Intelligent polling method for advertising

Abstract

The invention belongs to the technical field of advertisement putting management, and discloses an intelligent polling method for advertisement putting, which comprises the following steps of: automatically pulling advertisement putting data from a plurality of media platforms through a timed task; analyzing an advertisement plan name, and extracting key information such as a vehicle type, a delivery position, a material number, a document number and a page number according to a preset rule; vehicle type matching inspection is carried out on the advertisement materials and the copywriting, vehicle type information in the materials and the copywriting is compared with vehicle type information in the advertisement plan name, and early warning is triggered when the materials and the copywriting are inconsistent with the vehicle type information in the advertisement plan name; according to the invention, through integration of multiple platforms, originally dispersed media are integrated in one platform for unified management, employees are prevented from repeatedly skipping to the platform for checking and operation, the rule is flexible, the inspection rule can be customized and configured by the employees, different inspection requirements of different users are met, and a large number of advertisement plans are subjected to in-project and timed inspection. Manual inspection is replaced by a machine, so that the efficiency of staff is greatly improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of advertising delivery management, and in particular relates to an intelligent inspection method for advertising delivery. Background Art

[0002] With the rapid development of Internet technology, advertising has become an important means for enterprises to promote their products and services. However, the current multi-platform advertising delivery system faces many challenges, which mainly stem from the diversity of system configuration, inconsistent information and isolation between platforms. In the practice of multi-platform advertising, advertisers often need to manage multiple advertising platforms at the same time, such as social media, search engines, video platforms, etc. Each platform has its own unique advertising delivery system, which not only has different interface designs, but also has different configuration parameters and delivery strategies. Therefore, when advertisers place advertisements, they need to perform complex operations such as designing advertising creatives, setting target audiences, allocating budgets, and planning delivery time periods on each platform. This decentralized advertising delivery model is not only inefficient, but also increases the risk of errors.

[0003] Due to the independence and differences of the advertising delivery systems of various platforms, advertisers often need to form dedicated teams to be responsible for the advertising delivery work on different platforms. This model of dedicated people for specific tasks not only increases labor costs, but may also lead to problems such as poor communication between teams and information islands. At the same time, as the number of advertising delivery platforms continues to increase, the management difficulty will be further increased, and higher requirements will be placed on the management and resource allocation capabilities of advertisers.

[0004] Existing multi-platform advertising delivery systems can only provide standardized delivery functions and templates, which are difficult to meet the diverse needs of advertisers. For example, some platforms may not support specific ad formats or delivery strategies, resulting in advertisers being unable to freely customize according to their own brand characteristics and market strategies. In addition, even if some platforms provide customized services, they are often accompanied by high costs and complicated operating procedures, which increases the burden on advertisers.

[0005] Due to the serious data barriers and information islands between the advertising delivery systems of various platforms, it is difficult for advertisers to achieve cross-platform data integration and analysis. As a result, during the advertising delivery process, advertisers cannot fully understand the performance of advertisements on various platforms, nor can they make timely adjustments and optimizations based on real-time data. Therefore, it is easy to cause accidents such as misdelivery (i.e., delivering advertisements to irrelevant audiences) or omission (i.e., failing to deliver advertisements to the target audiences), which seriously affects the advertising effect and return on investment.

[0006] Therefore, an intelligent inspection method for advertisement placement is proposed to solve the above problems. Summary of the invention

[0007] The object of the present invention is to provide an intelligent inspection method for advertising placement to solve the problems raised in the above-mentioned background technology.

[0008] To achieve the above object, the present invention provides the following technical solution: an intelligent inspection method for advertising placement, including the following steps: S1: Automatically pull advertising placement data from multiple media platforms through scheduled tasks; S2: Analyze the advertising plan name, and extract key information such as vehicle model, placement location, material number, copywriting number, page number, etc. according to preset rules; S3: Conduct vehicle model matching inspection on advertising materials and copywriting, compare the vehicle model information in the materials and copywriting with the vehicle model information in the advertising plan name, and trigger an alarm when they are inconsistent; S4: Conduct compliance inspection on advertising copywriting, compare whether the copywriting content contains the vocabulary in the preset violation word library, and trigger an alarm when it contains; S5: Conduct consistency verification on the targeting settings of the advertising plan, compare the month in the targeting package name with the month in the advertising plan name, and trigger an alarm when they are inconsistent; S6: Record the inspection results and warning information, and push them to relevant personnel through instant messaging tools.

[0009] Preferably, the advertising materials include pictures and videos.

