchase and advertising material delivery. The relevant data includes traffic request parameters, landing page acquisition parameters, server-side legal acquisition parameters, exposure metrics, 302 metrics, click metrics, arrival metrics, and second-hop metrics.

[0092] The feature data acquisition module is used to determine the credibility of traffic features 1 based on exposure metrics and 302 metrics; determine the credibility of traffic features 2 based on click metrics and arrival metrics; determine the credibility of traffic features 3 based on arrival metrics and second-hop metrics; determine the credibility of traffic feature IP based on traffic request parameters, landing page acquisition parameters, legal parameters acquired by the server, and IP of traffic exposure reporting; determine the credibility of traffic feature UA based on traffic request parameters, landing page acquisition parameters, legal parameters acquired by the server, and browser standard protocol header of traffic exposure reporting; and determine the credibility of device features 6 based on traffic feature IP credibility 4 and traffic feature UA credibility 5.

[0093] The media traffic credibility detection module is used to determine whether the purchased media traffic is credible based on the obtained credibility scores and a pre-built media traffic credibility detection model.

[0094] While one or more embodiments of the present invention have been described above, those skilled in the art will recognize that the present invention can be implemented in any other form without departing from its spirit and scope. Therefore, the embodiments described above are illustrative and not restrictive, and many modifications and substitutions will be apparent to those skilled in the art without departing from the spirit and scope of the invention as defined in the appended claims.

Claims

1. A method for detecting the credibility of media traffic, characterized in that, include: Media traffic is purchased using a DSP platform, and advertising creatives are delivered based on the purchased media traffic. Acquire relevant data involved in media traffic purchase and advertising material delivery, including traffic request parameters, landing page acquisition parameters, server-side legal acquisition parameters, exposure metrics, 302 metrics, click metrics, arrival metrics, and second-hop metrics; The credibility of traffic features is determined based on the following criteria:

1. Exposure metrics and 302 metrics; 2. Click metrics and arrival metrics; 3. Arrival metrics and second-hop metrics; 4. Traffic feature IP credibility is determined based on traffic request parameters, landing page acquisition parameters, server-acquired legal parameters, and the IP address reported for traffic exposure; 5. Traffic feature UA credibility is determined based on traffic request parameters, landing page acquisition parameters, server-acquired legal parameters, and the browser standard protocol header reported for traffic exposure; 6. Device feature credibility is determined based on traffic feature IP credibility 4 and traffic feature UA credibility 5. Based on the obtained credibility levels, and using a pre-built media traffic credibility detection model, it is determined whether the purchased media traffic is credible.

2. The media traffic credibility detection method according to claim 1, characterized in that, The specific steps for determining the credibility of traffic features 1 based on exposure metrics and 302 metrics are as follows: The ratio of the exposure index to the 302 index is used as the confidence level of the traffic feature.

3. The media traffic credibility detection method according to claim 2, characterized in that, The specific steps for determining the credibility of traffic features based on click metrics and arrival metrics are as follows: The ratio of arrival metrics to click metrics is used as the confidence level of traffic features.

4. The media traffic credibility detection method according to claim 3, characterized in that, The specific steps for determining the credibility of traffic characteristics based on arrival indicators and two-hop indicators are as follows: The ratio of the arrival index to the second-hop index is used as the confidence level of the traffic feature.

5. The media traffic credibility detection method according to claim 4, characterized in that, The step of determining the credibility of traffic feature IPs based on traffic request parameters, landing page acquisition parameters, server-acquired legal parameters, and traffic exposure reported IPs further includes: Obtain the consistency between each traffic request parameter and the IP address reported for traffic exposure (1); 2. Obtain the consistency between the parameters obtained from each landing page and the IPs reported for traffic exposure; 3. Obtain the consistency between the legal parameters obtained by each server and the IPs reported for traffic exposure; The average of consistency scores 1, 2, and 3 is used as the traffic feature IP credibility score 4.

6. The media traffic credibility detection method according to claim 5, characterized in that, The step of determining the credibility of the user agent (UA) based on traffic request parameters, landing page parameters, server-obtained valid parameters, and browser standard protocol headers reported for traffic exposure further includes:

4. Obtain the consistency between each traffic request parameter and the browser standard protocol header for traffic exposure reporting; The consistency between the parameters obtained from each landing page and the browser standard protocol headers for traffic exposure reporting is 5.

6. Obtain the consistency between the legal parameters obtained by each server and the browser standard protocol headers for traffic exposure reporting; The average of consistency scores 4, 5, and 6 is used as the traffic characteristic IP credibility score 5.

7. The media traffic credibility detection method according to claim 6, characterized in that, The specific steps for determining device characteristic confidence 6 based on traffic characteristic IP confidence 4 and traffic characteristic UA confidence 5 are as follows: The product of traffic feature IP credibility 4 and traffic feature UA credibility 5 is taken as device feature credibility 6.

8. The media traffic credibility detection method according to claim 7, characterized in that, The media traffic credibility detection model is implemented using a machine learning model or a neural network model.

9. The media traffic credibility detection method according to claim 8, characterized in that, The steps for obtaining relevant data during media traffic purchase and advertising material delivery are implemented based on monitoring links bound to advertising materials.

10. A media traffic credibility detection system, characterized in that, include: The programmatic advertising buying module is used to purchase media traffic based on the DSP platform and to deliver advertising creatives based on the purchased media traffic. The raw data acquisition module is used to acquire relevant data involved in the process of media traffic purchase and advertising material delivery. The relevant data includes traffic request parameters, landing page acquisition parameters, server-side legal acquisition parameters, exposure metrics, 302 metrics, click metrics, arrival metrics, and second-hop metrics. The feature data acquisition module is used to determine the credibility of traffic features 1 based on exposure metrics and 302 metrics; determine the credibility of traffic features 2 based on click metrics and arrival metrics; determine the credibility of traffic features 3 based on arrival metrics and second-hop metrics; determine the credibility of traffic feature IP based on traffic request parameters, landing page acquisition parameters, legal parameters acquired by the server, and IP of traffic exposure reporting; determine the credibility of traffic feature UA based on traffic request parameters, landing page acquisition parameters, legal parameters acquired by the server, and browser standard protocol header of traffic exposure reporting; and determine the credibility of device features 6 based on traffic feature IP credibility 4 and traffic feature UA credibility 5. The media traffic credibility detection module is used to determine whether the purchased media traffic is credible based on the obtained credibility scores and a pre-built media traffic credibility detection model.

Images