[0010] Preferably, the intelligent inspection method for advertising placement further includes a vehicle model matching algorithm, and this algorithm realizes the accurate matching of vehicle model information in S3 through the following steps: A1: Receive the original string to be analyzed, and this string contains the vehicle model information in the advertising plan, material or copywriting; A2: Sort a list containing known vehicle type names from long to short according to the type name length; A3: Traverse the sorted vehicle type list, compare each vehicle type name with the original string, if the original string contains this vehicle type name, record this type name as the matching result, and remove the matched part from the original string to avoid repeated matching; A4: According to the matching result, return the list of vehicle type names that are matched. If the list contains multiple types, return the "multiModels" identifier. If the list is empty but the original string contains words indicating multiple vehicle models such as "full series", also return the "multiModels" identifier, otherwise return an empty string indicating no match.

[0011] Preferably, the intelligent inspection method for advertising placement further includes a vehicle model mismatch inspection algorithm based on the vehicle model matching algorithm, and this algorithm realizes the inspection of vehicle model information in advertising placement through the following steps: B1: Extract the expected vehicle model information from the name of the advertising plan; B2: According to the material number and the copy number, the corresponding material and copy content are searched from the database, and the vehicle model information is extracted using the vehicle model matching algorithm; B3: Compare the expected vehicle model information in the advertising plan name with the vehicle model information in the materials and copy; B4: If any inconsistency is found, an alert is triggered and an inspection report containing details of the inconsistency is generated.

[0012] Preferably, the intelligent inspection method for advertising placement also includes, during the vehicle type mismatch inspection process, if the vehicle type information in the advertising plan name, material or copy is incomplete or cannot be accurately extracted, deciding whether to continue the inspection or record it as an abnormal situation based on preset skip rules.

[0013] Preferably, the advertising delivery intelligent inspection method further includes a step of verifying the advertising targeting settings, which step includes: C1: Get the set of targeting package IDs from the ad group or ad plan; C2: Obtain the targeted package information, including the targeted package name, by calling the third-party interface; C3: Compare the month in the targeting package name with the month in the advertising plan name. If they are inconsistent, an alert is triggered.

[0014] Preferably, the intelligent inspection method for advertisement delivery further includes a detailed record of the warning information, and the record includes: The specific content of the inconsistency or violation; Unique identifiers such as the name of the advertising plan involved, material number, copy number, etc. The time when the warning is triggered; Information about the recipient of the warning.

[0015] Compared with the prior art, the present invention has the following beneficial effects: The present invention integrates multiple platforms to integrate the previously scattered media into one platform for unified management, reducing the need for employees to repeatedly jump to other platforms for viewing and operating. The rules are flexible and the inspection rules can be customized by employees to meet the different inspection needs of different users. A large number of advertising plans are being invested, and regular inspections are carried out. Machines are used instead of manual inspections, which greatly improves employee efficiency. When errors are found during inspections, early warning messages are promptly sent to employees and corrections are made in a timely manner, thereby reducing company losses to a greater extent. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 It is a flow chart of the present invention. DETAILED DESCRIPTION

[0017] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0018] See also Figure 1 As shown, the present invention provides the following technical solutions: Embodiment 1 An intelligent inspection method for advertisement placement includes the following steps: S1: Automatically pull advertising data from multiple media platforms through scheduled tasks; S2: Parse the name of the advertising plan and extract key information such as vehicle model, placement location, material number, copy number, page number, etc. according to preset rules; S3: Carry out vehicle model matching inspection on advertising materials and copywriting, compare the vehicle model information in the materials and copywriting with the vehicle model information in the advertising plan name, and trigger an alert if there is any inconsistency; S4: Conduct compliance checks on advertising copy to see if the copy contains words in the preset illegal word library, and trigger an alert if it does; S5: Perform consistency check on the targeting settings of the advertising plan, compare the month in the targeting package name with the month in the advertising plan name, and trigger an alert if there is a mismatch; S6: Record inspection results and warning information, and push them to relevant personnel via instant messaging tools.

[0019] In addition, in the present invention, the advertising materials include pictures and videos.

[0020] Furthermore, in the present invention, the intelligent inspection method for advertising placement also includes a vehicle model matching algorithm, which achieves accurate matching of vehicle model information in S3 through the following steps: A1: Receive the original string to be analyzed, which contains the vehicle model information in the advertising plan, material or copy; A2: Sort a list of known car type names by type name length from longest to shortest; A3: Traverse the sorted car type list and compare each car type name with the original string. If the original string contains the car type name, record the type name as a matching result and remove the matched part from the original string to avoid repeated matching. A4: According to the matching result, return the list of the names of the matched car types. If the list contains multiple types, return the identifier "multiModels". If the list is empty but the original string contains words indicating multiple models such as "full series", also return the identifier "multiModels". Otherwise, return an empty string to indicate no match.

[0021] Moreover, in the present invention, the intelligent inspection method for advertisement placement further includes a vehicle model mismatch inspection algorithm based on a vehicle model matching algorithm. This algorithm realizes the inspection of vehicle model information in advertisement placement through the following steps: B1: Extract the expected vehicle model information in the advertisement plan name; B2: Respectively according to the material number and the copywriting number, query the corresponding material and copywriting content from the database, and use the vehicle model matching algorithm to extract the vehicle model information therein; B3: Compare the expected vehicle model information in the advertisement plan name with the vehicle model information in the material and the copywriting; B4: If any inconsistency is found, trigger a warning and generate an inspection report containing the details of the inconsistency.

[0022] Moreover, in the present invention, the intelligent inspection method for advertisement placement further includes that during the vehicle model mismatch inspection process, if the vehicle model information in the advertisement plan name, the material or the copywriting is incomplete or cannot be accurately extracted, it is decided whether to continue the inspection or record it as an abnormal situation according to the preset skipping rules.

[0023] Moreover, in the present invention, the intelligent inspection method for advertisement placement further includes a verification step for the advertisement targeting settings, and this step includes: C1: Obtain the set of targeting package IDs from the advertisement group or the advertisement plan; C2: Obtain the targeting package information, including the targeting package name, by calling a third-party interface; C3: Compare the month in the targeting package name with the month in the advertisement plan name. If they are inconsistent, trigger a warning.

[0024] Moreover, in the present invention, the intelligent inspection method for advertisement placement further includes a detailed record of the warning information, and this record includes: The specific content of the inconsistency or violation; The unique identifiers such as the advertisement plan name, the material number, the copywriting number, etc. involved; The time when the warning is triggered; The information of the warning recipient.

[0025] Embodiment 2 Intelligent Inspection of Advertisement Placement Based on a Single Media Platform Background: A certain automobile brand has placed a series of advertisements on an advertising platform. In order to ensure that the advertisement material is consistent with the model information of the copy, the copy content is compliant, and the targeting setting is correct, it is decided to adopt the intelligent inspection method for advertisement placement described in the present invention.

[0026] Implementation steps: Data Pull: Through scheduled tasks, the previous day’s advertising data is automatically pulled from the advertising platform API interface at 1 a.m. every day, including the advertising plan name, materials (pictures, videos), copy content, and targeting settings.

[0027] The pulled data is stored in the local database, and preliminary data cleaning and verification are performed.

[0028] Key information extraction: Parse each advertising plan name and extract key information such as vehicle model, placement location, material number, copy number, etc. according to preset rules (such as regular expressions).

[0029] Vehicle model matching inspection: For each advertising creative and copy, a vehicle model matching algorithm is used to extract vehicle model information and compare it with the vehicle model information in the advertising plan name.

[0030] If any inconsistency is found, it will be recorded as a warning message and marked as pending.

[0031] Compliance check of the copywriting: Load the preset illegal vocabulary and perform text comparison on the advertising copy to check whether it contains illegal words.

[0032] If any illegal content is found, it will also be recorded as a warning message and marked as pending.

[0033] Directed settings consistency check: Get the targeting package ID from the advertising plan, and obtain the targeting package information (including the targeting package name) through the API interface provided by the advertising platform.

[0034] Parse the month information in the targeting package name and compare it with the month information in the ad plan name.

[0035] If any inconsistency is found, it will be recorded as a warning message and marked as pending.

[0036] Inspection result recording and push: All inspection results (including normal conditions and warning information) are recorded in the database and an inspection report is generated.

[0037] Inspection reports and early warning information are pushed to the person in charge and relevant personnel of the advertising department through the corporate email system for timely processing.

[0038] Effect evaluation: After a month of trial operation, the car brand successfully discovered and handled multiple advertisements with inconsistent model information, illegal copywriting, and incorrect targeting settings, significantly improving the accuracy and compliance of advertising. At the same time, through automated inspections, the cost and time of manual inspections were greatly reduced.

[0039] Embodiment 3 Deployment of intelligent inspection system for advertising placement across media platforms background: An e-commerce company has placed a large number of advertisements on multiple media platforms. In order to uniformly manage and monitor the advertising, it decided to deploy a cross-media platform advertising intelligent inspection system.

[0040] Implementation steps: System architecture design: Design a distributed and scalable intelligent inspection system for advertising, including data collection module, data processing module, inspection logic module, result recording and push module, etc.

[0041] Adopting a microservice architecture, different functional modules are split into independent services, and communication and data exchange are carried out through API interfaces.

[0042] Data source connection: Negotiate API interface connection with multiple media platforms to obtain access rights to advertising data.

[0043] Write corresponding data pulling scripts or services to realize cross-platform data collection.

[0044] Data processing and storage: The collected data is cleaned, verified and formatted to ensure the accuracy and consistency of the data.

[0045] Use distributed database systems (such as Hadoop, HBase, etc.) to store processed data to ensure high availability and scalability of data.

[0046] Inspection logic implementation: According to business needs, implement inspection logic such as vehicle model matching inspection, document compliance check, and directional setting consistency check.

[0047] Introduce machine learning algorithms and NLP technology to improve the accuracy and intelligence of inspections.

[0048] Result recording and push: Design an inspection result recording mechanism, store the inspection results and warning information in the database, and generate a detailed inspection report.

[0049] Inspection reports and early warning information are pushed to relevant personnel through instant messaging tools and email systems to ensure timely communication and processing of information.

[0050] System monitoring and maintenance: Deploy system monitoring tools to conduct real-time monitoring and early warning of the operating status of the advertising intelligent inspection system.

[0051] Regularly perform system upgrades and maintenance to fix known vulnerabilities and defects and improve system stability and security.

[0052] Effect evaluation: By deploying a cross-media platform advertising intelligent inspection system, the e-commerce company successfully achieved unified management and monitoring of advertising on multiple media platforms. The system automatically completed the inspection of advertising materials, copywriting and directional settings, significantly improving the accuracy and compliance of advertising. At the same time, through real-time monitoring and early warning mechanisms, it effectively reduced the risks and problems in the advertising process.

[0053] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. An intelligent inspection method for advertisement placement, characterized in that: It includes the following steps: S1: Automatically pull advertising placement data from multiple media platforms through scheduled tasks; S2: Parse the advertising plan name, and extract key information including vehicle model, placement location, material number, copywriting number, and page number according to preset rules; S3: Conduct vehicle model matching inspections on advertising materials and copywriting, compare the vehicle model information in the materials and copywriting with the vehicle model information in the advertising plan name, and trigger an alarm when they are inconsistent; S4: Conduct compliance checks on advertising copywriting, compare whether the copywriting content contains the vocabulary in the preset violation word library, and trigger an alarm when it contains; S5: Conduct consistency verification on the targeting settings of the advertising plan, compare the month in the targeting package name with the month in the advertising plan name, and trigger an alarm when they are inconsistent; S6: Record the inspection results and warning information, and push them to relevant personnel through instant messaging tools.

2. The method for intelligent inspection of advertisement placement according to claim 1, characterized in that: The advertising materials include pictures and videos.

3. The method for intelligent inspection of advertisement placement according to claim 1, characterized in that: It also includes a vehicle model matching algorithm, which realizes the accurate matching of vehicle model information in S3 through the following steps: A1: Receive the original string to be analyzed, which contains the vehicle model information in the advertising plan, materials or copywriting; A2: Sort a list containing known vehicle type names in descending order according to the length of the type name; A3: Traverse the sorted vehicle type list, compare each vehicle type name with the original string. If the original string contains the vehicle type name, record the type name as the matching result, and remove the matched part from the original string to avoid duplicate matching; A4: According to the matching result, return the list of vehicle type names that match. If the list contains multiple types, return the "multiModels" identifier. If the list is empty but the original string contains the vocabulary indicating multiple vehicle models such as "all series", also return the "multiModels" identifier, otherwise return an empty string indicating no match.

4. The method for intelligent inspection of advertisement placement according to claim 1, characterized in that: It also includes a vehicle model mismatch inspection algorithm based on the vehicle model matching algorithm, which realizes the inspection of vehicle model information in advertising placement through the following steps: B1: Extract the expected vehicle model information in the advertising plan name; B2: Respectively query the corresponding material and copywriting content from the database according to the material number and copywriting number, and use the vehicle model matching algorithm to extract the vehicle model information therein; B3: Compare the expected vehicle model information in the advertising plan name with the vehicle model information in the materials and copywriting; B4: If any inconsistency is found, trigger an alarm and generate an inspection report containing the details of the inconsistency.

5. The method for intelligent inspection of advertisement placement according to claim 1, characterized in that: It also includes that during the vehicle model mismatch inspection process, if the vehicle model information in the advertising plan name, materials or copywriting is incomplete or cannot be accurately extracted, it is decided whether to continue the inspection or record it as an abnormal situation according to the preset skip rules.

6. The method for intelligent inspection of advertisement placement according to claim 1, characterized in that: It also includes the verification steps for advertising targeting settings, and these steps include: C1: Obtain the set of targeting package IDs from the ad group or advertising plan; C2: Obtain the targeting package information, including the targeting package name, by calling a third-party interface; C3: Compare the month in the targeting package name with the month in the advertising plan name. If they are inconsistent, trigger an alarm.

7. The method for intelligent inspection of advertisement placement according to claim 1, characterized in that: It also includes the detailed record of warning information, and this record includes: The specific content of the inconsistency or violation; Unique identifiers such as the name of the advertising plan involved, material number, copy number, etc. The time when the warning is triggered; Information about the recipient of the warning.

Images